2014 Issue 1

2014 Issue 1
Technology
Today
HIG HL IG HTI N G RAYTHE ON’S T ECH NOLOGY
2014 ISSUE 1
Raytheon Research
Developing Tomorrow’s Technology Today
A MESSAGE FROM
Mark E. Russell
Vice President of Engineering, Technology and Mission Assurance
Raytheon has always valued a focused research portfolio that both continuously
improves our products and offers revolutionary new capabilities. Best performing,
affordable solutions are important in today’s fiscally constrained environment, and
Raytheon’s research provides a mechanism to identify and nurture technologies that
help develop best-in-class solutions.
Raytheon’s research approach is multifaceted where Raytheon laboratories enterprisewide perform research that impacts our core sensing, effects, C3I (command, control,
communications and intelligence), mission support and cyber markets. Collaboration
and partnerships with universities, small businesses and other contractors are critical
elements of our approach. As part of a formal process, research topics and results from
across the company are integrated and continuously assessed against Raytheon products
and customer mission needs to ensure focus on the most promising and needed areas.
This issue highlights the depth and breadth of Raytheon research, including our more
than 30-year commitment to multifunction, wideband and high-power active electronically scanned array (AESA) technologies and longer range research areas including quantum computers. While it may take years to realize the benefits in a particular
area, conducting research is nonetheless a critical part of maintaining excellence for our
future products. Our development of gallium nitride semiconductor device technology
began nearly 15 years ago and today provides the foundation for our most advanced
radar, communication and electronic warfare development programs.
In our Leaders Corner, Raytheon Chief Technology Officer Bill Kiczuk and the other
Technology Leadership Council members discuss their roles as technology leaders and
how Raytheon collaborates and nurtures research across the company. In our Eye on
Technology section, Raytheon Australia’s sapphire resonator oscillator technology,
developed by recently acquired Poseidon Scientific Instruments, is presented.
Our special interest section highlights two of Raytheon’s university partnerships; our
partnership with the Franklin W. Olin College of Engineering, where a team of engineering students work for a year with Raytheon engineers on real-world engineering
projects, and our collaboration with Worcester Polytechnic Institute that includes
science, technology, engineering and mathematics (STEM) education, course development and research.
On the cover: The Raytheon BBN
Technologies quantum computing lab
where employee Colm Ryan analyzes
results from a recent experiment.
2
2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
Mark E. Russell
View Technology Today online at:
www.raytheon.com/technology_today
INSIDE THIS ISSUE
Feature: Research
Technology Today is published
by the Office of Engineering,
Technology and Mission Assurance.
Vice President
Mark E. Russell
Chief Technology Officer
Bill Kiczuk
Raytheon Research: Delivering Discriminating Technologies
4
Raytheon AESA Research: Past, Present and Future
8
State-of-the-art RF Semiconductors for Military Systems
14
Computational Imaging Technology
18
BareMetal: a New Cybersecurity Technology
21
Next Generation EO/IR Detectors
22
Quantum Computers: Big and Small
24
Particle Flow Filters to Solve Near Impossible Problems
28
Automated Language Translation
30
Managing Editor
Tony Pandiscio
Optimization Algorithms for Decentralized Planning and Control
32
Feature Editor
John Zolper
Partially Observable Decision Processes
36
Senior Editors
Corey Daniels
Eve Hofert
Art Director
Susan DeCrosta
Photography and Art
Fran Brophy
Stephen Delisle
Daniel Plumpton
Website Design
Nick Miller
Publication Distribution
Rose McGovern
Contributors
Paul Bailey
Steve Klepper
Tony Marinilli
Lindley Specht
Nora Tgavalekos
Raytheon Leaders Corner
Q & A With the Raytheon Technology Leadership Council
40
Eye on Technology
Raytheon Australia’s Poseidon Scientific Instruments
42
Special Interest
Raytheon’s Collaboration With Worcester Polytechnic Institute
44
Raytheon’s Partnership With the Franklin W. Olin College of Engineering 46
Patents
48
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
3
FEATURE
Raytheon Research:
DELIVERING DISCRIMINATING TECHNOLOGIES
Raytheon is a leading technology and innovation company
that consistently provides innovative solutions, services
and mission support to our
global customers. Our vision is
to be the most admired defense and aerospace systems
company through our worldclass people, innovation and
technology.
T
o achieve this vision, Raytheon maintains
a leadership position in the technologies that enable our products to deliver
mission solutions. A robust, diverse research
portfolio is essential to our strategy. Raytheon
research is not solely an internally focused
activity done in a central corporate laboratory.
Rather, Raytheon research is a highly dynamic,
collaborative process where good ideas and
novel solutions come from many sources.
Raytheon has developed a well-structured,
mission-to-technology assessment process
to identify technology areas whose development will yield new or enhanced products and
system capabilities. The process starts with
customer roadmaps and mission requirement
documents plus Raytheon’s deep domain
knowledge, and drives down to identifying the
critical enabling technologies. Synergies are
identified across product families (e.g., core
technologies for multiple radars) to leverage
our research investments. Technology development can be achieved via various project and
funding mechanisms, including customer
program-funded technology development,
customer contracted research and development
(CRAD), independent research and development (IRAD), partnerships and alliances that
include universities, small businesses and
4
2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
government laboratories and mergers and
acquisitions. This approach leads to a research
and technology portfolio that leverages expertise both within Raytheon and externally.
chains enables Raytheon to identify technology needs and opportunities, and to focus
research and development to best meet the end
customer’s system requirements.
An important part of Raytheon’s research
portfolio is the in-house development and application of advanced semiconductor devices.
Figure 1 is an electro-optical/infrared (EO/IR)
example of Raytheon’s continuum of technology development, which spans materials,
devices, modules, systems and platforms. The
figure includes images of an EO/IR system
evolution that starts with a mercury cadmium
telluride (HgCdTe) crystal boule, progressing
to wafers fabricated with EO/IR focal plane
arrays (FPAs), to the formation of complete
detector modules, to the integration of imaging
subsystems, and finally to integraton on host
platforms. Raytheon also has a completely
integrated radio frequency (RF) roadmap,
extending from semiconductor monolithic
microwave integrated circuits (MMICs), to
transmit/receive modules, to active arrays and
ultimately to full radars. The deep domain
knowledge across these integrated technology
Sensor and Device Technologies
The article by Sarcione et al., “Raytheon AESA
Research: Past, Present and Future,” presents
Raytheon’s more than 40 year history developing and productizing multifunction, wideband
and high-power active electronically scanned
array (AESA) technologies. AESA technologies
have progressed significantly during the past
four decades at Raytheon, and our research
progress provides our domestic and international customers with advanced unprecedented
capabilities and performance. Applications
span surface, space and airborne platforms
for radar, communications and electronic
warfare missions and the article highlights key
past, present and potential future technology
advances in areas such as MMICs, low noise
RF sources and packaging.
Complementing the AESA article, the
article by Whelan et al., “State-of-the-art RF
PLATFORMS
SYSTEMS
MODULES
DEVICES
MATERIALS
Figure 1. Representation of electro-optical/infrared (EO/IR) technology progression from
materials to platforms. The images are for EO/IR focal plane array (FPA) technology going from mercury cadmium telluride (HgCdTe) crystal boules, to FPA wafers, to complete
detectors, to imaging subsystems integrated on platforms.
FEATURE
Semiconductors for Military Systems,” reports
on recent progress in developing, maturing
and applying high-performance MMICs to an
array of Raytheon products. The article covers
low-noise metamorphic high electron mobility
transistors (mHEMTs) for RF receivers as well
as high-power gallium nitride (GaN) MMICs
that are critical enablers for new Department
of Defense (DoD) systems. As noted in the
article, the Office of the Secretary of Defense
(OSD) honored Raytheon in 2013 for successful completion of a Defense Production Act
(DPA) Title III GaN production improvement
program that culminated in the demonstration
of a manufacturing readiness level (MRL1) of
8 for the GaN MMIC technology. This is the
first MRL 8 certification of GaN MMICs in the
industry and indicates a readiness to enter low
rate initial production.
The article by Johnson and Smith, “Next
Generation EO/IR Detectors,” reports on
recent progress in several EO/IR detector
technologies that enable increases in image
resolution, target discrimination and reduced
system cost. The first area covered is the
emerging field of strain-layer superlattice high
operating temperature mid-wave IR detectors
having fundamental properties that in theory
perform better than HgCdTe and can cost less
by leveraging the III–V semiconductor manufacturing infrastructure. Also described are
dual-band mid-wave/long-wave detectors that
afford superior target discrimination because
of the distinct image spectral characteristics of
the two bands. Finally, progress in uncooled
microbolometer technology is presented with
the successful realization of a 2048 x 1536
pixel long-wave FPA that is the world’s largest
uncooled sensor of this type. Dramatic improvements in image resolution are
reported for this uncooled FPA.
The article by Gann and Gibbons,
“Computational Imaging Technology is
Revolutionizing Digital Imaging,” discusses
an additional path Raytheon has explored to
further improve sensor imaging systems, i.e.,
Figure 2. One of the quantum computer research areas at the Raytheon BBN Technologies
Bits and Waves lab. The suspended cylinder to the left is a dilution refrigerator used to
cool the quantum computing devices to an operating temperature below 50 milli-Kelvin.
computational imaging (CI). CI leverages
advances in digital processing along with novel
algorithms and sensor architectures to enable
new strategies for optimizing imaging systems.
The essence of computational imaging is the
idea that significant gains can be attained by
inserting the computational step earlier in the
image acquisition or formation chain — not
just performing processing post-image formation — to produce improved and specialized
forms of scene information. Raytheon has been
collaborating with several universities and
small businesses in this field.
The final sensor and device technology article,
“Quantum Computers: Big and Small,”
describes one of Raytheon’s farthest reaching research areas: quantum computers. The
promise of quantum computers to provide
exponentially faster computation, compared
to conventional computers, has been widely
touted. But while the “physics works ” for a
very select set of problems, some of which
1 MRL is a measure used by the DoD and many of the world’s major companies to assess the maturity of manufacturing
readiness.
are of high interest to Raytheon’s customers,
the computer engineering to produce a truly
robust large-scale computer remains a research
topic. The authors of this paper, Dutton and
Silvia, are from the Quantum Information
Processing (QuIP) group at Raytheon BBN
Technologies, a recognized leader in quantum
computer technologies. The authors report
on the status of their quantum computing research, highlighting their progress in extending
the de-coherence time of their superconducting quantum bit (qubit) and achieving logical,
versus physical, qubits such as are required for
computationally useful quantum computers.
Their research is conducted at the BBN Bits
and Waves laboratory in Cambridge, Mass.
(Figure 2).
Algorithms and Processing
The previous section discusses research in
CARCO Rate table supplied by Ideal Aerosmith, LLC.
sensor and
device technology, but in many
of Raytheon’s systems, backend algorithms
and processing also play a crucial role.
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
5
FEATURE
Advancements in mathematics, as well as
algorithmic strategies and progress in digital
processing, are exploited to more effectively
and more efficiently turn sensor data into actionable information. The end-system solution
must trade off computational precision and
accuracy with the execution speed and power
requirements. A highly accurate solution that
is achieved only after mission completion
has little value; nor is a solution useful that
requires more computational power than the
platform can provide.
The first article in this area, “Particle Flow
Filters to Solve Near Impossible Problems”
by Daum, represents a significant advance in
the use of particle filters to efficiently calculate
high-accuracy estimates of nonlinear system
states such as a ballistic missile’s position and
velocity. Prior to this work, particle filters
were known to be highly accurate but required
extensive computational resources in run time
and processor capability such that they were
not viable for most real-time applications.
Daum invented a new variant of the particle
filter, the particle flow filter (PFF), and his
solution is orders-of-magnitude computationally faster than standard particle filters. The key
improvement is a new way to select the particle
locations based on the underlying physics
describing the system’s dynamics. For some
applications, a speed increase of 10 orders of
magnitude over conventional particle filters is
achieved with an estimation accuracy superior
to the widely used extended Kalman filter.
These results could have wide-ranging impacts
on a broad range of Raytheon products.
Effective automated language translation
(ALT) capabilities are needed more than ever,
primarily because of the massive amount
of data available on the Internet, half the
content of which is in languages other than
English. Raytheon BBN Technologies has been
researching automated language translation
for over a decade, and the article by Abib,
Makhoul and Andrews, “Automated Language
Translation,” presents groundbreaking work
on ALT with their statistical machine translation (SMT) approach. Instead of defining
translation rules manually, SMT uses a
machine learning approach to develop a large
set of translation rules automatically based
on statistical models of translation. SMT has
6
2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
Raytheon Research: Delivering Discriminating Technologies
revolutionized automated language translation
by quickly and cheaply enabling the development of translation systems for new languages
and domains.
Raytheon BBN Technologies has developed
and matured the SMT approach under several
Defense Advanced Research Projects Agency
(DARPA) initiatives, including the Global
Automatic Language Exploitation (GALE)
program and the Translation for Tactical Use
(TransTac) program. They have also created
several turnkey solutions for both the government and commercial markets based on their
translation technology.
A new Raytheon Pikewerks product called
BareMetal is highlighted in the article
“BareMetal: A New Cybersecurity Technology”
by Fraser, Tanen and Egalite. BareMetal
is a small, portable device that allows an
information technology (IT) professional
(not necessarily a firmware expert) to quickly
characterize the firmware on their IT systems
and identify potential threats. IT equipment
uses firmware to load the operating system
and initialize system components. Firmware is
evolving from the basic input/output system
(BIOS) type used extensively in the past to
new firmware based on the Unified Extensible
Firmware Interface (UEFI) standard. Although
this standard enables more structured and efficient firmware development, it also potentially
introduces new opportunities for adversaries to
inject low-level malware and low-level vulnerabilities. Most IT equipment security products
monitor and protect the application software,
but BareMetal specifically focuses on threats to
the low-level firmware where some of the most
persistent and hidden malware can hide.
Figure 3. A simulated urban area is used to test and refine cooperative unmanned aerial
vehicle control algorithms that optimize the ability to find and track ground objects
of interest.
FEATURE
Autonomous Systems
Autonomy relates to systems which have a set of intelligence-based
capabilities that allow them to respond to situations that were not preprogrammed or anticipated in the design. Autonomous missions are
typically conducted in unstructured and dynamic environments where the
systems must have a degree of self-governance and adaptability.
In 2011, OSD identified seven science and technology area priorities that
they would invest in from 2013 to 2017. These technologies are expected
to greatly impact future OSD systems and mission needs. One of these
technology areas is autonomy. In response to this need, Raytheon formed
a cross-company team of subject matter experts in fields related to autonomy to develop key enabling technologies in this area and a strategy
for their application to Raytheon products. Two articles that describe
autonomy-enabling decision and control frameworks are included in this
research edition.
The first article, “Optimization Algorithms for Decentralized Planning
and Control ” by Hirsch and Schroeder, presents research on optimization algorithms for distributed planning and control. This work supports
our customer’s goal of doing more with less. Specifically, there is interest
in having multiple unmanned aerial vehicles (UAVs) be controlled by a
single operator. This entails a more distributed and autonomous control
of some UAV functions versus the current centralized operator-intensive
control. Hirsch and Schroeder describe their approach to this problem
as a hybrid methodology combining greedy randomized adaptive search
procedures (GRASP) and simulated annealing (SA). Their approach has
been tested in simulation using a scenario that contains multiple UAVs
searching for multiple targets in an urban area (Figure 3), and a summary
of their results is presented.
The second article, “Partially Observable Decision Processes” by Moore
and Vitali, presents an analytic framework for making decisions based on
only a partial understanding of a situation, i.e., partial information of the
system state. The framework is based on the theory of partially observable
Markov decision processes (POMDPs) and uses a belief state to quantify
the uncertainty in the estimate of a system’s true state. Once calculated,
the belief state is used to select the best course of action (COA) at that
time. The article describes several application areas for the framework,
including autonomous sensor control for conducting intelligence,
surveillance and reconnaissance (ISR) missions and autonomous COA
generation for an airport security monitoring system.
Other Research at Raytheon
Research and technology maturation continue to play critical roles at
Raytheon in developing the best products to meet our customers’ mission
needs. This issue of Technology Today highlights some of our current research projects in the areas of sensor and device technologies, algorithms
and processing, and autonomous systems. There are many other projects
ongoing in these areas and in other areas, all with the common goal of
helping Raytheon maintain its ability to provide best value solutions to
our customers. •
John C. Zolper, Ph.D.
ENGINEERING PROFILE
John Zolper,
Ph.D.
Vice President,
Research and
Innovation
John C. Zolper is
vice president of
research and innovation for Raytheon
Engineering,
Technology and
Mission Assurance
(ET&MA). He joined
Raytheon in 2007 as
part of the corporate
technology team.
Zolper partners with
the chief technology
officer to develop and
implement an integrated technology and research vision and
strategy for Raytheon. He leads the Raytheon Innovation
Challenge to inspire creative solutions that address critical
customer needs.
Zolper points to the importance of his role of ensuring that “the
company is constantly considering new and emerging customer
needs by understanding customer capability gaps, assessing the
technology available today and determining the technologies
Raytheon should be developing for the future.” He spends a lot
of his time “working to understand where customers are headed
and what gaps they will have.”
Prior to his position with Raytheon ET&MA, Zolper served in
roles of increasing responsibility at the Defense Advanced
Research Projects Agency (DARPA). As director of the
Microsystems Technology Office (MTO), he was responsible for
the strategic planning and execution of a portfolio of more than
75 research programs with an annual budget of more than $400
million covering all areas of advanced component technology,
including electronics, photonics, MEMS (microelectromechanical
systems), algorithms and component architectures. Before joining
DARPA, Zolper was a program officer at the Office of Naval
Research (ONR) and a principal member of the technical staff
at Sandia National Laboratories.
Zolper traces his path to Raytheon through his long involvement
in high performance semiconductor technology, “I joined Sandia
National Laboratories in 1989 where I applied my understanding
of semiconductor materials and device technology to high performance electronics, performing early research on gallium nitride
(GaN) transistors. In many ways, that work on GaN transistors
led me to where I am today. I left Sandia to lead the wide bandgap semiconductor electronics programs at the ONR and then
from there helped start the wide bandgap technology program
at DARPA. That DARPA program was critical to establishing the
viability of the GaN monolithic microwave integrated circuit
technology that Raytheon is leveraging today in their designs.”
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
7
FEATURE
RAYTHEON AESA RESEARCH:
Past, Present and Future
Introduction
Multifunction, wideband and high-power
active electronically scanned array (AESA)
technology has progressed significantly during
more than 40 years of development at Raytheon. The progress of our research provides
domestic and international customers with
unprecedented capabilities, performance and
affordability. AESA applications span surface, space and airborne platforms for radar,
communications and electronic warfare (EW)
missions. The key technologies of monolithic
microwave integrated circuits (MMICs), waveguide to printed circuit radiating elements, low
noise sources, radio frequency (RF) and digital
electronics, thermal and power conditioning, packaging, interconnects and processing
continue to evolve and provide our customers
with a wide range of solutions for their most
challenging problems.
The AESA has been, and will continue to be,
the workhorse for the majority of RF sensors
and other high-end RF systems. However, the
evolution of RF packaging, device technologies
and architectures continues to mold the AESA’s
capability and affordability and their use is
now extending beyond traditional military
applications to include commercial weather
and air traffic surveillance and wireless communication. Agencies such as the Department
of Homeland Security (DHS), the Department
of Commerce (DoC), the National Oceanic and
Atmospheric Administration (NOAA), the National Weather Service (NWS), the Department
of Transportation (DoT) and the Federal Aviation Administration (FAA) as well as the commercial wireless industry now also recognize
the utility and cost-effectiveness of AESA-based
sensor and communication systems. In addition, multifunction AESA-based systems are
becoming more capable with dynamic transmit
and receive functionalities that enable the same
system to perform multiple missions.
Looking at the evolution of AESAs from the
early 1970s to today, we see a dramatic increase in capability enabled by key technology
8
2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
developments in MMICs, radiating elements,
receiver/exciters, beamforming and signal/data
processing and packaging (Figure 1). The early
focus was at the high frequency (HF) to ultraHF (UHF) bands, but today, AESAs operate
from L-band through Ku-band and continue to
extend into the millimeter-wave band, enabled
by advances in gallium arsenide (GaAs), silicon
(Si) and mixed signal devices. Gallium nitride
(GaN), now in production at Raytheon, has
enabled the next generation of higher power,
more efficient AESA-based products. These
technology developments enable architectures
that provide cost and performance trades with
scalability and modularity that are now more
easily integrated into a wider variety of platforms and applications.
The production of AESAs for many military
applications and some high-end commercial
applications (e.g., Iridium) began in earnest in
the early 1990s. Packaging of the microwave
electronic circuitry in most cases required hermetically sealed modules with corresponding
interconnects, thermal control and a maintenance philosophy that dictated the weight, volume and maintainability of the AESAs. Today’s
AESAs have evolved to become lighter, even
with increased packaging density, while some
still have hermetic packages, others exploit
alternative environmental protection technologies. This evolution along with technology improvements in MMICs, interconnects, thermal
control, etc., have realized a 50 percent savings
in both weight and cost.
The industry began investing in architectures
and technologies several years ago to further
improve the affordability and streamline the
integration of phased arrays. The next generation of affordable AESAs are enabled by emerging MMIC technologies capable of supporting
higher RF power per unit area (e.g., GaN),
providing more functionality per unit area
(e.g., RF complementary metal oxide
semiconductor [CMOS] and silicon germanium) and leveraging maturing RF microelec-
tromechanical systems (MEMS). In addition,
higher levels of circuit board integration with
fewer manufacturing steps and surface mount
assembly connectors to eliminate expensive
interconnects further reduce costs. Environmental protection technologies also eliminate
the need for hermetic packages enabling more
affordable and lightweight circuit card assembly (CCA) arrays as the major building block of
a broad class of phased arrays.
Radiating Elements
Phased array radiator evolution has advanced
through many improvements since the first
explorations of array mutual coupling. Today’s
radiators have the advantages of high efficiency,
advanced polarization purity, wide bandwidth
and scan volume, as well as reduced component and system cost. The evolution of the
devices has been matched by an evolution in
understanding, modeling and simulating complex electromagnetic phenomena and interactions; evolving from the first analytic expansion of the dipole electric field to the complex
frequency-time domain three-dimensional
full-wave solvers used today.
Some of the first dipole phased array radiators
were used by Raytheon in early warning radar
systems such as the Precision Avionics Vectoring Equipment Phased Array Warning System
(PAVE PAWS) in a dual polarization configuration. At the same time, important advances
were made in waveguide radiators which began
use in our missile defense family of radars.
Printed circuit radiators, such as the patch and
stacked patch, emerged in prominent systems
including the Iridium global communications system. These radiators have all of the
characteristics of waveguide cavity radiators,
but with boundary conditions stemming from
the array environment instead of the physical
unit cell properties. The stacked patch radiators
expanded the operating frequency bandwidth
by as much as a factor of two compared to
prior approaches.
FEATURE
RADIATING ELEMENTS
Thumbtack
Printed Circuit
Flared Notch
Dielectric Loaded
D
Waveguide
W
Tightly Coupled
Dipole With
MagnetoDielectric Loading
Printed Circuit
Stacked Patch
Duall PPolarization
FFolded
ld d D
l i
Dipole
ACTIVE DEVICES & MONOLITHIC MICROWAVE INTEGRATED CIRCUITS (MMICs)
GaN, CMOS
MMICs
Discrete
Devices
Si Bipolar
MASTER
ER OSCILLATORS
MESFETs, pHEMTs, mHEMT
mHEMT, HBTs
HBTs, SSiGe
GaAs, MMICs, Si ASICs
Quartz
Oscillators
Q
(photo
courtesy of Wenzel)
(
BEAM FORMING AND SIGNAL DATA PROCESSING
Analog
Coaxial, Stripline, Microstrip
Analog Combiners
PACKAGING
Brick Architecture
Packaging &
Electronics
Orthogonal
to Aperture
Surface Acoustic
(SAW) Wave Oscillators
EVOLUTION FROM ANALOG TO DIGITAL
Transmit/
Receive
Modules (TRMs)
1980s
Digital
SubarrayLevel
ArrayLevel
ADC
ElementLevel
Circuit Card Assembly
(CCA) Architecture
Packaging & Electronics
Parallel to Aperture
Multichannel
TRMs
1990s
ADC
ADC
ADC
ADC
ADC
Digital Summation
Beam 1
Beam 2
Beam 3
Moduleless TRIMMs
With Integrated
Radiating Elements
Transmit/Receive
Integrated Multichannel
Modules (TRIMMs)
Tile Architecture
Packaging &
Electronics Parallel
to Aperture
1970s
Sapphire Oscillators
E A R LY 2 0 0 0 s
MMIC Chips on
Circuit Card Assemblies
With Embedded
Interconnects
TO D AY
TOMORROW
RAYTHEON SYSTEMS*
PAVE PAWS &
BMEWS EWRs
F-15 & F-18
Airborne Radars
SPY-3
CJR
F-18G NGJ Pod
JLENS SuR/FCR
DDR
ROTHR
Iridium
Main
Mission
Antennas
AMDR
GBR, AN/TPY-2, SBX
Missile Defense Radars
CEC
DDS
Radio
*Partial
ASTOR
SYSTEMS LEGEND
AMDR
= Air and Missile Defense
Radar
AN/TPY-2 = Army Navy/Transportable
Radar Surveillance –
Model 2
ASTOR
= Airborne STand-Off
Radar
BMEWS = Ballistic Missile Early
Warning System
CEC
= Cooperative Engagement
Capability
CJR
= Cobra Judy Replacement
DDR
= Dismount Detection
Radar
DDS
= Data Distribution System
TECHNOLOGY LEGEND
EWR
= Early Warning Radar
ADC
= Analog-to-Digital Converter
GBR
= Ground Based Radar
ASIC
= Application Specific Integrated Circuit
JLENS SuR/FCR = Joint Land-attack Elevated
Networked Sensor Surveillance Radar/Fire Control
Radar
NGJ
= Next Generation Jammer
PAVE PAWS
= Precision Avionics Vectoring
Equipment Phased Array
Warning System
COTS
= Commercial off the shelf
GaAs
= Gallium Arsenide
GaN
= Gallium Nitride
HBT
= Heterojunction Bipolar Transistor
HF
= High Frequency
MESFET = Metal-Semiconductor Field Effect Transistor
mHEMT = Metamorphic High-Electron-Mobility Transistor
= Relocatable Over The
Horizon Radar
MMIC
SBX
= Sea Based X-Band radar
SAW
= Surface Acoustic Wave
SPY-3
= Navy Shipboard Radar
designation
Si
= Silicon
SiGe
= Silicon Germanium
ROTHR
= Monolithic Microwave Integrated Circuit
pHEMT = Pseudomorphic High-Electron-Mobility Transistor
Figure 1. Active electronically scanned array (AESA) technology evolution
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
9
FEATURE
Wideband radiators, such as the Vivaldi flared
notch emerged to address the needs of wide
bandwidth systems. The flared notch provides
broadband performance with nominally linear
polarization. Because of its broad beamwidth,
it is ideal for electronically scanning antenna
arrays with scan volumes of at least 60 degrees.
For the past decade, Raytheon has pioneered
taper optimization approaches which compress
the notch length significantly below that of
common exponentially tapered notches while
also extending bandwidths up to 10:1. With a
pair of orthogonally polarized elements, virtually any arbitrary polarization can be achieved
at any given beam scan position when amplitude and phase control is employed.
Raytheon AESA Research: Past, Present and Future
balun embedded in the circuit board below the
ground plane or printed on the substrate of the
radiator structure. The design can be scaled to
various bands ranging from HF to Ku frequencies. The thumbtack radiator has a coincident
and stationary phase center over bandwidths
up to 10:1. In addition, the thin and planar
structure produces a very low cross polarization component, a desired feature for many
advanced systems.
Tightly coupled dipole arrays (TCDAs) are
also being developed for AESA-based systems
requiring ultra-wideband operation and lowprofile packaging (Figure 2). The TCDA design
takes advantage of novel magneto-dielectric
Raytheon’s long-term commitment to the
development of GaN technology began nearly
15 years ago, and has leveraged its long history
of GaAs semiconductor work, as well as partnerships
with industry, academia and government.
Printed circuit radiating elements opened the
door to further cost reduction through low
cost manufacturing techniques. The thumbtack
radiator is a low profile antenna which matches
to free space directly without a long flared
notch to perform an impedance transformation. Each thumbtack radiator is fed by a
Figure 2. Tightly coupled dipole array
(TDCA) enables wideband and low profile
solutions.
10 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
loading materials with advanced dielectric wide
angle impedance matching (WAIM) structures
to push efficient operational bandwidths into
the range of 20:1. This may be achieved with
antenna depths of 1/40th of a wavelength at
the lowest frequency of operation, an orderof-magnitude thinner than traveling wave
radiators of the same bandwidth. The dipole
nature of the radiator naturally lends itself
to dual polarization collocated phase center
architectures, thus allowing synthesis of arbitrary polarizations. Due to its reduced height
and planar construction, it exhibits improved
polarization purity, especially during diagonal
plane scanning, compared to deeper or tapered
radiators.
Active Devices and MMICs
In the 1970s as an alternative to vacuum tubes,
discrete transistors, mainly composed of
silicon, were used in a hybrid configuration in
arrays. Large, lumped element capacitors and
inductors were used to match these transistors
at microwave frequencies to standard impedances.
Industry attention and movement from hybrid
microwave integrated circuits to monolithicbased approaches began due to the requirements for low cost, mass production compatible circuits providing increased integration,
reliability and multi-octave performance.
Development of MMIC devices began in the
early 1980s when gallium arsenide (GaAs)
was emerging as the semiconductor of choice
for efficiently amplifying and phase shifting
RF signals. Efforts ramped up in the 1990s,
as GaAs-based MMICs were introduced into
large production programs for a new generation of phased array radars. Over time, the
performance requirements of military systems
became more challenging thus requiring further improvements to MMIC power, efficiency
and low noise performance.
GaN MMIC technology was pursued to help
meet these new higher performance military
system requirements. Raytheon’s long-term
commitment to the development of GaN
technology began nearly 15 years ago, and has
leveraged its long history of GaAs semiconductor work, as well as partnerships with industry,
academia and government. Raytheon’s development history with GaAs provided the needed
infrastructure and lessons-learned experience
to accelerate GaN’s development. This included
the growth of starting material, the modeling of
a GaN transistor’s RF performance, the semiconductor fabrication facility, the microwave
and module design and the testing capabilities.
Through early strategic partnering with Cree,
the University of California Santa Barbara
and U.S. government labs during the Defense
Advanced Research Projects Agency (DARPA)
Wide BandGap Semiconductors (WBGS)
Phase II program, the team was able to shorten
the cycles of learning and leverage each other’s
findings to more quickly advance the state of
GaN transistors (Figure 3).
Raytheon’s focus on early reliability demonstrations and transition to 4-inch wafers, to
leverage the existing GaAs manufacturing facility, resulted in an industry leading manufacturing readiness level (MRL) of 8 accomplished
under Raytheon’s Office of Secretary
of Defense Title III program.
FEATURE
gies (GaN HEMTs and Si CMOS) could
coexist on the same Si substrate. The performance impact to both the Si CMOS and GaN
HEMT circuits was minimal. This technology provides an opportunity to enable digital
control and optimization of RF and mixed
signal circuits (e.g., amplifiers with on-chip
digital control and calibration), adaptable or
linearized bias controlled power amplifiers
(PAs), high-power digital-to-analog converters
(DACs) and many other novel on-chip capabilities with better RF performance than silicon
germanium (SiGe) devices.
Figure 3. Early GaN wafer developed and
tested as part of the Raytheon and DARPA
GaN MMIC maturation programs.
In addition to GaAs and GaN MMIC development, our research and investments extend
to customizing the many unique and prolific
MMIC functions that make up our modules.
Our GaAs pseudomorphic high electron mobility transistor (pHEMT) technology development focused on the MMICs which provide
the amplitude and phase control and enabled
an efficient digital interface. The pHEMT
process mixes RF and logic functions on the
same GaAs MMIC. This optimizes efficient
serial or parallel logic interfaces to a silicon
(Si) controller chip which extends to the beam
steering electronics. This unique capability in a
GaAs MMIC enables logic functions locally and
minimizes the number of off-chip components
and interfaces, improving reliability, producibility and reducing the AESA’s size and cost.
Master Oscillators
AESA based radar systems require very low
noise sources for their master oscillators to take
advantage of the high dynamic range enabled
by the AESA. High-performance, low-phasenoise sources have evolved over the decades
from frequency-multiplied quartz crystal
oscillators in the earliest AESA-based systems
to frequency-multiplied surface acoustic wave
(SAW) oscillators in the 1980s. This technology
evolution enabled higher levels of sensitivity
and dynamic range with Raytheon’s novel SAW
resonator and oscillator technologies delivering outstanding radar system performance for
several decades.
The emergence in the mid-1990s of new
threats and the need to operate in high clutter
environments called for even more demanding radar sensitivity requirements that pushed
the limits of SAW-based exciters. By the early
1990s, Poseidon Scientific Instruments (PSI)
developed a compact sapphire resonator. It
employed a single synthetic sapphire resonator that offered orders of magnitude reduction in microwave oscillator noise over our
SAW-based solutions. By 2000, PSI had fully
productized their compact sapphire-based oscillator into a shoebox size form factor suitable
for radar applications. This sapphire oscillator
technology now enables even higher performance and continues to provide Raytheon
radar systems with industry leading levels of
performance.
Beamforming and Signal/Data Processing
Beamforming in early AESAs at the low frequencies (e.g., UHF) was primarily just coaxial
cabling and commercial off-the-shelf (COTS)
connectorized combiners. At higher frequencies, waveguide manifolds, printed circuit (Figure 4) or microstrip combiners were typically
employed for analog beamforming due to their
cost, weight and size advantages.
Today, we still use variations of the same core
technologies, however the manufacturing capabilities and material choices are more diverse.
Future passive analog RF beamforming components will continue to leverage new materials
as they become available, further improving
integration, interconnects and reducing cost.
The emergence of digital beamforming and
signal processing began in earnest in the 1990s
when analog-to-digital converter (ADC)
technology began to be commercialized in the
RF domain for the wireless industry. Major
The extension of the pHEMT process to other
heterogeneous MMIC integrations has continued under independent research and development (IRAD) and DARPA investments. Our
research interests now include technologies
that directly integrate GaAs, indium phosphide
(InP) and GaN RF devices with high density Si
CMOS logic on a single Si wafer. This heterogeneous integration process and technology
provides a path for optimization of specific
functionality and digital control using the best
active device technology and minimizes further
the number of interconnects.
We demonstrated the world’s first heterogeneously integrated MMIC using GaN high electron mobility transistor (HEMT) and Si CMOS
devices. A GaN RF amplifier using a Si CMOS
gate bias control proved that the two technolo-
Figure 4. The Rotman-Turner lens is an example of a stripline beamformer design that
can offer low-cost, wideband performance in a compact size.
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
11
FEATURE
improvements in resolution due to lower jitter
and power dissipation opened up new
applications. Many military AESA radar
systems benefited from these improvements
as well. Much of the commercial industry
research focused on Si, however other higherend military systems developed converters
based on InP electronics for higher sampling
rates and dynamic range.
Most of today’s AESA radar systems have only
a handful of digitized receiver channels, thus
the entire dynamic range burden is placed on
only one or a few converters. The next generation of AESA systems will have digitization at
the subarray level, and some will extend to the
element level. Thus hundreds to thousands
of digital channels and degrees of freedom
will be available for advanced signal and data
processing. This will enable unprecedented
performance and new capabilities for radar,
communications and EW missions such as
creating and processing multiple simultaneous
beams, increased polarization diversity and
improved dynamic range. Digitization at the
AESA element level will also simplify transmit/
receive module (TRM) functionality by no
longer requiring the MMICs for amplitude and
phase control.
Digitization of AESAs is not limited to just
advances in ADC technology. Advances in the
ability to move and process received data in
the digital domain also continues to rapidly
advance. Signal and data processors have
benefited from Moore’s law1 to become much
more capable, lower power, smaller and more
affordable, enabling the implementation of
new and more sophisticated processing of the
received data.
Packaging
During the 1980s, many airborne AESAs were
constructed using hermetically sealed housings, typically aluminum, with glass to metal
sealed RF and direct current (DC) connectors. However, AESA TRMs used for UHF
surface-based radar applications used sealed,
but not hermetic, assemblies due to their large
size. Many of the components were mounted
to ceramic carriers with fluxless solders such
as gold-tin. These carriers contained discrete,
bipolar transistors, field-effect transistors
(FETs) or simple microwave integrated circuits
(MICs). Due to the coefficient of thermal
12 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
Raytheon AESA Research: Past, Present and Future
expansion (CTE) mismatch between the carrier
and housings, the carriers had to be mounted
to the housings with screws. These modules
were large, complex and costly, but were successfully used in the first X-band solid-state
phased array (SSPA) that employed fundamental frequency amplification for transmit. The
multiple thermal interfaces between the heatgenerating regions of the active devices and
the cooling system had a major impact on the
packaging, reliability and replacement strategy
of the modules. The X-band airborne module
was approximately 1.5 x 5.2 x 0.5 inches while
the UHF module was at least twice the size.
Major thrusts during the 1990s were launched
to reduce cost and increase performance of
modules and arrays. Advances in MMIC
technology yielded chips with greater power
and more on-chip capabilities, resulting in a reduced numbers of chips per module. The new,
more capable MMICs and the development of
low CTE housings with ceramic feedthroughs,
ionically pure die attach adhesives and fine
feature thick films helped usher in the next generation of modules. These modules, developed
for airborne and surface radars, are fabricated
using automated assembly methods and are
much smaller than their module predecessors.
These MMIC modules were assembled into
a brick-style transmit/receive integrated
multichannel module (TRIMM). The MMIC
modules, RF and DC circuit cards and radiating element were attached by a combination of adhesive bonding and fasteners to an
aluminum cold plate. Interconnection between
the MMICs, RF and DC circuit cards and the
radiating elements was accomplished with
ribbon connectors in place of rigid connectors.
The elimination of the rigid connectors greatly
reduced the cost and weight of the assembly.
The next evolution of hermetic module packaging was the development of multichannel
modules, for example the High Density Tile
Module – 4 channel, mark number design
variant 1 (HTM4 Mk1). The HTM4 Mk1
module consists of a multilayer high temperature thin-film ceramic (HTTC) substrate that
contains the RF circuitry, a multilayer thick
film substrate with DC and logic circuitry, and
an interposer which connected the two substrates into a single package. The HTM4 Mk1
design allowed four complete transmit/receive
(T/R) functions to fit into a 1 x 1 inch package.
Further evolution of the multichannel module
package resulted in a single multilayer HTTC
substrate and a copper-molybdenum thermal
plane with a kovar ring frame and lid, thereby
reducing both cost and weight. HTM4 modules
evolved into what is known today as the tile
module. The tile module has RF and DC inputs
and outputs on different planes (opposite
sides) that are assembled into the array in a
tiled x-y matrix. Interconnections between
the module, radiating aperture and control
circuitry are made by spring pin interconnects
rather than more traditional RF connectors or
wire bonds, allowing for simplification of the
array assembly.
The AESA’s next packaging evolution was the
development of arrays that did not rely on
hermetic modules to package the T/R functionality. Module-less TRIMMs were developed
where the RF and DC routing was accom-
Major thrusts during the 1990s were launched to
reduce cost and increase performance of modules
and arrays. Advances in MMIC technology yielded
chips with greater power and more on-chip
capabilities, resulting in a reduced numbers
of chips per module.
1 Moore’s law is the observation that the number of transistors in an integrated circuit doubles approximately every two years.
Increased computer processing and memory capabilities are strongly linked to the density of transistors in an integrated circuit.
FEATURE
TRIMM Features
• Liquid (higher power density) or air cooled
• Packaged face up or flipped MMICs
• Small number of circuit layers
• Primarily touch labor assembly
• Drives the AESA depth
CCA Features
• Liquid (medium power density) or air cooled
• Packaged face up or flipped MMICs
• Large number of circuit layers
• Automated fabrication, pick and place
• Enables a thinner AESA
Figure 5. Evolution from transmit/receive integrated multichannel module (TRIMM)
to circuit card assembly (CCA) packaging provides size, weight and cost reductions.
plished using advanced printed wiring boards
(PWBs). PWB technology had advanced to
a point where high performance RF laminates could be combined with traditional DC
laminates to produce multichannel substrates
suitable for AESA applications. These TRIMMbased module-less arrays were first developed
with channel counts ranging from four to
60. These TRIMMs consisted of a multilayer
RF/DC PWB populated with a mixture of
traditional surface mount technology (SMT)
devices and chip and wire RF devices. These
substrates could include either radiating elements or simple nonhermetic RF connectors.
Environmental protection for initial applications was provided via a sealed enclosure. For
future systems, Raytheon is developing coating
technologies to eliminate the need for a sealed
enclosure.
Circuit card assembly (CCA) arrays that
incorporate advanced printed circuit radiating
elements were developed to provide reduced
antenna depth over the module-less TRIMMs
(Figure 5). The CCA approach uses stacked
layers to produce arrays which are exceptionally thin and lightweight. Advanced RF/DC PWB
substrates are used to produce subarrays with
channel counts more than 1,000. Typically,
small daughter cards with either flip-chip or
upright components that contain the RF components are used in CCA construction. The
daughter cards are attached either mechanically
or through SMT soldering to a larger mother
board containing the DC distribution network, capacitors and other components. The
aperture, RF, DC, control layers and cold plates
are all mechanically fastened together to form
a single CCA subarray. Subarrays can then be
joined to create a full array of the needed size.
The development of GaN-based MMICs has
required new packaging to handle the higher
operating voltages and increased thermal
loads. The higher power levels obtainable
with GaN-based TRMs allow arrays to be
greatly reduced in overall size and element
count, however, this reduction has resulted
in unprecedented power levels and thermal
loads to be dissipated at each module. As such,
hermetic housings constructed of new materials with higher thermal conductivities are being
developed, including mixed-metal composites,
nano-material based thermal interface materials and ultra-high performance cold plates.
Power and Cooling
As with packaging, AESA thermal management and power conditioning has changed
dramatically during the last 30-plus years.
The RF devices in an AESA can generate large
heat flux levels, and the thermal management
of AESA systems has spurred development of
higher performance liquid flow-through cold
plate technology, thermal interface materials, and controlled CTE packaging materials.
Arrays in the 1980s used heat pipes to reduce
the temperature rise between the electronics
and the ambient environments. As packaging
approaches evolved to higher densities, thermal
management solutions evolved to using smaller
air or liquid heat sink assemblies.
Liquid cooling is typically the choice for
thermal management of high performance
AESA-based electronics, and is the most efficient and affordable approach to maintaining
the required MMIC and TRM temperatures
needed to achieve high reliability and electrical
performance. Some liquid-cooling solutions
provide the cooling channels directly under
the TRMs or MMICs and others conduct the
heat away from the TRMs or MMICs to a
liquid cooled manifold. For further cooling
efficiency, new liquid-cooling designs position
the fluid even closer to the MMICs using integrated nano- and micro-size channels in planar
volumetric cooling architectures and novel heat
spreader designs. Air-cooled approaches are
also used for some applications, especially for
small low-power devices. Air-cooled approaches are often simpler and lower cost versus
liquid-cooled designs and can be used when
the system requirements allow it.
Power conditioning technology has also improved over the years with the development
of planar magnetics and smaller devices. Many
of the smaller AESAs use centralized power
conditioning due to the proximity opportunities of the supply and AESA. The larger AESAs
must distribute the power conversion electronics in order to minimize the size and weight
of the conductors and to improve efficiency.
The use of higher voltage GaN allows smaller
conductors to be used and improves the overall
power efficiency allowing significantly lighter
power conditioning systems.
Summary
Raytheon, including legacy Hughes, Texas
Instruments Defense and E-Systems companies, provides unparalleled AESA research,
development, manufacturing and product
experience for surface, airborne and space
missions in radar, communications and EW
systems. Working with our government investment partnerships for more than 30 years,
Raytheon has pioneered and matured AESAs
and their dependent technologies into fielded
solutions for our nation’s warfighters and
allies. Raytheon continues to be an industry
leader in developing advanced technologies
such as high-power GaN MMICs, integrated
RF electronics packaging and interconnects,
more efficient power and thermal conditioning
solutions and new subarray and element level
digital beamforming and signal/data processing
techniques to continue providing affordable
solutions for a diverse customer base. •
Mike Sarcione, Porter Hull, Colin Whelan,
Doug Tonomura, Thomas V. Sikina, Jim Wilson
and Robert E. Desrochers II
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
13
FEATURE
State-of-the-art RF Semiconductors
FOR MILITARY SYSTEMS
M
odern military systems require the
highest performance radio frequency
(RF) semiconductor technology.
For example, in radars in transmit mode, RF
energy needs to be distributed to each active
antenna array element, precisely phase shifted
and then greatly amplified before being radiated, and in receive mode, the small return signal
must be amplified with great fidelity. Likewise,
electronic warfare (EW) and communication
systems require the same sort of transmit and
receive functionality with even more emphasis
on signal fidelity in both transmit and receive
modes. The final active array amplification
of transmitted RF signal at each element is
performed by a compact RF power amplifier
(PA) circuit chip that is a type of monolithic
microwave integrated circuit (MMIC). This
PA technology must be small enough to fit
within the radar’s element spacing, be able to
generate high RF output power and also be
very efficient in converting direct current (DC)
power into RF power to minimize prime power
consumption and waste heat generation. In
receive mode, the first stage of amplification of
the reflected signal is performed by a low noise
amplifier (LNA) MMIC which must increase
the desired signal strength while minimizing
additional noise degradation and the introduction of nonlinear distortions. The phase shifter
MMIC within each radar element allows dynamic beam steering of the radiated RF energy.
Raytheon, leveraging its own trusted foundry
(Figure 1), has a long and successful legacy of
developing next generation, high performance
semiconductor processes for PA, LNA and
phase shifter MMICs as well as for inserting
them into highly reliable, fielded, phased array
radars. This MMIC design and fabrication
capability coupled with expertise in semiconductor material growth, MMIC-based module
development, and integration and test, provide
a leading edge RF MMIC capability that supports current and emerging Department of
Defense (DoD) system needs at a lower cost
and shorter timeline than alternative approaches. Specifically, today’s commercial RF
capabilities, that are driven primarily by commercial technology demands, cannot meet the
most stressing DoD system requirements.
Semiconductors for Combined RF and
Logic Functionality
Traditionally, gallium arsenide (GaAs) has
been the semiconductor of choice for efficiently amplifying and phase shifting RF
signal in radars. Throughout the 1990s,
Raytheon was a pioneer in inserting GaAsbased MMICs into the first modern phased
array radars. As the performance requirements
of these military systems have increased to
meet ever-growing threats, so too have the
Figure 1. Raytheon’s trusted semiconductor foundry facility fabricates gallium arsenide and gallium nitride monolithic microwave
integrated circuits.
14 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
FEATURE
Raytheon Process
Description
Radar Functionality
GaAs pHEMT
Traditional PAs and
LNAs
GaAs pHEMT with
logic
Multifunction MMICs
Heterogeneous
integration
RF with complex logic
or control
mHEMT
Higher performance
LNAs
GaN
Higher performance
PAs
During the past six years under independent research and development (IRAD) and
Defense Advanced Research Projects Agency
(DARPA) funding, Raytheon has taken the
integration of RF and logic functions to the
next level. Raytheon has developed technology
to directly integrate GaAs, indium phosphide
(InP) or gallium nitride (GaN) RF devices with
high density Si complementary metal oxide
semiconductor (CMOS) logic on a common
Si substrate. This heterogeneous integration
technology enables greater levels of functionality and digital control of RF circuits as well as
a reduction in cost over using the traditional
separate RF and logic chips.
Table 1. Raytheon’s customized semiconductor processes optimized for each radar
function.
power, efficiency and low noise requirements
on these MMICs. During that time, Raytheon
has continued to customize and optimize its
semiconductor processes for each specific
radar function (Table 1). One such Raytheon
GaAs pseudomorphic high electron mobility
transistor (pHEMT) technology, customized
for multifunction MMICs containing phase
shifters, attenuators and gain stages, combines
RF and logic functions on the same MMIC,
providing a serial or parallel logic interface
to a separate silicon (Si) controller chip. By
designing a process that allowed the combining of some logic functions on the RF MMIC
circuitry, the number of off-chip components
and interfaces between chips was minimized,
reducing size and cost while improving radar
manufacturability.
Using this integration approach, Raytheon
demonstrated the world’s first GaN HEMT
and Si CMOS heterogeneously integrated chip;
a GaN RF amplifier with in-situ Si CMOS
gate bias control (Figure 2). This circuit was
a proof of concept that demonstrated that
GaN HEMT and Si CMOS devices could
be integrated on the same Si substrate with
minimal performance impact to the Si CMOS
and GaN HEMT technologies. The circuit also
serves as a building block for digitally assisted
RF and mixed signal circuits, such as amplifiers
with on-chip digital control and calibration,
reconfigurable or linearized PAs with in-situ
adaptive bias control, high-power digital-toanalog converters (DACs) and on-chip power
distribution networks.
GaN
HEMT
Si
CMOS
Figure 2. Optical micrograph of the first
GaN and Si CMOS heterogeneously integrated MMIC.
Semiconductors for Low Noise
Amplification
For low noise amplifications such as what is
used in radar receivers, Raytheon developed
high indium-content metamorphic HEMT
(mHEMT) technology and manufacturing processes for higher gain performance with even
lower noise contribution than traditional GaAs
HEMT technology that is limited to 19 percent indium in the channel region where the
electrons travel. Previously, only InP HEMT
devices with high indium content (53 percent)
in their InGaAs quantum well channels could
fill this ultimate low noise function, but at a
higher cost due to the challenges associated
25 GHz Fmin (2 x 50 μm)
1.0
)
cale
ft S
e Le
(Us
8,000
0.8
6,000
GaAs
4,000 Devices
ale)
t Sc
igh
se R
(U
0.6
InP
Devices
0.4
2,000
0.2
Fmin 19% In pHEMT
Fmin 33% In mHEMT
1.25
Fmin (dB)
10,000
Quantum Well Depth (eV)
Mobility of Electons (cm2/V-s)
1.5
1.2
12,000
Fmin 60% In mHEMT
1
0.75
0.5
0.25
mHEMTs
0.0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Indium Channel Content (In(x)Ga(1-x)/As)
Figure 3a. With metamorphic high electron mobility transistor
(mHEMT) technology, as indium (In) is added to the quantum well
channel, both electron mobility and quantum well depth increase.
0
0
50
100
150
200
250
Current (mA/mm)
Figure 3b. As indium (In) is added to the quantum well channel,
the minimum noise figure (Fmin) decreases.
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
15
FEATURE
State-of-the-art RF Semiconductors for Military Systems
with manufacturing MMICs on fragile InP
wafers.
mHEMT technology, with up to 60 percent
indium composition in the channel, offers the
gain and low noise performance advantage
of InP HEMTs and the manufacturability
and cost advantages of GaAs MMIC wafers.
Through the growth of a metamorphic
material layer on a GaAs substrate, the lattice constant of GaAs can be transformed to
allow the growth of higher indium content
InP HEMT devices on GaAs wafer substrates.
Additionally, metamorphic growth allows classically forbidden indium contents to be grown,
i.e., those not lattice matched to GaAs or InP
substrates, thus enabling the device designer
to explore and exploit the properties of a new
set of HEMT devices. Figure 3a shows that the
measured mobility of channel electrons
increases with increasing channel indium
content, due mainly to the reduction in
electron effective mass and reduced scattering
in the deeper quantum well. These improvements give rise to higher electron channel
velocities, allowing operation at higher gain,
higher frequency and lower noise figure than
GaAs devices. Figure 3b shows the reduction
in minimum noise figure, Fmin1, achieved as
one moves from a traditional GaAs pHEMT
device with 19 percent indium content to that
of a mHEMT with a 60 percent indium content
InGaAs channel. When fabricated into LNAs,
these mHEMT MMICs greatly improve the
signal-to-noise ability of the system, extending its range while reducing prime power over
GaAs technology as the high mobility enables
them to operate at lower DC bias.
Semiconductors for High-Power
Transmission
For improvement on the transmit side,
Raytheon developed and matured GaN
technology for PAs, enabling the next generation of DoD systems. The power, efficiency
and bandwidth performance of GaN-based
MMICs is unsurpassed, revolutionizing the
design of radars by creating not only higher
performance but also lower system cost. With
over 5 Watts/millimeter (W/mm) of RF output
power density compared with GaAs at 1 W/
mm, GaN RF amplifiers deliver five times the
power per element compared to GaAs within
the same footprint, providing up to 50 percent
more radar range or the ability for the radar
to search five times the volume of space in the
same amount of time (Figure 4). Fewer highpower GaN MMICs could be used to replace
many low-power GaAs MMICs, decreasing
radar size by half while maintaining search
TRACK
SEARCH
Search volume with GaAs
Range using GaAs
Radar Array Face
1.5x Range using GaN
Reference
target size
5x volume with GaN
Identical
target
Radar Array Face
Range Window
1) Five times the search volume in the same
amount of time.
2) 50 percent increase in range with the same
sensitivity.
OR
APERTURE SIZE
Search volume with GaAs
Reference area with GaAs
50 percent area with GaN
Radar Array Face
Greater volume with GaN
3) 50 percent reduction in antenna size with greater search volume.
Assumes GaN power = 5x GaAs power
Figure 4. GaN technology provides significantly improved radar performance compared to GaAs at reduced size and cost.
1
Fmin is a measure of signal to noise during RF signal amplification.
16 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
FEATURE
Log – Normal Failure Distribution
Failure Distribution
(Arbitrary units)
0.6
0.5
0.4
0.3
0.2
30 year mission
requirement
0.1
0.0
1.E+3 1.E+4 1.E+5 1.E+6 1.E+7 1.E+8 1.E+9 1.E+10 1.E+11 1.E+12
Time (Hr)
Figure 5. Calculated failure distribution at 150 °C based on measured
accelerated GaN device testing. Lifetimes, denoted by the failure
distribution, greatly exceed the 30-year DoD system mission life.
performance and increasing efficiency or, alternatively, equal-power GaN
chips can be made dramatically smaller in size. Thus, GaN can reduce
overall system costs while enabling new smaller size-constrained systems.
The higher drain current that GaN offers makes the broadband matching
of high-power EW MMICs simpler and more efficient than GaAs, while
the approximately seven-fold improvement in the thermal conductivity
enables efficient amplifier cooling. Finally, the wide bandgap that is intrinsic to GaN material provides large critical breakdown fields and voltages,
making a more robust amplifier and easing system implementation.
Recently, Raytheon was honored by the Office of the Secretary of Defense
(OSD) for their successful completion of the Defense Production Act
(DPA) Title III GaN production improvement program, culminating over
a decade of government and Raytheon investment in GaN technology.
Raytheon also demonstrated that the reliability of our GaN technology
exceeded the 30 year mission requirement for insertion into production military systems (Figure 5). This maturation of GaN resulted in a
manufacturing readiness level (MRL2) of 8, the highest level obtained by
any organization in the defense industry for this technology. Also, through
the Title III program, Raytheon demonstrated that GaN MMICs could
generate RF power for one third the cost of GaAs MMICs, enabling higher
performance systems at a lower cost.
Future Semiconductor Technology
Raytheon continues to pioneer the development and manufacturing of
cost-effective, next generation RF semiconductor technologies to deliver
needed capabilities for current and future defense electronic systems.
One such technology for power amplifiers is polycrystalline diamond
which provides very high thermal conductivity, allowing GaN to reach
its full potential in RF power density and compactness while maintaining
reliable operating temperatures. For next generation receivers, InGaAs
with indium contents exceeding 60 percent are being explored to further
improve noise figure and lower DC power consumption. All of Raytheon’s
semiconductor research technologies strive to achieve higher system performance for the warfighter at lower costs than existing capabilities. •
E N G I N E E R I N G P R O F I L E FEATURE
Steven
Bernstein
Principal
Engineering
Fellow, Advanced
Technology
Programs
Steven Bernstein
is a program and
capture manager
at the Raytheon
Integrated Defense
System Advanced
Technology Programs
organization with
more than 25 years
of experience in
semiconductors,
electronic devices,
transmit/receive modules, transmit/receive integrated multichannel
modules (TRIMMs) and radar systems. He is currently responsible
for radar component technology development including both RF
component and power system technologies. “Through the years,
my focus has changed,” Bernstein acknowledges, “from optical
and electronics materials development to radar component technologies … My interest in developing technology, however, has
always remained the same.”
Bernstein is the Raytheon program manager for several Defense
Advanced Research Projects Agency (DARPA) and Office of Naval
Research (ONR) programs as well as several Raytheon independent
research and development activities. These include the DARPA
Near Junction Thermal Transport (NJTT) program which is involved
in the development of gallium nitride (GaN)-on-diamond technology, the DARPA Wide BandGap Semiconductor (WBGS) Phase 3
program which is developing GaN-based T/R modules and the
ONR Compact Power Conversion Technology program which is
developing the next generation of power system technology.
Previously, he was the Raytheon program manager for the DARPA
Wide BandGap High Power Electronics program and several thermal management programs. He received the Raytheon Excellence
in Technology award for his work on GaN technology.
Earlier in his career, Bernstein was on the staff of Raytheon’s
Research Division. His work at the Research Division focused on
gallium arsenide (GaAs) process engineering, the development
and demonstration of novel electronic ceramic thin-film devices
based on high temperature superconductors and ferroelectric
materials, and the development of novel infrared window solutions
for high speed missiles.
When asked about some of the more exciting aspects of his job,
Bernstein said, “I work many different programs and technologies
so every day is different. I am also excited when we take a risk on
something new and accomplish something that has never been
done before.”
Colin Whelan, Nicholas Kolias, Joseph Smolko,
Thomas Kazior and Steven Bernstein
2 MRL is a measure used by the DoD and many of the world’s major companies to assess the
maturity of manufacturing readiness.
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
17
FEATURE
COMPUTATIONAL IMAGING TECHNOLOGY
Revolutionizing Digital Imaging
New capabilities allow for the practical application of exciting computational imaging
technologies that will revolutionize digital imaging systems to meet the expanding sensing needs
of our customers.
I
maging technologies of the 21st century
are advancing at a fast pace. Key technical advances are being made daily in the
performance and capabilities of imaging detectors, high-definition displays, high-bandwidth
networks and distributed high-performance
computing. These new capabilities allow for the
practical application of exciting new imaging
technologies that are the synergistic fusion
of optics and detectors, and allow the use of
computing capabilities and algorithms that will
revolutionize the field of digital imaging. This
emerging technology area, known as computational imaging, has the potential to provide our
customers with the actionable information
they actually need, instead of drowning them
in data and hoping they can find the desired
information amongst all the raw imagery.
As a leader in developing and manufacturing
advanced electro-optical (EO) imaging systems
for a wide variety of intelligence, surveillance
and reconnaissance (ISR), security and effects
targeting needs, Raytheon is exploring the powerful potential emerging from the application
of computational imaging technologies.
Addressing Current Sensing
Systems Needs
For the past four centuries, imaging systems
were designed as two separate entities: the image forming components (i.e., optics), and the
image detection components (i.e., eye or film).
The transition to electronic focal plane array
detectors as a replacement for the human eye
or silver halide film has done little to change
this design architecture. However, during the
past decade, the rapid increase in practically
available computational power has led many to
explore whether there might be an advantage
to the joint design of these optical detection
systems.
18 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
Computational imaging is an emerging new
technology area, one with the potential to dramatically impact EO sensing system capabilities over the coming decade. Computational
imaging refers to the synergistic combination of
traditional and nontraditional optics, detectors and computational resources in order
to achieve gains in system performance (e.g.,
increased detection ranges; decreased system
size, weight and cost; additional and improved
extraction of desired information from scenes
beyond just imagery). These system attributes
are the key technical performance measures
customers are looking for in determining
solutions for their sensing system needs.
The essence of computational imaging is the
concept that there are significant gains to be
had from inserting the computational step
earlier into the image acquisition and formation chain — not just by performing processing
post-image formation — to produce improved
and specialized forms of scene information.
This scene information can be purpose
rendered to meet the operational information needs of the end user. The application
of computational imaging technologies to
current-day ISR, security and targeting systems
can potentially provide improved performance
for many technical attributes:
• Reduced physical size and weight of
imaging systems.
• Reduced demands on high-bandwidth
communications/networks.
• Improved performance attributes relating
to multispectral and hyperspectral data, threedimensional imagery, image spatial resolution
and dynamic range.
• Improved image exploitation attributes
relating to automated target cueing, scene
change detection, detection and tracking of
objects of interest and improved situational
awareness over large areas under surveillance.
Exploring New Sensing Capabilities
Raytheon businesses span a diverse base of EO
sensing products, covering a broad range of
military, civil and international markets.
Computational imaging technology similarly
spans a wide spectrum of potential capabilities and applications. Key application areas
where the use of computational imaging design
concepts and principles could potentially be
employed include:
• Ultra-wide field-of-view (FOV) imaging
for ISR and security applications.
• Spatial resolution enhancement for ISR,
targeting and security applications.
• Novel multispectral and hyperspectral
imaging for ISR and security applications.
• Reducing the size, weight and cost of optics
and imagers for hand-held applications.
• Digital processing to focus the operators’
attention on image content of interest
to reduce data overload, increase the speed
of understanding and improve efficiency
and effectiveness.
As much of the innovative research and development activity in this technology area occurs
within the academic environment, Raytheon is
collaborating with leading research universities
to cooperatively explore computational imaging. In particular, Raytheon has been collaborating with Duke University, Rice University
and the University of Arizona. Additionally,
Raytheon has been collaborating with several
small technology businesses identified through
the Small Business Innovation Research (SBIR)
program. These collaborative research efforts
have focused on exploring these new technologies and their potential applications.
FEATURE
During these joint investigations, a wide range
of potential applications were discovered. In
particular, three areas were concentrated on for
technical investigation:
• Computational resolution enhancement.
• Computational hyperspectral sensing.
• Compact imaging optics.
Computational Resolution
Enhancement
In a current trend mirroring that of consumer cameras, the pixel count in EO imaging
systems continues to increase. The driving
requirement for this is the need for greater
scene spatial resolution. One example of the
use of computational imaging technologies
for this paradigm is being explored by a team
led by Duke University as part of the Defense
Advanced Research Projects Agency (DARPA)
Advanced Wide FOV Architectures for Image
Reconstruction and Exploitation (AWARE)
program. The Duke University team has constructed wide FOV multiscale visible imaging
cameras capable of producing diffractionlimited images as large as 10 gigapixels. An example of the high-resolution images generated
by a Duke University 2-gigapixels prototype
visible camera is shown in Figure 1. This is an
exciting new imaging capability that has many
applications in the surveillance, security and
entertainment domains.
Not all EO sensing applications, however, can
afford or require the size, weight, power and
bandwidth resources required by a gigapixelclass imaging camera. For these applications,
Raytheon has been investigating techniques for
collecting high resolution imagery with fewer
actual pixels in the focal plane; a research area
termed “compressive sensing.” Two different
approaches are under development:
• The single pixel camera imaging design
concepts explored by Rice University.
Through the use of a fast dynamic scene
modulator, these imaging cameras use a low
number of physical imaging detectors (less
than 50) to capture compressed imagery
containing up to a megapixel.
• The coded aperture camera imaging
design concepts pioneered by Duke
University and the University of Arizona.
Through the use of a high resolution coded
aperture located in an intermediate image
plane, imagery can be generated with an effective resolution of three times greater than the
physical focal plane array in each dimension.
Figure 1. Prototype gigapixel camera is capable of providing very wide field-of-view images while also providing very high image resolution and scene detail (photos courtesy of
Duke University).
Computational
imaging techniques
applied to hyperspectral sensing systems
potentially provide
data similar to the
traditional designs at
a much lower size,
weight, power
and cost.
Computational Hyperspectral Sensing
Raytheon has a significant technology presence with conventional hyperspectral sensing
systems, ranging from the spaceborne Visible
Infrared Imager Radiometer Suite (VIIRS) meteorological sensor and the Advanced Responsive Tactically Effective Military Imaging Spectrometer (ARTEMIS) to the Airborne Cueing
and Exploitation System Hyperspectral (ACES
Hy) tactical surveillance sensor. All of these
units are traditional spectrometer designs.
In collaboration with Duke University,
Raytheon is evaluating the design concepts
for a coded aperture hyperspectral sensor.
This sensor concept, known as a coded
aperture snapshot spectral imager (CASSI), is
a snapshot multispectral or hyperspectral imaging sensor that utilizes a coded optical
aperture with a dispersing prism or grating.
The resulting dispersed spectra are overlaid
on a staring focal plane array, and recovery
of the image is performed by using an itera-
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
19
FEATURE
Computational Imaging Technology is Revolutionizing Digital Imaging
tive data recovery algorithm. Figure 2 shows
a prototype CASSI camera operating in the
shortwave infrared (SWIR) spectral region
(1–1.7 m). Simulated hyperspectral imagery
is also shown in this figure along with the
anticipated processed results compared to
measured spectral feature data.
0.025
LAKE
0.02
0.015
0.01
0.005
0
-0.005
0.1
Compact Tactical Imaging Optics
Raytheon is known as a major provider of
tactical hand-held thermal imaging systems to
support warfighter effectiveness and increase
their survivability in combat operations and
environments. Examples of current and
developmental tactical situational awareness
and targeting systems include the family of
Thermal Weapon Sights (TWS), PhantomIR
and the Integrated Day-Night Sight Technology (IDNST) imager.
0
10
20
30
10
20
30
10
20
30
BUSH
0.08
0.06
0.04
0.02
0
0.012
0
ROAD
0.01
0.008
0.06
0.004
0.02
One of the major contributors to the size,
weight and cost of these smaller military handheld imaging systems is the optics. Using computational imaging concepts, optical design
alternatives for tactical situational awareness
imagers were explored by Duke University,
the University of Arizona and the University of California San Diego through their
participation in the DARPA Multiple Optical
Non-redundant Aperture Generalized Sensors
(MONTAGE) program. As illustrated in Figure
3, these activities yielded several innovative
lightweight optical designs suitable for tactical
imagers. The Duke University team developed
0
0
Figure 2. Coded aperture snapshot spectral imager (CASSI) hyperspectral imaging camera
is capable of providing multispectral and hyperspectral imagery in a compact package
suitable for tactical ground and airborne platforms.
an optical design that was only 5 millimeters
thick and is significantly lighter than conventional optical design approaches. The University of Arizona and University of California
San Diego team developed a compact folded
optical design that has been evaluated by the
U.S. Army Night Vision and Electronic Sensors Directorate (NVESD) for use on the TWS
family of thermal night sights. These research
and evaluation activities continue in an effort
to provide enhanced tactical imaging solutions
to warfighters.
Future Directions
Raytheon has been an active industry participant in the evaluation and development of
computational imaging technologies over the
past decade. Continuing these technical evaluation, development and prototyping activities
will provide the objective insight to make
informed technology deployment and investment decisions in the coming years. Working
closely with university and technology-based
small business partners, Raytheon continues
to develop enhanced image generation and
exploitation capabilities that meet current
and future customer needs. •
Randy Gann and Bob Gibbons
DARPA MONTAGE Multi-aperture Lens
DARPA MONTAGE Folded Imaging Optics
Figure 3. Compact lightweight optical systems were developed under the DARPA Multiple
Optical Non-redundant Aperture Generalized Sensors (MONTAGE) program to significantly
reduce the size and weight of tactical imaging systems (Photos courtesy of Duke University
[left] and the University of California San Diego [right]).
20 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
FEATURE
BAREMETAL:
a New Cybersecurity Technology
The U.S. government and cleared
defense contractors rely heavily on a
commodity information technology (IT)
infrastructure to plan and coordinate
business practices and perform data
processing. Unfortunately, commodity
IT equipment is often manufactured
in untrusted environments (e.g., foreign factories), shipped by untrusted
companies and received by potentially
malicious insiders, exposing new attack
surfaces and putting the U.S. at risk
for attack.
While many organizations attempt to secure
commodity desktop and server systems with
personal security products and hardened
operating systems, there is no verification
that adversaries have not exploited their firmware through supply chain or remote attacks
to facilitate cyberespionage or to disrupt
operations.
All commodity IT equipment uses firmware
to initialize system components and to load
the operating system. In desktops, laptops
and servers, basic input/output system (BIOS)
firmware has served this purpose. In recent
years, however, unified extensible firmware
interface (UEFI) firmware has started replacing
legacy BIOS firmware. This new UEFI standard
enables the development of highly modularized and well-formatted firmware, which allows
developers to construct higher quality firmware
more rapidly. The UEFI standard also lowers
the barrier to entry for adversaries to inject
low-level malware and exploit low-level vulnerabilities.
Raytheon Pikewerks has developed a small,
portable device called BareMetal that can be
used by a nonspecialist to acquire, process
and characterize computer firmware in a
matter of minutes. BareMetal, shown with the
cover removed in Figure 1, focuses on threats
to UEFI and option read-only memory (ROM)
firmware used to initialize peripherals like
video and graphics cards. The device consists of
an ARM-based1 computer-on-module (COM),
a self-contained power source and light emitting diodes (LEDs) to inform users of analysis
results and errors. BareMetal is intended for
operators who know little about firmware or
firmware threats. Operators need only configure a machine to boot off BareMetal and the
device does the rest. Thus, BareMetal can be
used to inspect firmware during provisioning
and before the equipment is deployed within
the IT infrastructure.
Figure 1. The BareMetal device (shown with
cover removed) contains processor, storage,
user interface and power supply elements
packaged into a small (3.0 x 2.1 x 1.2 inch)
form factor enclosure.
BareMetal can also be deployed in the form of
an agent to provide situational awareness of
unauthorized firmware modifications across
an enterprise. A server-side component communicates with the agent to acquire firmware
and process it to determine if any unauthorized
changes have been made.
Additionally, BareMetal is designed to cooperate with the Intel® Active Management
Technology (AMT) so it can receive messages
when a computer’s chassis is opened, when a
computer unexpectedly reboots, or when a new
peripheral device is inserted into the motherboard. All these events are considered possible
indicators of a firmware or hardware attack.
BareMetal, in its enterprise form, is designed
to perform the following actions:
• Monitor the contents of the firmware using
applications and/or out-of-band mechanisms on the device under test (DUT).
• Monitor events that can indicate attempted
firmware modifications using applications
and/or out-of-band mechanisms on the
DUT.
• Use server-side components to communicate with the sensors on the DUT, collecting
the data and processing it to verify firmware
consistency or to determine whether a
mission is ongoing to modify firmware.
• Present firmware health to an administrator using a security information and event
management (SIEM) solution.
Whenever a new security capability is integrated into IT operations there is a cost associated
with purchasing the capability, integrating it
into existing security processes, and training
personnel to use the capability properly. Since
BareMetal is intended for nonspecialists and
its analysis is completed in less than two
minutes, it can be added to the receiving and
provisioning process with minimal impact both
to staffing and provisioning time. This cost
efficiency, along with the relatively inexpensive
purchase cost of the BareMetal device, allows
organizations to cost effectively improve the
security of their IT system by introducing
BareMetal into their supply chain risk management procedures. •
Adam Fraser, Joe Tanen and Eric Egalite
1 ARM refers to a family of reduced instuction set computer based (RISC-based) processors designed and licensed by British company ARM Holdings.
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
21
FEATURE
NEXT GENERATION
EO/IR Detectors
Electro-optical/infrared (EO/IR) system capabilities are continuously being expanded and improved. This is in part due
to the development of high-performance focal plane array (FPA) technology. Raytheon Vision Systems (RVS) is actively
advancing the state-of-the-art for new and discriminating FPA technologies in the 3–5 micrometer (µm) mid-wavelength
infrared (MWIR) and the 8–12 µm long-wavelength infrared (LWIR) spectral regions by using a variety of semiconductor materials and FPA architectures. These include advanced III–V1 semiconductor strained-layer-superlattice (SLS) FPAs,
II–VI2 semiconductor mercury cadmium telluride (HgCdTe) dual-band FPAs and uncooled bolometer FPAs using microelectromechanical systems (MEMS). Each of these FPA technologies address different mission requirements and are being
developed to not only increase sensor capability and performance, but also to provide improved manufacturability and
reduce costs.
Advanced III–V Materials
WIR FPAs with higher operating
temperatures are advantageous for
reducing the size, weight and power
of cooled systems. For this reason, high operating temperature (HOT) FPAs are being developed based on molecular beam epitaxy (MBE)
grown indium arsenide/gallium antimonide
(InAs/GaSb) strained-layer-superlattice (SLS)
bandgap engineered barrier device structures.
The goal is to develop materials that have the
producibility advantage of traditional III–V
indium antimonide (InSb) technology, but the
operating temperature advantage of HgCdTe.
The InAs/GaSb family of SLS is the only known
IR detector material with fundamental properties that provide a theoretically predicted
higher performance than HgCdTe (lower dark
currents at the same wavelength and temperature, with a comparable absorption coefficient).
This SLS family has features whereby unique
device architectures that provide performance
advantages can be realized using multiple
lattice-matched layers with a wide range of
bandgap energy on a GaSb substrate as shown
in Figure 1. These III–V semiconductors have
greater latitude in process parameters (such as
temperature) and tolerance to process-induced
defects, and there is also a broader industrial
base such that larger substrates and merchantsupplier MBE growth houses are available
to produce the material and contribute to
the technology development. Together these
factors offer the promise of improved performance, higher yield and reduced cost for FPAs
fabricated on SLS material.
M
2.5
0.5
GaP
Ga Ga Ga Ga Ga Ga
Sb Sb Sb Sb Sb
AlAs
0.6
2.0
Ga Ga Ga Ga Ga Ga
Sb Sb Sb Sb Sb
Ga Ga Ga Ga Ga Ga
Bandgap (eV)
Sb Sb Sb Sb Sb
0.8
AlSb
1.5
InP
GaAs
1.0
1.0
0.5
1.5
GaSb
GaInAsP
Ga 0.47 In0.53 As
GaInAsSb
3.0
InSb 5.0
10.0
InAs
0
5.5
5.7
2.0
5.9
6.1
6.3
In In In In In In
As As As As As
In In In In In In
As
As
As
As
In As
Interface
As
In In In In In In
As As As As As
Ga Ga Ga Ga Ga Ga
GaSb
Interface
InAs
Sb Sb Sb Sb Sb
GaSb
Ga Ga Ga Ga Ga Ga
Sb Sb Sb Sb Sb
Ga Ga Ga Ga Ga Ga
Sb Sb Sb Sb Sb
SiO2
6.5
Lattice constant (A)
Figure 1. The semiconductor bandgap versus lattice constant graph on the left shows
various binary compounds that can be used to create, for example, an InAs/GaSb strainedlayer superlattice (SLS) structure (middle) whose bandgap can be varied by the thickness
of the individual layers. The picture on the right is a high resolution transmission electron
microscopy (TEM) image of an SLS structure which shows the amorphous nature of the
SiO2 passivation layer in contrast to the atomistic nature of the crystalline InAs/GaSb SLS.
22 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
Dual-band FPAs
Raytheon has developed an industry leading
dual-band HgCdTe infrared FPA architecture
(Figure 2) that is in production for advanced
missile interceptor applications, and is being further refined and expanded as a third
generation upgrade to existing tactical systems
for ground and airborne applications. The
availability of spectral information from a
second spectral band enables the discrimination of absolute temperature and unique
signatures of objects in a scene (Figure 3).
When coupled with advanced signal processing algorithms, two-color infrared detection
provides improved sensitivity compared to that
of single-color devices. These advanced FPAs
are necessary in order to improve acquisition,
discrimination and the tracking of ballistic
missile warheads, or the identification and
engagement of tactical battlefield targets. This
FPA architecture utilizes MBE-grown HgCdTe
on 6 x 6 centimeter (cm) cadmium zinc telluride (CdZnTe) substrates, but there is also a
significant development focus on using 6-inch
diameter silicon (Si) substrates to substantially
lower cost. Achieving the highest dual-band
FPA performance, particularly for the LWIR
band, is more challenging when using a Si
1
The III–V semiconductors are compounds composed of
elements from the IIIA and VA CAS-standard groups of the
periodic table or equivalently groups 13 and 15 based on the
modern group-numbering standard.
2
The II–VI semiconductors are compounds composed of
elements from the IIB and VIA groups of the periodic table
based on the CAS group-numbering standard. These are
groups 12 and 16, respectively, using the modern groupnumbering standard.
FEATURE
is illustrated in Figure 4, which compares the
uncooled LWIR images of the Santa Barbara
Mission taken with 640 x 480 and 2048 x 1536
pixel FPAs. Note that these images were taken
at different times of the year so the details
around the mission vary, e.g., the truck on the
left in the top image was gone when the bottom
image was taken.
Unit Cell
Indium Bump
Contact
Band-2 Absorber
P-Type Cap Layer
Band-1 Absorber
CdZnTe Substrate
Shorter
Wavelength
Longer
Wavelength
Figure 2. The left figure shows a schematic cross-section of Raytheon’s single-contact,
single-mesa dual-band HgCdTe detector architecture, and the right shows a scanning
electron microscope image of dual-band detectors with indium bumps on individual
diode mesas.
MWIR
substrate due to the very large lattice and
thermal mismatch between Si and HgCdTe
that contributes to elevated material dislocation densities. Improvements in device design
and material growth makes the goal of low cost
dual-band HgCdTe FPAs on 6-inch diameter Si
substrates attainable.
eters which are fabricated monolithically on
200 millimeter (mm) Si read-out integrated
circuit (ROIC) wafers. Over the last few years,
this technology has been further enabled by
Raytheon partnering with a high-volume commercial semiconductor company, Freescale
Semiconductor, to fabricate the bolometers.
Uncooled FPAs
LWIR uncooled FPAs employing MEMS-based
microbolometer structures are now widely
used in both commercial and U.S. Department
of Defense (DoD) applications. By its nature,
uncooled technology is lower cost because it is
operated at room temperature and can utilize
simpler, smaller packaging without the need for
a cooler. Additionally, it is fabricated on large
area Si substrates using commercial manufacturing technology. Raytheon uncooled technology uses vanadium oxide (VOx) microbolom-
Typical uncooled FPA formats use 640 x 480
pixels or smaller with a pixel size of 25 µm.
Raytheon, in partnership with the Army Night
Vision and Electronic Sensors Directorate
(NVESD) and the Defense Advanced Research
Projects Agency (DARPA), is also developing
three mega-pixel-class uncooled FPAs with a
2048 x 1536 format for very large format, high
resolution sensors.
An example of the dramatic resolution improvement using this very large format FPA
LWIR
Figure 3. Example of imaging using a MWIR/LWIR dual-band FPA produced using HgCdTe
grown on a low-cost Si substrate as part of an Army Night Vision and Electronic Sensors
Directorate (NVESD) development program. The image from the MWIR band on the left
shows transmission through the MWIR filter but not the LWIR filter, while the LWIR image
on the right shows the opposite effect.
Additionally, RVS in partnership with DARPA
is developing wafer scale manufacturing processes to enable uncooled camera-on-a-chip
technology as part of the Low Cost Thermal
Imager – Manufacturing (LCTI-M) program.
The goal of this effort is to dramatically lower
the cost of this technology to enable wide use
of infrared imaging technology and insertion
in compact systems.
Uncooled LWIR 640 x 480
Uncooled LWIR 2048 x 1536
Figure 4. The dramatic resolution improvement using large format FPAs is illustrated
in the images above that compare an uncooled LWIR image of the Santa Barbara
Mission taken with a 640 x 480 pixel FPA
(top) with an uncooled 2048 x 1536 pixel
FPA (bottom).
Through these and other activities, RVS continues to advance the state-of-the-art in FPA technology to achieve higher performance, lower
cost and smaller size EO/IR sensor systems. •
S.M. Johnson and E.P.G. Smith
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
23
FEATURE
QUANTUM COMPUTERS:
Big and Small
Q
uantum information
processing is enjoying
a new level of excitement
as researchers learn to
engineer quantum systems.
The Quantum Information
Processing (QuIP) group at
Raytheon BBN Technologies
is tackling both the long-term
problems of building tools
for the design and analysis of
large scalable quantum computers and the more immediate
problems of implementing
record-breaking, proof-ofconcept devices to demonstrate the power of quantum
devices in computation, imaging and communication tasks.
Introduction
On a fundamental level, quantum computing
is a quest to harness the ultimate processing
power allowed by the laws of physics. On a
practical level, it offers the promise of massive improvements in the time to solve certain
problems of extreme import. Most famously,
an implementation of Shor’s algorithm1 could
factor large numbers with computational
resources proportional to a polynomial function of the key length. This polynomial-time
factoring capability would have huge consequences as it could enable attacks on many of
today’s public encryption algorithms whose
security relies on the need for an exponentially
large amount of computing power to crack
the encryption code. Other examples where
quantum computing could have a significant
impact include the use of an algorithm called
Grover’s algorithm2 to speed up brute force
search tasks and the development of exponentially more efficient algorithms for calculating
the dynamics of complex many-body systems.
Recent work by Raytheon has identified how
quantum processors can improve imaging and
communication capabilities both at optical and
radio wavelengths.
Quantum processors require completely new
physical implementations, where quantum bits
or “qubits” replace the traditional bits encoded
as voltages in a complementary metal-oxide
semiconductor (CMOS) circuit. Like classical
bits, qubits have two logical states (“0” and
“1”). However, they differ in that they can be
manipulated into any of a continuum of superpositions, whereby they exist simultaneously in
mixtures of the two states with a well-defined
relative phase between them. These superpositions probabilistically collapse purely into “0”
or “1” upon measurement (or readout) of the
qubit. The power of quantum computation
comes about when one maintains the superpositions during the entire algorithm, measuring
the state of the system only to obtain the final
answer at the end.
Analogous to traditional computation, quantum computation requires a universal set of
one and two qubit logic gates in order to perform any quantum processing algorithm. Also
analogous to classical bits sent over communication networks, quantum systems can implement error correction codes through the use
of large collections of physical bits to encode
a single logical qubit. However, these codes
must be implemented in very particular ways
so as to avoid measuring the logical quantum
information directly, which would destroy the
superposition and thus the quantum advantage. In order to realize quantum computation,
a technology must be able to perform a universal set of gates with sufficiently low error rates
to be fault tolerant with suitable error correction. Additionally, one must be able to scale the
technology to large systems of qubits, each with
its own control and read-out circuitry.
Quantum Computer Implementations
There are several candidate qubit implementation approaches, each possessing its own set
of challenges to obtain sufficiently low error
quantum gates and scaling to system sizes
needed for real-world applications. These
include ionic atoms trapped by electric fields,
neutral atoms confined by lasers, photons
traveling through integrated circuits, nitrogenvacancy color centers in diamond, and
electron spins in semiconductors. In addition,
Raytheon, along with IBM, Northrop Grumman and several academic institutions around
the world, is developing quantum processors
in superconducting circuits using Josephson
Junctions. Circuits comprised of superconducting metal and cooled below their critical
temperature can support the dissipationless
flow of electron pairs (Cooper pairs) and act
in a quantum coherent manner. Josephson
1 Shor’s algorithm is a quantum computing algorithm, named after mathematician Peter Shor, that efficiently calculates
the prime factors of an integer.
2
24 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
Grover’s algorithm is a quantum computing algorithm, named after computer scientist Lov Grover, that efficiently searches an unsorted
database.
FEATURE
Junctions are thin resistive layers fabricated into
the circuits that provide a nonlinear behavior,
allowing them to act as qubits. Information is
encoded in the direction and pattern of electron
current traveling through the circuits.
Over the past five years, the superconducting
approach has seen very significant progress,
demonstrating one and two qubit logic operations with low error rates (<0.2 percent and
<1 percent, respectively). One of the main
reasons for this progress is that error rates are
often driven by unwanted decoherence of the
qubit (often decay into the “0” state) after a
short period of time. These qubit lifetimes have
recently improved by three orders of magnitude, to greater than 100 microseconds. This
provides enough time to perform more than
1,000 operations and is thus sufficient for fault
tolerant operation. Researchers are now beginning to focus on the challenges associated with
hardware scalability. Raytheon is collaborating
with IBM on a number of scalability topics
as part of the Intelligence Advanced Research
Projects Activity (IARPA) Multiqubit Coherent
Operations (MQCO) program.
Inset 1 in Figure 1 shows a superconducting
chip containing three qubits and resonators
that couple the qubits together and enable measurement, i.e., read-out, of the qubit state. The
chip is put into a dilution refrigerator (large
cylinder in Figure 1) which cools the chip to
below 50 milli-Kelvin (-272.95 °C and just 0.05
degrees above absolute zero) where coherent
quantum operations can be executed. Recently,
Raytheon demonstrated gates with approximately a 0.2 percent error rate, approaching the
fault-tolerant threshold for scalable computation. Current work is focused on the design
and fabrication of an eight qubit device using a
design scalable to much larger sizes.
Quantum logic gates and the read-out of qubits
for these superconducting devices is done by
applying radio-frequency pulses at frequen-
cies in the range 5–10 GHz. This requires that
scalable quantum processors include flexible, reliable and high performance waveform
generators of reasonable cost. Toward this end,
Raytheon has developed the arbitrary pulse
sequencer (APS) shown in Figure 1, Inset 2.
The APS boxes provide 14-bit, 1.2 giga-samples
per second (GS/s) analog output, allowing the
sophisticated pulse shaping necessary for high
fidelity gates and supporting sequences of up to
8,192 pulses. The Raytheon developed firmware
for the APS boxes provides easy and rapid
implementation of complex quantum gate
sequences.
Quantum Optical Communications
In parallel to the superconducting quantum
processor effort, Raytheon has been carrying out research on quantum optical receiver
implementations with an objective of understanding and quantifying the ultimate limits
of communication and imaging system
information transfer using optical technoloInset 2
Inset 1
Resonators
Qubits
Figure 1. A chip (Inset 1) containing co-planar waveguide resonators and superconducting circuits serving as quantum bits (qubits) is
cooled in a dilution refrigerator (blue cylinder in photo) so that coherent quantum operations can be executed. The customized arbitrary
pulse sequencer (APS) device (Inset 2) is a high performance waveform generator developed by Raytheon BBN Technologies to control
qubit experiments.
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
25
FEATURE
gies. Through the Defense Advanced Research
Projects Agency (DARPA) Information in a
Photon (InPho) program, Raytheon has been
exploring the ultimate information efficiency
of light. This research involves identifying
the maximum power efficiency, in bits per
photon, allowed by quantum physics on a
free space optical communications channel as
well as in imaging systems such as laser radars
(LADARs). So far, two important conclusions
have come from this work: 1) the maximum
bits per photon can be much greater than one
when multiple spatial or temporal degrees of
freedom are employed (though there is a clear
trade-off between degrees of freedom used
and photon efficiency); and 2) the maximum
power efficiency (for a given number of degrees
of freedom) is much greater than even the
ideal limit obtainable with any current optical
receiver approach such as direct, homodyne or
heterodyne detection.
Motivated by the theoretical result, Raytheon
BBN Technologies proceeded to identify physical implementations of these quantum optical
Quantum Computers: Big and Small
receivers to demonstrate the ability to surpass
the information transfer limits of current detection systems. An example of such a receiver
is the green machine which can improve the
photon efficiency for a binary phase shift keyed
communication system from its current maximum of 1.44 bits/photon to a much higher,
potentially unlimited, value. However, there
is a fundamental trade-off which imposes a
corresponding decrease in the rate of information transfer in a given bandwidth (spectral
efficiency).
Any loss or scatter in an optical receiver results
in wasted power and thus a reduction in
photon efficiency. Thus, one of the challenges
of implementing optical receiver devices at the
quantum limits of efficiency is the necessity for
highly accurate alignment and low losses. Implementation in large-scale or fieldable systems
will likely require implementation in integrated
optical waveguide devices made from low loss
material such as silicon. Raytheon collaborated
with the optical nano-photonics group at Oak
Ridge National Labs to fabricate and test an
integrated silicon waveguide implementation
of the green machine, shown in Figure 2. It is
an eight input and eight output device containing 12 beam splitters and encompassing an
area of approximately 10 square millimeters
(mm2). By comparison, such a device implemented via bulk optics would require an area
of approximately 300 square centimeters (cm2)
and would likely not be capable of meeting
the phase alignment requirements. One of the
most interesting results derived from this effort
was that ultimately devices with the computational power of a universal quantum computer
will be required to reach the power efficiency
limits possible using quantum optical communications and imaging methods. This is another
interesting new potential application area for
quantum processing, i.e., high performance
military communications and imaging.
Quantum Computer Tools
and Programming
As advances in the physical implementation of
quantum processing devices have progressed,
the need to develop high level tools to under-
Optical Fiber Inputs
B
Beam
Splitter
P
Phase
Splitter
Figure 2. A 10 mm2 silicon integrated optical circuit built collaboratively between Raytheon and Oak Ridge National Laboratory. This
circuit can be used to process received optical communications at capacities well beyond the classical Shannon limit.
26 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
FEATURE
stand and manage the complexity of large multiple-device systems has
become apparent. For example, with quantum computation it becomes
important to understand the impact of low-level device control issues
on the fault-tolerance of larger multi- device systems and to be able
to describe large and complex algorithms in a clear, compact manner,
abstracted from the physical details of the machine implementation.
Under IARPA’s Quantum Computer Science (QCS) program, Raytheon
has developed a Quantum Functional Programming Language (QuaFL)
specifically designed for implementing quantum algorithms. An important feature of QuaFL is the automatic inclusion of constraints imposed
by the laws of quantum physics that govern the operation and error
model of the quantum processing device. Like traditional functional
programming languages, QuaFL provides familiar abstractions to free
the programmer from thinking in terms of physical circuits, while at
the same time guaranteeing that forbidden operations such as copying
data, which is not allowed due to the no-cloning theorem of quantum
mechanics, are reported as compilation errors. Also under this program,
Raytheon BBN Technologies has done extensive work optimizing
error correction codes used for quantum computing. In particular,
analytic results on one of the best known quantum error correcting
codes, the surface code, were extended to show that the surface code
method is a favorable approach in terms of resource consumption and
its implementation on realistic qubit architectures. This is an extremely
important result in scalable quantum computation because the physical
qubit resources required to implement logical qubits is a big driver in
the complexity of a full quantum processor. It is also important because
it allows optimization and resource estimation of systems that are much
too complex to perform in the way they are done on the relatively small
systems in laboratories today.
The Future
The development of quantum processors at a scale interesting to applications will be a long but exciting process. What is particularly exciting
today is the rapid progress in the understanding and manipulation of
certain physical qubit implementations, including the Josephson Junction superconducting circuits being developed by Raytheon BBN Technologies and other laboratories around the world. This progress has
spawned new research focused on scaling these building blocks to large
system implementations. Raytheon has so far focused its research on
designing multiqubit architectures, developing scalable radio-frequency
control and read-out electronics, developing programming languages
to compile large-scale quantum algorithms, and finding interesting new
applications of quantum processing in communications and imaging.
Additional important and difficult challenges also on the horizon include reliable, large-scale fabrication of qubits and the efficient characterization and debugging of large-scale devices. Recent progress in these
areas indicates that quantum processing will experience rapid development in the U.S. and around the world over the next several years and it
is expected to eventually be adopted as a powerful and game-changing
computing technology. •
Zac Dutton, Ph.D., and Marcus Silva
Contributors: Blake Johnson, Monika Patel, Saikat Guha,
Colm Ryan, Thomas Ohki and Jonathan Habif
We acknowledge support for this work from the IARPA MQCO program under contract
W911NF-10-1-0324, the IARPA QCS program under contract number D11PC20166,
and the DARPA InPho program under contract number HR0011-10-C-0162.
E N G I N E E R I N G P R O F I L E FEATURE
Zachary
Dutton, Ph.D.
Raytheon BBN
Technologies
Dr. Zachary Dutton
has been manager of
the quantum information processing
business unit at
Raytheon BBN
Technologies since
July 2012. His work
focuses on improving
communications and
sensing techniques
utilizing quantum
optics effects. He
was the BBN principal investigator (PI)
on the Defense Advanced Research Projects Agency (DARPA)
Quantum Sensors program, investigating quantum enhanced
LADAR, and the DARPA Quiness program, focusing on developing optical quantum key distribution methods that achieve higher
rates and continental scale distances. Additionally, he has
worked on architectures for superconducting circuit based quantum computation in collaboration with experimental colleagues
developing these systems.
“Our customers are focused on investigating novel quantum
effects seen in physics laboratories.” He states. ”We try to understand if and how they can enhance current communications,
sensing and computation systems, and focus on the engineering
challenges in realizing these enhancements. As a manager, I get
to meet and interact with our customers on a regular basis to
understand their goals and work with them to achieve those
goals. And in my more technical role, I get to propose and
research solutions to these difficult challenges.”
Dr. Dutton received his doctorate in theoretical atomic physics
from Harvard University in 2002. In his thesis work, he performed
seminal work on coherent nonlinear optical effects in cold atomic
ensembles and Bose-Einstein condensates, including electromagnetically induced transparency, slow light and coherent optical
storage. He was a National Research Council post-doctoral fellow
at the NIST-Gaithersburg and a staff physicist at the Naval
Research Lab before joining Raytheon BBN Technologies in 2007.
“My academic training involved the odd situation of working as
a theorist in a primarily experimental group. I believe being
forced to work in such close collaboration with experimentalists
and having to analyze real experimental data (as opposed to
ideal theoretical models) gave me a very pragmatic, problemsolving approach which has served me well throughout my
research career. It has allowed me to quickly broaden to new
topics and attack new problems. I believe this approach has even
served me well in the business and management aspects of my
current role, which requires quick decision making and decisive
action based on the best available data, which can often be
incomplete or approximate.”
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
27
FEATURE
Particle Flow Filters to Solve Near
IMPOSSIBLE PROBLEMS
The EKF uses a linear approximation of the
system dynamics and works well for certain
applications, but it can give surprisingly bad
accuracy for systems with difficult nonlinear or
non-Gaussian noise dynamics (Table 1). Particle filters (PFs) have the promise of optimal
accuracy for arbitrary nonlinear non-Gaussian
problems, but at a huge cost in real-time
computational complexity. A PF approximates
the probability density of the state vector using
numerical values at points in the state space
(so-called “particles”), and hence it can represent highly non-Gaussian densities by using
enough particles. A PFF moves the particles to
the correct regions in the state space to represent the probability density accurately using
physics rather than Monte Carlo simulations.
It is much better to use 1,000 correctly chosen
particles in the state space rather than 10 million randomly or poorly selected particles in
the wrong locations.
The PFF can be several orders of magnitude
more accurate than the EKF for difficult nonlinear dynamical system problems that may
even contain non-Gaussian noise (see Figure
1). In addition, the PFF is many orders of
magnitude faster than standard PFs for three
reasons: 1) PFFs use many fewer particles than
PFs to achieve optimal accuracy; 2) the PFF
software implementation is much faster than
the PF implementation; and 3) PFF computations can be parallelized for efficiency. In
particular, standard PFs require millions or
billions of particles to achieve the same accuracy as the PFF for high dimensional problems
(e.g., Figure 1).
Systems engineers can exploit the particle flow
filter in several ways: 1) lower system cost owing to less expensive sensors or less expensive
computers; 2) achieve better system accuracy;
3) increase system capacity to track more targets or targets at longer range or stealthier tar-
ITEM
EXTENDED KALMAN FILTER (EKF)
STANDARD PARTICLE FILTER (PF)
PARTICLE FLOW FILTER (PFF)
What is it?
Linear approximation
of nonlinear
dynamics and nonlinear
measurements
Monte Carlo approximation
of non-Gaussian probability
density
Exact computation of
non-Gaussian probability
density using physics rather than
Monte Carlo simulation
Estimation accuracy for linear
Gaussian problems
Optimal
Optimal if you use enough
particles
Optimal if you use enough
particles
Estimation accuracy for certain
(easy) nonlinear problems
Nearly optimal
Optimal if you use enough
particles
Optimal if you use enough
particles
Estimation accuracy for difficult
Very poor
nonlinear non-Gaussian problems
Optimal if you use enough
particles
Optimal if you use enough
particles
Real-time computational
complexity
Extremely fast (even for high
dimensional problems)
Extremely slow for high dimensional problems
Many orders of magnitude faster
than standard particle filters for
high dimensional problems
Representation of the probability
density of the state conditioned
on the set of all measurements
Mean and covariance matrix of
Gaussian density
Arbitrary non-Gaussian
density using numerical values
at N points in state space
(so called “particles”)
Arbitrarily smooth, nowhere
vanishing, non-Gaussian density
using particles
Table 1. Attributes of the Kalman filter, particle filter and particle flow filter.
28 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
gets; 4) use much smaller computers which can
be used in applications requiring less volume,
less power or less weight, such as airborne or
space-based sensors or systems requiring very
compact computers for mobility or stealth;
and 5) solve problems that were previously
thought to be impossible to solve.
The PFF is three to four orders of magnitude faster per particle because it avoids the
bottlenecks of standard PFs, and, in addition,
it uses three or four orders of magnitude
fewer particles. For example, for a typical
radar tracking problem the filter estimates
the position and velocity of a ballistic missile,
107
VELOCITY ERROR METERS/SECOND)
T
he Raytheon particle flow (nonlinear)
filter (PFF) provides superior system
performance and lower cost for tracking, prediction, navigation, guidance, control,
robotics, multiple sensor data fusion, autonomous decisions and learning, image processing, communications, weather forecasting,
compressive sensing, multi-input multi-output
(MIMO) communication systems, MIMO
radar systems, MIMO navigation systems,
cryptography, cyberwarfare, medical diagnosis,
science and many other important applications. A filter is an algorithm that estimates the
state of a dynamical system using a set of noisy
measurements. For example, we use Kalman
filters (KFs), and for nonlinear systems
its extended Kalman filter (EKF) variant, to
locate the position of your car using noisy
range measurements provided by multiple
Global Positioning System (GPS) satellites. KFs
and EKFs were invented more than 50 years
ago, and they are used for object tracking and
estimation in essentially all modern radars,
sonars, optical systems and other sensor
systems.
106
105
N = 1,000 particles
100 Monte Carlo trials
20 dB SNR
104
Standard Particle Filter
103
102
101
100
0
Extended Kalman Filter
Particle Flow
20
40
60
80
100
TIME (SECONDS)
Figure 1. Particle flow filter beats the
extended Kalman filter velocity estimation
accuracy by an order of magnitude for long
range radar tracking of ballistic missiles.
hence, the dimension of the state vector is six.
A standard PF typically requires 10 million
particles to obtain optimal accuracy for a six
dimensional state space, whereas the PFF needs
only one thousand particles, for a savings of
four orders of magnitude in computer speed.
Moreover, the PFF is typically three to four
orders of magnitude faster per particle computation than a standard PF, and therefore the
net reduction in computer throughput is seven
to eight orders of magnitude without parallel
processing. This means that problems can now
be solved that engineers thought were impossible to solve because the computer would be
much too expensive or much too large for a
given application using standard PFs.
FEATURE
ENGINEERING PROFILE
Fred Daum
Principal
Engineering
Fellow
Standard PFs suffer from the curse of dimensionality and particle degeneracy, whereas the PFF solves these problems through particle flow. The
root cause of the PF problem is Bayes’ rule, which is the computation of
the probability density of the state vector using each measurement. In
particular, the probability density after the measurement is the product
of two densities: 1) the probability density of the state vector before the
last measurement and 2) the density of the last measurement given the
state vector. The basic idea of our new algorithm is to compute Bayes’
rule using particle flow equations borrowed from physics rather than
the standard approach of multiplying two functions. The particle flow
is designed by solving a partial differential equation (PDE), analogous
to solving Maxwell’s equations for antenna design. In fact, our PDE is
the first of Maxwell’s equations, i.e., the divergence of the electric field is
equal to the charge density, and the rich history in physics of solving such
equations is exploited to reduce the computational complexity of PFF
computations. For example, we use incompressible flow and irrotational
flow as well as Coulomb’s law to move the PFF particles. Incompressible
flow is an excellent approximation for subsonic flight in air, which was
invented by physicists and mathematicians more than 200 years ago, and
it greatly simplifies the PDE used in fluid dynamics to design aircraft. Our
PFF uses the exact same idea. However, filter designers do not need to
solve PDEs in order to design PFFs. Rather, they can use several derived
implementation methods (18 at latest count) to solve the PFF PDE equations, resulting in algorithms like the ubiquitous KF and EKFs that are
easily programmed on computers. This approach is analogous to antenna
designers who do not attempt to solve Maxwell’s equations from scratch,
but rather open the handbook of antenna design and use the relevant solution with some good approximations (e.g., far field, narrowband, small
aperture approximations). There are many different kinds of antennas,
and each one uses a different set of approximations, which is analogous to
our 18 distinct methods to solve the PFF PDE.
Standard PFs attempt to repair the damage done by Bayes’ rule using
Monte Carlo methods by resampling particles (i.e., throw away the useless
particles and sample new particles to replace the bad ones), but with very
limited success. This approach is analogous to a basketball player taking
shots at the basket randomly with her eyes closed; if the ball misses the
basket, then she takes another shot, and another, and another, as in a
Monte Carlo method (e.g., Metropolis-Hastings). Such methods work for
low dimensional problems, but they fail catastrophically for high dimensional problems, because there are so many (wrong) directions in high
dimensional space; this is called the curse of dimensionality. In contrast,
the PFF keeps its eyes on the basket and calculates the predicted trajectory
of the particles using physics. Physics is much better than blind, random
guessing.
In summary, Raytheon’s PFF is applicable to a range of mission problems and can be expected to pay dividends in system performance and/or
reduced complexity. •
Frederick E. Daum
is an IEEE Fellow
and a Principal
Engineering Fellow
with Raytheon
Integrated Defense
Systems. He has
been with Raytheon
for more than 44
years. Daum is a distinguished IEEE
lecturer and a graduate of Harvard
University. He was
awarded the Tom
Phillips prize for technical excellence in recognition of his ability
to make complex radar systems work in the real world.
When asked about interests and what got him started along
his career path, Daum talks about his time as an undergraduate
in college, when he first heard about Raytheon, “I was told that
I could work on interesting, diverse high-tech radar systems
with Raytheon; this had great appeal for me, and it turned out
to be completely true. I was lucky to be given challenging tasks
and large responsibility from the very beginning of my career.”
As a radar systems engineer, Daum developed, analyzed and
tested real-time algorithms for essentially all the large long
range phased array radars built by the United States in the last
four decades, including Cobra Dane, the Precision Avionics
Vectoring Equipment Phased Array Warning System (PAVE
PAWS), Cobra Judy, the Ballistic Missile Early Warning System
(BMEWS), the Terminal High-Altitude Area Defense (THAAD)
system, the Relocatable Over-the-Horizon Radar (ROTHR), the
Upgraded Early Warning Radar (UEWR) and the Sea-Based
X-Band (SBX) Radar, as well as several shipboard fire control
systems and air traffic control systems. He developed and tested
algorithms in radar waveform scheduling, Bayesian discrimination, data association, discrimination of satellites from missiles,
calibration of tropospheric and ionospheric refraction and target
object mapping. Daum’s exact fixed finite dimensional nonlinear
filter theory generalizes the Kalman and Beneš filters.
Daum is elated to do research in nonlinear filters. “I get to
work with the world’s experts in radar system engineering,”
he states. “I have the freedom to try out new ideas, and I get
to work on extremely interesting and challenging technical
problems.”
Daum has published nearly 100 technical papers and has given
invited lectures at the California Institute of Technology,
Massachusetts Institute of Technology, Technion (Israel)
Institute of Technology and multiple other universities and
research facilities in the United States and internationally.
Fred Daum
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1 29
FEATURE
AUTOMATED LANGUAGE TRANSLATION
Breaking the Language Barrier to Cross-lingual Information Access
Accessing Information From
Foreign Content
More than half the content on the Internet is
in a language other than English, and three out
of four Internet users are not native speakers
of English. Although much useful information — some of it critical to economic success
and physical security — can be gleaned from
these foreign language sources, the value of this
information often decreases over time. This
presents a challenge because current translation by humans is too slow and too expensive
to provide the quick, reliable access to foreign
language information that governments and
businesses need.
Foreign language translation is changing,
however. Advances in automated language
translation technology have opened the possibility of breaking the language barrier for
both information access and in-person communication. Figure 1 illustrates the central
role that automated language translation plays
in enabling communication and information
access across different languages. The goal of
an automated language translation system is to
ingest sentences in a source language and produce a correct, fluent, semantically equivalent
sentence in the target language.
History of Automated Language
Translation
Despite the difficulty of automated translation problems, early attempts were made to
tackle it dating back to the 1950s. Figure 2
shows a timeline of the major milestones in
the evolution of automated translation. The
prevailing approach in the early decades was to
analyze the structure of the input sentence and
determine the possible senses of its ambiguous
words, and then apply translation rules crafted
by expert linguists to generate the translation.
A drawback of this rule-based approach is a
lack of flexibility for adding new translation
rules and ensuring consistency with the existing rules. Moreover, using this approach to
build a translation system for a new language
pair requires a linguist who is an expert in both
languages. Also, the process of writing the requisite set of translation rules is a slow, difficult
and labor-intensive undertaking. These are
major disadvantages.
A data-driven approach to automated translation began in the early 1990s. Instead of specifying translation rules manually, this methodology uses a parallel corpus that consists
of sentences from the source language along
with their translations in the target language.
Example sentence translations from the corpus
are used to derive automatically a large set of
translation rules between smaller units (e.g.,
words or phrases), together with an associated likelihood for each rule. The rules and
likelihoods are then applied to translate a new
input sentence from the source language. This
approach, called statistical machine translation
(SMT), revolutionized automated language
translation by enabling translation systems for
new languages and domains to be developed
quickly and cheaply. With the advent of SMT,
the need for linguists who are experts in two
or more languages to design translation rules
manually is no longer necessary. Translation
rules are now automatically derived from
sentences translated by bilingual speakers
who are not necessarily linguistic experts.
This improves development time and reduces
translation cost.
Early SMT models focused on learning the
translations of individual words. Thanks to
the increased availability of data and cheap
computational power, more complex models
that can learn the translation of phrases or
syntactic structures are being developed. The
current approach to SMT, shown in Figure 3,
uses statistical models dependent on linguistic
information and context to produce translations that preserve sentence structure. SMT is
still a fledgling technology despite the significant advances made over the last two decades.
Current research continually incorporates
advances in machine learning theory and linguistic modeling to improve the state of the art
in SMT technology.
Automated Language Translation
Challenges
The automatic translation process is easy to
SMT Research at Raytheon
depict, but difficult to achieve. A general propBBN Technologies
erty of all human languages is the prevalence of
Research in machine translation started at
ambiguity in the meaning of individual words
Raytheon BBN Technologies in 2003. In two
as well as in the relationship between parts of a
sentence. Humans are usually very efficient at
TWO-WAY
INFORMATION
TRANSLATION WORKFLOW FOR
resolving these ambiguities when interpreting
SOURCES
HETEROGENEOUS DOCUMENTS
SPEECH-TO-SPEECH
linguistic input, often without being aware of
their existence. They rely on past experience
and the context surrounding the speech or text
to perform the task. This knowledge is very
AUTOMATED LANGUAGE TRANSLATION
hard to model in a computer system. Without access to such knowledge, an automated
language translation system still has to meet the
BROADCAST AND SOCIAL
HARDCOPY DOCUMENT
challenges of selecting the correct translation
MEDIA MONITORING
TRANSLATION
of a word, rearranging the translated words according to the grammar of the target language,
and producing a correct and natural-sounding Figure 1. Automated language translation enables translingual communications and
translation.
information access.
30 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
FEATURE
years, Raytheon established itself as a leading
player in the field of language translation by
leveraging its experience in statistical modeling
for speech recognition. Most of the research
was done under Defense Advanced Research
Projects Agency (DARPA) sponsored programs
that had a great impact in advancing SMT
technology. The alignment of these programs
to the major milestones in SMT evolution can
be seen in Figure 2. Between 2005 and 2011,
Raytheon BBN Technologies participated in
the DARPA Global Autonomous Language
Exploitation (GALE) program, whose goal was
to develop technologies to absorb, analyze and
interpret huge volumes of speech and text in
multiple languages. Raytheon was consistently
ranked top performer in the program’s official
evaluations organized by the National Institute
of Standards and Technology (NIST). Raytheon was also the top performer in DARPA’s
Spoken Language Communication and Translation System for Tactical Use (TRANSTAC)
program, which aimed to develop technology
for real-time speech-to-speech translation from
English to a foreign language and vice versa.
As part of these and other research programs,
Raytheon made several significant innovations
to improve the state of the art of SMT. For example, Raytheon BBN Technologies developed
a translation model that produces translations
with improved semantic coherence by using information about the grammatical relationship
between words that occur far apart in sentences. Raytheon also developed a procedure
for combining the outputs of multiple automated translation systems to produce a better
translation than any of the individual outputs.
Pre-DARPA GALE and TransTac
1994
2000
Rule Based Word Based
Phrase Based
Language dependent
Labor-intensive rules
Requires expert linguists
Case-by-case approach
Source-Target Corpus
Trainer
Training
Phase
Syntax,
Context,
Semantics
Statistical Model
Input in
Source Language
Translator
Translation
Phase
Output in
Target Language
Figure 3. Statistical machine translation uses a large collection of translated sentence pairs
to develop statistical models that then translate source language text.
Raytheon researchers have likewise developed
techniques for detecting names and handling
names properly in translations, and for using
confidence scores on the alignment between
phrases to improve the translation quality.
Raytheon BBN Technologies is currently part
of the DARPA Broad Operational Language
Translation (BOLT) program. DARPA
launched BOLT in 2011 to address the U.S. Department of Defense’s need for quick, reliable
access to the large volume of foreign language
information generated by users online. One of
the program’s goals is to create SMT technologies that can correctly translate informal text
generated by online users, which often contains
spelling and grammatical anomalies. Another
goal is to deal with the problem of communicating with non-English-speaking local populations in foreign countries in person. Raytheon
researchers have made significant progress in
the short time since the program started, in-
GALE and TransTac
2006
DARPA BOLT
2012
Hierarchical, Syntax Driven
Context, Semantics Driven
Language independent
Statistical modeling
Automatic training on parallel corpus
Rapidly portable
Figure 2. Automated language translation approaches have evolved from initial rule-based
approaches to more current statistical methods. This evolution has been helped, in part,
by Defense Advanced Research Projects Agency (DARPA) program initiatives such as the
Global Autonomous Language Exploitation (GALE) program, the Spoken Language Communication and Translation System for Tactical Use (TRANSTAC) program and the Broad
Operational Language Translation (BOLT) program.
cluding developing abilities to robustly process
errors in input text, better model syntax and
semantics, and improve the statistical models
using neural networks. Advances in speechto-speech translation include modeling of a
conversation’s context and detecting speech
recognition errors during translation to limit
any harmful effects on the translation output.
The BBN team ranked first in the formal evaluations of all BOLT machine translation tasks.
Raytheon BBN Technologies Automated
Translation Solutions
In addition to conducting leading-edge research in SMT, Raytheon BBN Technologies
has created several turnkey solutions for both
the government and commercial markets based
on its translation technology. TransTalkTM, a
two-way speech-to-speech translation solution,
runs completely on a smartphone without
the need to call a remote server. It currently
supports translation between seven languages
(including Arabic, Pashto and Dari) and English, and it has been deployed for testing in Afghanistan. The Multimedia Monitoring System
(described in Technology Today, 2012, Issue
2, pp. 52–55) uses BBN’s SMT technology for
some of the foreign languages it supports. The
Multilingual Document Analysis and Translation System (MDATS) uses BBN’s optical
character recognition and SMT technologies to
translate Arabic document images into English.
All these systems have been deployed in a number of government locations for 24/7 use. •
Rabih Zbib, John Makhoul and Walt Andrews
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
31
FEATURE
Optimization Algorithms for
DECENTRALIZED PLANNING AND CONTROL
Introduction
Current U.S. Department of Defense (DoD)
missions for unmanned vehicles typically
require one dedicated pilot to control each remote vehicle. The DoD’s long-range vision is to
do more with less; pushing for more autonomy
in the appropriate phases of the mission. Its
priorities include:
• Developing trusted highly-autonomous
decision-making systems that enable
the replacement of, in whole or in part,
current human-intensive functions in order
to maintain a decisive combat advantage.
• Pursuing technologies and policies that
introduce a higher degree of autonomy to
reduce the manpower burden and reliance
on full-time high-speed communications
links while also reducing decision loop cycle
time.
• Developing collaborative autonomous
capabilities that enable a team of unmanned
systems to coordinate their activities to
achieve common goals without human
oversight.
• Developing algorithms that perform autonomous mission optimization under dynamic
circumstances, including dynamic task reassignment.
• Developing a decentralized system of
heterogeneous unmanned systems relying on
network-centric, decentralized control that is
flexible in its level of autonomy.
There has already been a significant amount
of research done in the area of autonomous
vehicle control for surveillance-type missions.
Almost all of that research has dealt with
centralized cooperative control, with little
research addressing the decentralized control
problem where multiple autonomous vehicles
are allowed to collaboratively work together to
achieve mission objectives.
In fact, the development of decentralized
cooperative planning and execution strategies
remains an open area of research and is the
focus of this article.
32 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
Decentralized Planning and Control
for Unmanned Aerial Vehicles
In principle, decentralized control offers advantages over centralized control that include
alleviation from single point of failure events,
lower latency sharing of information between
participants, and a more even distribution of
the information processing and mission planning burden among participants. A relatively
simple benchmark problem was created to help
develop and evaluate different decentralized
control algorithms and to help understand
their advantages and disadvantages relative to
each other and to centralized approaches.
The benchmark problem was to develop a collaborative planning and control approach that
utilizes unmanned aerial vehicles (UAVs) to
provide: 1) continual surveillance over a region
of interest; and 2) accurate tracking of all
targets detected in the region. It was assumed
that all UAVs were flying at the same fixed
altitude. The UAVs were modeled as point-
Start
mass objects that moved on a two-dimensional
fixed altitude plane and had a fixed minimum
turning radius. The targets were either moving
or stationary, and the UAVs did not know the
target dynamics.
The UAVs were tasked with collaboratively
searching the environment for targets and
tracking the detected targets. For each planning window, each individual UAV needed to
dynamically determine which assigned tasks it
should perform, in what order to perform the
selected tasks, and what trajectory it should fly
to accomplish its tasks. Each UAV’s planning
and control algorithm needed to consider
sensor field-of-regard constraints, vehicle kinematic constraints, communication bandwidth
constraints and possible target dynamics.
For this example, vehicles were assumed to
operate in a decentralized manner. At each
time step, each UAV independently executed
the same set of tasks as shown in Figure 1. Each
Move UAV
Receive neighboring UAV
positions, current
trajectories, sensor
characteristics,
uncertainty maps, and
sensed target maps
Score UAV
assessment
of world
Small
Score?
No
Create fused target
picture and updated
search uncertainty map,
i.e.,i generate UAV
assessment of world
End
Update time
t = t +Δ t
new
Yes
Solve cooperative
search and track
optimization
problem over
planning horizon
Update UAV
task
assignment
and trajectory
Figure 1. Each unmanned aerial vehicle (UAV) starts with an initial search and track plan.
Then each UAV executes the same set of dynamic planning tasks at each new performance
assessment time step using its own and neighboring UAV sensor information. If the UAV’s
assessment of the current plan is poor (indicated by a low score), then it executes a replanning set of tasks to change the plan and improve the predicted performance. We note that
the start and end designations clarify the starting and ending points of the decision loop at
each time-step.
FEATURE
UAV created its own assessment of the world,
based on on-board sensors and communication with neighboring UAVs. Each UAV then
determined whether its current plan moving
forward was still appropriate, based on its
assessment of the world. If it was not appropriate, then the UAV planning software was used
to solve a coupled tasking and route planning
optimization problem for itself and its neighbor UAV. The solution to this problem resulted
in updated tasking, including new flight routes
for itself and each neighboring UAV. Each of
the UAVs might determine different solutions as compared with their neighbors. This
is due to the fact that each UAV might have a
different perspective, and hence assessment of
the world. Currently, this solution mismatch
is rectified when each UAV receives information from neighbor UAVs. Future research will
involve incorporating alternative deconfliction
approaches.
The collaborative UAV search and track problem is a generalization of a standard vehicle
routing problem (VRP). While there are many
variants to the standard VRP, the general
form considers m vehicles starting at a depot
and tasked with delivering goods or services
to a set of n customers. Each customer has a
certain required demand, and each vehicle has
a maximum capacity. The VRP is known to
be NP-hard1, and thus by extension the UAV
planning problem is NP-hard as well.
To efficiently develop a solution to the UAV
search and track optimization problem, a
hybrid set of algorithms is used that combines
both greedy randomized adaptive search procedures (GRASP) and simulated annealing (SA)
techniques. GRASP is a multistart local search
procedure, where each iteration consists of two
phases; a construction phase and a local search
phase. In the construction phase, interactions
between greediness, i.e., moving the solution
in the local direction that yields the largest performance improvement, and randomization,
i.e., moving the solution estimate in a random
direction, generate a diverse set of quality solutions. Then in the local search phase, the state
space around the solutions generated in the
construction phase are further optimized.
SA is a method to find good-quality solutions
to optimization problems by using a process
that mimics the cooling process of metals. At
each step, a current solution is perturbed. If the
perturbation results in a better solution, then
the current solution is replaced. If the perturbed solution is worse than the current solution, then the perturbed solution might still
replace the current solution with a probability
based on the distance between the current and
perturbed solution values and the current
temperature parameter used in the annealing process. As the method progresses, the
temperature parameter is lowered, making it
more and more unlikely to replace the current
solution with a worse perturbed solution.
For the hybrid GRASP-SA method, GRASP
is used to determine the (near) optimal task
assignment for the UAV and the SA method is
used to determine the (near) optimal route that
the UAV should fly to accomplish its tasks. The
best solution over all of the GRASP multistart
iterations is retained as the final solution. A
complete solution, for a given UAV, consists
of the tasking and flight path for that UAV, as
well as the tasking and flight path of that UAV’s
neighbors. The flight path trajectories are
calculated over a selected planning time horizon.
The hybrid GRASP-SA method as applied
to the UAV surveillance and track planning
problem is based on the assumption that only
one task can be performed by a vehicle over
the entire planning horizon. In reality, this is
not the case. There are many instances when a
UAV is assigned a task but is unable to perform
that task during certain sub-intervals of the
planning horizon because, for instance, the
target to be tracked is not in the field-of-regard
of the UAV’s sensor. To account for these taskfree sub-intervals, an additional method, the
collection opportunity update (COU) method,
was implemented that evaluates the GRASP-SA
solution and determines where in each UAVs
trajectory additional tasks can be performed,
thus improving the quality of the solution.
Simulation Results
A simulation of UAVs performing a surveillance and tracking mission over an urban area
was developed to test the decentralized planning and control algorithms. The GRASP-SACOU optimization algorithms were coded in
C++ and inserted into a simulation environment that provided target and UAV kinematics, a scenario laydown, a display, and data
1 An NP-hard problem is a class of computational problems that are at least as hard to solve as the hardest problem that can be
solved in polynominal time.
logging and analysis routines. For comparison
to the baseline UAV planning approach (a
decentralized, cooperative approach), two
other solutions were simulated; a centralized
planning approach and a decentralized selfish
approach. The centralized approach collects all
UAV information at a central location and then
generates all UAV tasks and flight paths using
this information. Clearly, this approach is the
most communication bandwidth intensive of
all the approaches, but its access to all UAV
sensor and tasking information make it the solution with the best task planning performance.
The selfish approach allows individual UAVs
to plan their own actions as in the baseline
approach, but information from neighboring
UAVs is not shared and the UAVs develop
independent, uncoordinated tasking and flight
plans. Two new metrics were developed to
measure performance. The cost of decentralization is a measure of the solution quality of the
decentralized approach versus a centralized approach. The price of anarchy metric compares
the solution quality between the selfish and the
cooperative approach.
In Figure 2, we show the difference in solution
quality before and after the COU algorithm.
The figure represents a top-down view of
a simulated urban environment. The gray
rectangular objects represent the tops of buildings. The winged objects that have numbers
near them and a curve extending from them
represent the UAVs. The curves represent the
projected flight path over the planning horizon;
red segments indicate that the UAV is performing a searching task, a green segment indicates
the UAV is performing a tracking task. The
black outlined square, extending from each
UAV, represents its sensor footprint. The motorcycles and cars represent the targets, and the
light and dark blue colored background represents the composite probability of successful
search regions. Light blue indicates a greater
than 90 percent probability of that region being
successfully searched, and dark blue indicates a
greater than 95 percent probability.
The GRASP-SA solution, shown on the left
of Figure 2, shows the UAV route plans at
one particular time step without the COU
algorithm applied. UAVs 2 and 4 are tasked
to search for targets, while UAVs 1 and 3 are
tasked to track a previously detected target
that is moving downward. It can be seen that
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
33
FEATURE
Optimization Algorithms for Decentralized Planning and Control
GRASP-SA
GRASP-SA-COU
4
2
4
1
3
2
1
3
Legend
UAV Computed Surveillance Route
UAV Computed Track Route
Identified Objects of Interest
Unidentified Objects of Interest
Current UAV Field of Regard
Current >90% Probability of Successful Search Region
Current >95% Probability of Successful Search Region
Figure 2. UAV sensor utilization is improved using the GRASP-SA-COU approach (right) versus the GRASP-SA approach (left). The GRASPSA approach switches UAV 1 and 3 trajectory tasking from track to surveillance based on an estimate of when the target being tracked will no
longer be in the UAV’s field of regard.
the UAV 1 and 3 trajectories will not keep the
target in their fields of regard and therefore will
not be able to maintain a track over their entire
trajectory planning horizons.
for those UAVs having the same tasking. The
closer the flight paths are to each other (from
the centralized and decentralized approaches),
the lower the cost of decentralization.
The trajectories for the UAVs, with the COU
algorithm applied, are shown on the right in
Figure 2. None of the UAV trajectories change.
However, UAVs 1 and 3 have modified their
tasking so that once the extrapolated target
position is no longer in the field of regard of
their sensors, the tasking is changed to search,
resulting in better usage of UAV resources.
Figure 3 illustrates the comparison between
solutions derived from the decentralized
cooperative approach (left) and centralized
approach (right), at a given time. The decentralized solution determines that all four
UAVs should be in search mode over the
complete planning horizon. By contrast, the
centralized solution determines that three of
the four UAVs should execute only tracking
tasks and the fourth UAV should only perform
surveillance at the end of its trajectory. Hence,
the resulting cost of decentralization metric at
this time is high.
The cost of decentralization metric compares
the solution produced using a centralized approach with the solution produced using a decentralized approach. The measure computes
the normalized distance between trajectories
34 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
The cost
of decentralization
metric compares
the solution produced
using a centralized approach
with the solution produced
using a decentralized
approach.
FEATURE
The price of anarchy compares the solution
produced from a cooperative approach with
a selfish approach. The measure computes the
normalized distance between trajectories for
those UAVs having the same tasking. This is
one way to measure the benefit of cooperation, i.e., the benefit of sharing information
on detected targets and search uncertainty.
Figure 4 shows simulation results comparing
the decentralized cooperative approach solutions (left) and the selfish approach solutions
(right). There are significant differences in
both the UAV trajectories and tasking and
therefore the price of anarchy metric is large.
On average, the selfish approach produces
an 8 percent worse sensor resource utilization (based on a specific objective function)
than the decentralized cooperative approach,
primarily due to the lack of information
exchange between UAVs when using the
selfish approach.
Summary
The DoD is investigating autonomy as a
means of lowering the manpower required
to conduct missions. Decentralized cooperative control methods, such as the approach
outlined in this article, provide starting points
for further autonomy research and testing,
eventually allowing the DoD to increase the
amount of mission planning and execution
done by machines and freeing operators for
higher-level supervisory control to ensure
broad mission objectives are being met.
Decentralized Cooperative Approach
Centralized Approach
3
4
2
2
3
1
1
4
Legend
UAV Computed Surveillance Route
UAV Computed Track Route
Identified Objects of Interest
Unidentified Objects of Interest
Current UAV Field of Regard
Current >90% Probability of Successful Search Region
Current >95% Probability of Successful Search Region
Figure 3. A comparison between the decentralized cooperative approach and the centralized approach shows significant differences in solutions, resulting in a large cost of decentralization metric. On average, the centralized approach yields a 19 percent improvement
in sensor resource utilization as compared with the decentralized framework.
Decentralized Cooperative Approach
Selfish Approach
1
1
Rather than a dramatic shift, it is expected
that the implementation of autonomy will
follow a gradual path, starting out as intelligent decision aides and performing lower-level
mission execution functions, all with operators tightly controlling all critical decisions.
As the DoD’s experience and confidence using
autonomous methods increases, it is expected
that additional control will be performed by
machines, allowing humans to allocate more
time to defining top-level mission objectives
and supervising mission execution. •
Michael J. Hirsch, Ph.D., and Daniel Schroeder
2
2
4
4
3
3
Legend
UAV Computed Surveillance Route
UAV Computed Track Route
Identified Objects of Interest
Unidentified Objects of Interest
Current UAV Field of Regard
Current >90% Probability of Successful Search Region
Current >95% Probability of Successful Search Region
Figure 4. A comparison between the decentralized cooperative approach and the selfish approach shows there are some differences in solutions, resulting in a large price of anarchy
metric. On average, the selfish approach produces a sensor resource utilization 8 percent
worse than the decentralized cooperative approach.
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
35
FEATURE
Partially Observable
DECISION PROCESSES
Decisions associated with a known or observable condition are based on facts from direct observation.
For example, a normal pedestrian decides to observe the stoplight before proceeding to cross the
intersection. For a blind pedestrian, however, the stoplight is only partially observable, so the decision is
to listen for traffic and other pedestrians before proceeding. Listening infers the stoplight state based on
indirect but related observables. Inference in a partially observable process is known as belief. Belief is
the probability of being in a state and serves as a surrogate for fact when indirect but related observables
must be used to infer state. More often than not, whether in sports, business, medicine, warfare, etc.,
real-world decisions are based on belief rather than fact. This article describes a belief-based method
that Raytheon uses to optimize decision making in solutions we provide to customers.
Partially Observable Markov Decision
Processes (POMDPs)
ecision making can be defined as a
sequence of actions taken over time for
the purpose of producing a desired goal.
The important information of a sequence’s past
history is often captured in the present state of
the system (e.g., the confidence associated with
a prioritized intelligence or high-payoff revenue tip), and in these cases only the current
state is needed to determine the best action
for transitioning to a future desired state. The
formal model used to represent such a system
is a step-wise stochastic sequence known as a
Markov decision process (MDP).
D
10
CEE
D
9
PRO
8
7
VALUE
6
5
4
3
A completely observable MDP (COMDP) assumes that the state of a system (e.g., the color
of the stoplight) at any time is known to the
decision maker. Assigning reward (cost) for
the known state attained from a decision immediately determines the best decision as the
one with the highest reward (least cost). In a
partially observable MDP (POMDP), the decision maker has only belief based on indirect
observables (e.g., traffic and pedestrian sounds)
with which to infer the state. Since belief is
the probability of being in a state it can be any
real number in the unit interval. So, while a
COMDP typically has a finite or countable
36 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
2
1
1
0.9
0
0.8
LISTEN
0.05
0.7
0.15
0.6
0.25
0.5
0.35
0.4
0.45
RED Belief
0.3
0.55
0.2
0.65
0.75
GREEN Belief
0.1
0.85
0
0.95
Figure 1. LISTEN is the selected action (plane) for belief states valued  0. PROCEED, indicated by separately colored bands of equal value, is valued non-negative for belief state
pairs (GREEN  0.4, RED ~ 0). High value to PROCEED is only attained at near certainty
of GREEN which is obtained from listening to traffic flow.
FEATURE
number of states, a POMDP has a continuously
infinite number of possible belief states. Belief
value is introduced to overcome the difficulty
of belief cardinality in solving a POMDP. As in
a COMDP, a reward is assigned for the value of
belief state certainty (unit interval end points)
attained from a decision. Given N states, the Nvector of state rewards for each decision defines
a hyper-plane in N-dimensional “value” space
that maps underlying belief into belief value.
Solving a POMDP amounts to stepwise value
iteration, successively estimating the maximum
probable belief from observables at each step.
After a finite number of steps, the process
results in optimal decision hyper-planes that
intersect to form a piece-wise linear convex
surface known as the POMDP policy. For any
belief, the POMDP policy dictates the best
decision to attain a desired state.
The POMDP model of the blind pedestrian’s
optimal decision selection includes three stoplight states (GREEN [G], YELLOW [Y], RED
[R]), two decisions (LISTEN, PROCEED),
rewards (near 1 [0] for a good [bad] decision in
a state) and observables (heavy, medium, or no
traffic flow given states G, Y or R, respectively).
The model’s transition functions consist of
conditional probabilities associated with
transitioning from one state to another given
a decision. The model’s observation functions consist of conditional probabilities of an
observable (sound) occurring having attained
a state from a decision. Figure 1 illustrates the
optimal decision policy that results from solving the POMDP. Since belief is a distribution
of probability over three states, only two need
be considered; in this case R and G, i.e., Y = 1
 (G+R). The decisions, listen and proceed, are
represented by two intersecting planes in three
dimensional value space that map the belief
state ordered pairs (R, G), 0  R + G  1, into
belief values between 0 and 10. To visualize
the blind pedestrian’s POMDP policy space, it
is sufficient to consider (R, G) belief ordered
1
Jet
Bridge
Entrance
Jet
Bridge
Entrance
Camera
Camera
Greatest
Risk
Pedestrian
Other
Destination
Entrance
Other
Destination
Entrance
Pedestrian
Figure 2. Simulation snapshot of two active jet bridges and two independent cameras
observing approximately 150 passengers.
pairs whose component sums are less than or
equal to 1. As evident in the diagram, only for
belief states with R near 0 and G well above
0.5 does the policy highly value a decision to
PROCEED.
Autonomous Threat Assessment and
Course of Action Applications
Protecting jet bridges1 from a suspected terrorist is an unclassified, International Traffic and Arms Regulations (ITAR) compliant
Anti-Terrorism Advisory Council (ATAC)
scenario illustrating the design and operation
of a POMDP that provides optimal course of
action (CoA). The mission is to provide airport
waiting area security to prevent unauthorized
A jet bridge is an extendable bridge for loading passengers onto large commercial aircraft.
access to one or more jet bridges. The waiting
areas are viewed by one or more cameras with
narrow fields of view (FOV) that cover only
a portion (approximately 20 percent) of the
entire area, and this FOV is POMDP controllable (i.e., the application can decide where the
cameras point). In addition to pointing the
cameras, the POMDP has two additional decision options:
• Interdict a particular passenger before he
or she gains access to the jet bridge.
• Close the waiting area door when interdiction is not possible (threat is too close or
threat assessment decision was made too
late).
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
37
FEATURE
Partially Observable Decision Processes
TRUTH
2
INITIALIZE
EEI C Declared “YES”
1
N
Sensor Inputs Improve
Belief for EEI C
1.00
Y
Probability
Query B
0.50
N
Query C
1.75
0.50
0.25
N
0.00
0.96
A
B
D
E
F
G
H
I
Query
Y
Query D
Y
5
Query E
Contribution
1.00
0.50
Contribution
N
N
1.75
3
0.50
0.25
0.00
0.50
A
B
C
D
Y
EEI D Highlighted
Query
Query F
Query G
Action
0.50
Action Value
N
E
Tree Evaluation Identifies EEI
D as the Highest Contributor
Y
0.50
N
20
15
10
5
0
4
A
B
C
D
E
Query
Y
POMDP Indicates EEI D
now has Highest Benefit
The specifications are in the form of intelligence requests (IRs) which may be prioritized,
i.e., prioritized IRs (PIRs), and can be viewed
as questions to be answered. PIRs are decomposed into essential elements of information
(EEIs), some or all of which are needed to
answer (or satisfy) the PIR. The EEIs — when
decomposed to their lowest level — represent
individual pieces of information that can be
provided by a sensor. Often, these EEIs can
be depicted as a decision tree, where each
individual EEI belongs to a decision hierarchy
designed to answer the original PIR.
Action
Query D
Query I
N
0.50
True - False
True
0.55
0.45
1.00
1.75
Probability
False
0.50
0.25
0.00
True
State
False
Figure 3. Hierarchical POMDP determines at each step which questions, if any, can be
declared and which make the largest contribution and benefit to the decision process.
The process is continued until all questions are declared or the state is deemed acceptable.
The POMDP uses the camera observations
to generate a belief that each passenger is in a
particular state, e.g., a terrorist. The belief is
a probability distribution across all possible
states. Based on the belief distribution, the
POMDP decides on a CoA: either specify what
parts of the waiting area to view next or take
direct action against a passenger.
The jet bridge mission has been simulated (Figure 2) in a stressing scenario involving approximately 150 pedestrians (represented by aqua
dots) moving in the security area (background
grid) and two independently controllable
cameras (white dots) protecting two jet bridge
entrances (red dots with range rings). Each
pedestrian has a destination (one of the six red
dots on the screen) out of the security area and
moves toward that destination with random
speed but with some wandering (a random
variation on the direct path). One of these
pedestrians is a terrorist. At any given time, the
POMDP identifies (red box) which pedestrian
is the greatest risk, although this risk may not
be large enough to warrant further action.
Testing shows that the POMDP does identify
the terrorist every time; in about 85 percent of
the cases this identification occurs in time for
the interdiction to be performed (the lowercost option). The POMDP will autonomously
decide to collect additional observations until
38 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
the belief is sufficient to trigger a CoA.
POMDPs can also be applied to control
multifunction sensors where several mission tasks will compete for sensor resources.
Typically, this arises in sensor systems that
can perform both surveillance and fire control
functions, such as the F-15 radar and the
predator multispectral targeting system (MTS)
electro-optical system. In the course of a mission, these systems and their operators will
be faced with decisions that require balancing
the need to perform surveillance across a large
area to maintain situational awareness and the
competing needs inherent in a particular CoA
such as dwelling on a particular detected entity
in order to assess threat suspicion, tracking
entities that are declared suspicious and obtaining additional targeting data to support other
systems, such as weapon systems, that are used
to counter a declared threat.
Intelligence, Surveillance and
Reconnaissance Applications
The POMDP technique can also be applied
to pure intelligence, surveillance and reconnaissance (ISR) sensor missions, i.e., missions
having no follow-on fire control or weapon
phases. The ISR process begins with a top-level
commander specifying what information is
needed to make ongoing command decisions.
A notional PIR decomposed into its EEI decision tree is depicted on the left side of Figure 3.
This particular tree is a fictitious example but is
typical of an actual PIR and its associated EEIs
where the consequence of a poor assessment
is extremely serious but refining (improving
confidence in) the assessment is potentially
very costly. POMDPs can be constructed for
each statistically independent EEI to help
decide whether to declare the EEI question true
or false or to gather more information, i.e.,
task an appropriate sensor that can provide additional information about this EEI. The belief
states shown on the left side of Figure 3 capture
the current confidence that the EEI question is
true or false. With each new piece of information collected (which can be either positive or
negative), the POMDP transitions the belief
state from its current value to a new value that
reflects the new information gained from the
sensor collection. These individual POMDPs
are rolled up into the overall PIR confidence
level shown on the bottom right of Figure 3
based on their decision tree hierarchy.
Autonomous Dynamic Tasking
Applications
Autonomous dynamic tasking (ADT) is a class
of techniques and methods that optimize, and
continually adjust, the execution of a set of
tasks or activities based on the time-dependent
relative importance of tasks. ADT has applications in the Department of Defense (DoD) and
Intelligence Community (IC) for executing
missions, as well as in commercial industries.
An example of the latter is a commercial satellite imagery provider’s transition from quality
of service (QoS) to revenue-based scheduling
offered by POMDP value optimization. Paying
customers select the desired image quality from
a menu of cost-ranked options, and the highest
FEATURE
44.20
Do ADT
44.10
Value
44.00
43.90
PPT
ADT
3.80
Do ADT
43.70
43.60
0.0
0.2
0.4
0.6
0.8
1.0
Belief
Figure 4. Two-state policy for executing a dynamic task in lieu
of pre-planned tasking based on the belief in a high-payoff tip.
paying customer receives the highest QoS through satellite spatial and
temporal performance scheduling.
However, QoS may be deliberately compromised if the provider acts on
speculative high-payoff imaging opportunities (e.g., a sporting, political
or major weather event such as a hurricane). Given a speculative tip, the
ADT POMDP computes a policy for either adjusting pre-planned tasking
(PPT) to accommodate the potential high-payoff event or not; the goal
being to optimize revenue based on belief that the speculative event will
occur and that coverage of the event will result in a substantial payoff.
Figure 4 is an example of a policy consisting of two states, revenue gain
and loss, and two decisions, ADT or PPT, depending on the confidence
(belief) associated with a speculative tip. The POMDP decisions are
represented by lines (one-dimensional action hyper-planes) intersecting
in a two-dimensional value space. Value is the iterated expectation of the
reward based on task priority ranking. The POMDP policy selects the
decision associated with the line of maximum value for any belief. The
ADT line has maximal value for tip confidence above approximately 0.55.
Thus, only a relatively high confidence tip warrants sufficient revenue gain
to override pre-planned tasking. Nonetheless, the POMDP policy suggests
a strategic shift to a business model that integrates consumer-based QoS
and provider-based revenue optimization.
Future POMDP Application Areas
The application of POMDPs to support critical decision making and
actions is ubiquitous; ranging from autonomous threat detection and
course of action missions to ISR mission planning to revenue-based QoS
customer resourcing. New and promising applications of POMDPs to
game-theoretic strategy formulation within Raytheon have been developed and successfully implemented, and new application areas continue
to be researched. •
Kenneth Moore, Ph.D., and Robert Vitali
E N G I N E E R I N G P R O F I L E FEATURE
Kenneth
Moore, Ph.D.
Director,
Algorithms
and Intelligent
Systems
Dr. Kenneth L.
Moore has more
than 33 years of
experience in aerospace and defense
industry business
and technology leadership, program
management and
engineering. He has
been the technology
area director for
algorithms and intelligent systems at the Raytheon Space and Airborne Systems
business since 2007. As both director and subject matter
expert, he contributes to solutions and oversees independent
research and development (IRAD) investment definition and
execution associated with mission and system analysis, modeling and simulation, concept of operations development and
intelligent real-time algorithm development for airborne and
space-based intelligence, surveillance and reconnaissance
(ISR), cognitive electronic warfare (EW), on-board multi-intelligence (multi-INT) fusion and autonomous decision and control.
“Success at Raytheon requires a balance between customer
exposure, business pursuit and technical excellence,” Dr.
Moore states. “The requirements of Department of Defense
(DoD) and intelligence community (IC) customers are constantly changing as are their financial priorities. Leadership
and technical excellence is recognized and rewarded when
associated with high-priority customer problems.”
Dr. Moore’s involvement with customer challenges over the
past 15 years began with space-based missile defense and surveillance and has evolved to tactical airborne automatic target
detection, identification and tracking, autonomous payload
processing decision and control solutions associated with
actionable information, multi-INT resource management and
adaptive/cognitive strategy. Initially focused on counter terrorism and insurgency, he has transitioned his mission focus to
counter anti-access/area-denial (C-A2/AD), stand-in EW strike,
space-based ISR and space protection and situational
awareness.
He talks about, technologically, what a great time in history it
is to be working on research and innovation: “Modern computing and networking have advanced to a point where
decades of computationally overwhelming optimal control and
decision solutions can now be reduced to practice with common families of algorithms tailored to applications that
advance human reasoning.”
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
39
LEADERS CORNER
The Raytheon Technology Leadership Council
William F. Kiczuk
(Chair)
Steve Cummings,
Ph.D.
Chief Technology Officer (CTO)
and Vice President,
Corporate Technology
Vice President, Technology
Development and Execution,
Space and Airborne Systems
John Zolper, Ph.D.
(Co-chair)
Mark Hebeisen
Vice President,
Research and Innovation
Technology Today spoke with Technology
Leadership Council (TLC) members about
how Raytheon research and technology is
managed and the roles and responsibilities
of its technology leaders.
How does the TLC operate?
Bill K.: Raytheon provides leading-edge,
innovative technologies to our customers
to ensure mission success, and the TLC manages the strategies that ensure the correct
technologies are being developed and matured. As TLC Chair, I work with technology leaders across Raytheon to develop the
company’s technical roadmaps and oversee
the execution of the company’s research and
technology initiatives. The TLC prioritizes
Raytheon’s technologies based on our customers’ needs and our business plans and is supported by technology area directors (TADs)
and champions. TADs and champions engage
with the Raytheon technology network participants to assess our technology portfolio and
identify technology trends, disruptions and
innovations that will continue to differentiate
Raytheon and our products.
The company’s technical roadmaps integrate
hundreds of different activities spread across
numerous technology focus areas and include
contract research and development activities
(CRAD), independent research and development activities (IRAD) and university and
industry partnerships. The roadmaps are
reviewed and updated every year by the TLC,
TADs, and technology champions to ensure
the technology maturation strategies depicted
in the roadmap are executable and provide the
required technologies at the required times to
support the mission needs of our customers.
40 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
Strategic Architecture
Director and
Technical Director,
Integrated Defense
Systems
What excites you about technology at
Raytheon?
Bill K.: I’ve been involved in Raytheon technologies for more than 30 years in different
businesses and at corporate in my current role
as CTO. I am continually impressed with the
breadth of technologies being investigated and
developed. Raytheon’s technology portfolio
is always changing, with new areas emerging
every year that have the potential of providing major improvements to our existing
products or of being disruptive and providing a path to a whole new way of performing
a mission. Whether the technology is a new
radio frequency (RF) semiconductor device,
a high power laser, or an advanced analytics
algorithm, Raytheon is constantly investigating new technology areas and how these new
technologies can be advantageously applied
to our products and customer missions. I find
the constant learning required to keep pace
with these new and emerging technologies
very exciting. Also, my frequent interaction
with our talented and dedicated researchers
and technologists is very rewarding. There is
an excitement and energy you get from talking
with these researchers and learning about their
technology.
What is the role of a business technical
director (TD) and what are your day-today responsibilities as a TD?
Mark H.: Within each business, the TD is
responsible for the over-arching technology
strategy for their business and is responsible
for driving innovation, technical excellence,
and technology development and maturation.
The TD’s responsibilities within the business
include IRAD planning and execution, coordination of CRAD capture activities, fostering of
university and small company research efforts,
and supporting our intellectual property (IP)
Marty Rupp
Technical Director,
Missile Systems
Brad Whittington
Capabilities and
Technology Director,
Intelligence, Information
and Services
and legal teams to ensure our IP is properly
developed and documented. Each TD is also
a member of the TLC that is responsible for
the leadership and coordination of technology efforts across the company. The TD is the
primary point of contact within the businesses
for supporting cross-Raytheon technology
integration efforts.
In addition to these responsibilities, each TD
may have other business-specific responsibilities. For example, at Integrated Defense
Systems (IDS), I also serve as the director of
the strategic architecture engineering group.
In this capacity, I lead a team of product-line
chief engineers, technical directors and industry-recognized subject matter experts, many of
whom are top engineering fellows. Hence, at
IDS, the TD not only champions the technical strategy for the business, but also ensures
flawless execution of the engineering content
on new development and ongoing production
programs. The TDs at the other businesses
also have unique business-specific roles and
responsibilities per the preferences of their
leadership teams.
How do the business TDs collaborate?
Steve C.: The business TDs interact in a
number of ways, both formal and informal.
First, corporate events such as Technology
Integration Week (TIW), monthly TD meetings chaired by the CTO, and technology
focus area reviews provide formally scheduled
opportunities for the TDs to collaborate,
especially on companywide efforts and initiatives. Less formally, the TDs cooperate to
support each other’s business needs through
the formation and staffing of proposal and
architecture review teams and other sharing of technical and programmatic expertise.
Quite informally, the TDs correspond
throughout the year via email and phone
The Technology Leadership Council oversees Raytheon research, collaboration
and technology opportunities and is responsible for developing and executing an
integrated technology and research strategy for the company. The TLC is chaired
by the Raytheon Chief Technology Officer with the technical directors from
each Raytheon business as members.
calls on opportunities such as innovation
ideas, the sharing of IRAD results, and discussions on coordinating IRAD projects between
businesses.
How do you find and nurture innovation?
Brad W.: As a world-class aerospace defense
company, innovation not only invigorates and
excites our workforce, it drives our business
growth — especially in a budget constrained
environment that seeks the typically affordable gains of innovation versus the generically
expensive gains of invention. Thus, Raytheon
spends significant resources at both the corporate and business level to drive and leverage
innovative employee ideas. At the corporate
level, the Identify, Develop, Expose, Action
(IDEA) program and the Raytheon Innovation
Challenge (RIC) encourage employees to submit innovative ideas that can lead to funding,
customer interaction and leadership opportunities. Other corporate activities such as TIW
and the Fellows workshop drive real-time innovation by assembling Raytheon’s technology
leaders at a single location for presentations
and working groups on technology planning and assessments as well as innovation.
Each business also has its own methods and
programs for identifying and nurturing innovation; all focused on setting up an efficient
process for our technologists and engineers to
get their ideas out, obtain technical feedback,
and, potentially, obtain funding for further
investigation.
The corporate and business-specific initiatives
are important and effective ways to identify and
nurture innovation. However, the largest innovation creating process at Raytheon continues
to come from the engineer’s interaction with
their peers and mentors. Activities as simple
as buying a younger employee lunch, asking a
mentor a question at the water cooler or raising a hand during an engineering review can
lead to innovations that positively impact our
business.
Looking for innovation is a continual process and is not limited to a specific Raytheon
business or to Raytheon as a whole. Rather,
innovation can come from any Raytheon
business or from outside Raytheon; reapplying an innovation is again innovative.
We should all continue our innovative thinking
and look all around us for how to apply other
innovations to our businesses.
How does Raytheon decide when to
invest in internal research versus obtaining the technology externally, and what
role do universities play in the technology
strategy?
Marty R.: Generally, we invest internally in
technology that can provide a major competitive discriminator, and we rely on externally
available technology (e.g., commercial off
the shelf [COTS], etc.) for commodity items.
Examples of competitive discriminators are
advanced electro-optic/infrared and radio RF
sensors, analog and digital electronics, and
algorithms. While our electronics use many
COTS components, the architecture, packaging design, firmware and software provide the
discriminator. Both development and recurring
costs are also significant factors, and this constraint also drives the decision.
We are starting to place more emphasis on
universities as a technology source, mainly
with regard to low technology readiness level
(TRL) technology that needs to be matured
before we seriously consider its use. Across
Raytheon, we are involved with and participate in multiple university consortia, including the Defense Advanced Research Projects
Agency (DARPA) Semiconductor Technology
Advanced Research Network (STARnet), the
Arizona State University (ASU) Sensor Signal
and Information Processing (SenSIP) Center,
the National Science Foundation supported
Center for Surveillance Research (CSR), and
others. Membership in these groups provides
Raytheon the opportunity to obtain and leverage collaborative research results at reasonable
membership costs. We also work with individual universities, providing Raytheon opportunities to develop a campus presence, support
recruiting and collaborate with professors who
are performing research in areas with potential application to our products and customer
needs.
How does a person get involved in
Raytheon research, technology and
innovation?
John Z.: Research, technology and innovation are core to Raytheon’s business and culture. There are multiple ways to get involved,
whether at the front end in generating new
ideas or after a contract award in executing
successful programs. At the corporate level,
we foster early-stage technology identification,
maturation, and assessment through our IDEA,
RIC and core research programs. Each business also sponsors innovation activities and has
processes to identify and support IRAD projects that enable new capabilities. Raytheon also
works with our customers to develop technology through CRAD contracts and as part
of program upgrades.
A good mechanism for engineers and
technologists to learn more about Raytheon
technology activities is to participate in a Raytheon Technology Interest Group (TIG). TIGs
are community-of-interest groups that enable
Raytheon engineers and technologists with
similar interests to exchange information via
meetings, workshops and other activities. The
TIGs cover a broad range of technology areas
and are open to anyone who wants to attend. •
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
41
on
Technology
Raytheon Enhances Its Sensor Technology Portfolio With
the Acquisition of Poseidon Scientific Instruments
Introduction
Raytheon is committed to the success of our
global radar customers through our technological leadership in radio frequency (RF) sensor
systems. Our ability to provide next generation
capabilities has increased with the acquisition
of Australian technology company Poseidon
Scientific Instruments (PSI), developers of the
world’s highest-performance microwave signal
generators. PSI expands Raytheon’s brand in
world-class sensor systems by providing expertise in ultra-low-phase-noise signal generation
and companion measurement technologies.
These technologies provide new levels of system
performance and mission capabilities for the
tor technology that is critical to establishing
and sustaining oscillation.
the mid-1990s of new threats drove even more
demanding radar sensitivity requirements that
pushed the limits of SAW-based exciters.
High-performance, low-phase-noise microwave MO technology evolved over the decades
from klystron oscillators in the 1950s, to
frequency-multiplied quartz crystal oscillators,
to frequency-multiplied surface acoustic
wave (SAW) oscillators in the 1990s. This
technology evolution has enabled radar
systems to achieve higher levels of sensitivity.
PSI Technologies
Raytheon and the warfighter needed a leapahead technology to achieve the increased radar
sensitivity necessary to combat new emerging threats. Enter PSI and their ultra-highperformance sapphire resonator technology.
Founded in 1987, PSI is recognized worldwide
by international and U.S. defense and commercial industries as a leader in generating and
subsequently analyzing ultra-low-phase-noise
microwave signals for high sensitivity radar and
communications systems. By the early 1990s,
PSI’s compact sapphire resonator was a reality.
It employed a single synthetic sapphire resonator that offered orders of magnitude reduction
in microwave oscillator noise over our SAWbased solutions.
From the mid-1990s to 2000, Raytheon and PSI
engineers independently developed and demonstrated a sapphire-based MO, built around
PSI’s patented sapphire resonator technology.
Raytheon’s proof-of-concept MO and related
exciter architecture were used to demonstrate
a new level of radar performance. At the same
time, PSI had fully productized their compact
sapphire-based oscillator into a shoebox size
Figure 1. High performance sapphire resonator oscillator solutions: the Sapphire Loaded
Cavity Oscillator (SLCO) (left) and Shoebox Oscillator® (SBO) (right) — Setting the bar
for superior sensor sensitivity.
Radar Oscillator Background
Radar systems consist of several major subsystems, including a receiver-exciter, transmitter, signal data processor and power system,
each of which performs a specific role in radar
operation. For the receiver-exciter subsystem,
a master oscillator (MO) is used to provide
the phase reference for transmit, receive and
radar timing functions. It is the spectral purity
of the MO, measured as power spectral density
and conventionally expressed as amplitude
modulation or phase modulation noise, that, in
part, establishes the receiver-exciter noise and
in turn, the radar’s detection sensitivity. In the
presence of large clutter, MO noise can mask
small targets and thus limit a radar’s detection
sensitivity. Moreover, at the heart of every MO
is a narrow-bandwidth, high-stability, resona-
42 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
Furthermore, Raytheon’s novel SAW resonator and oscillator technologies have delivered
outstanding radar system performance for
several decades. However, the emergence in
Phase Noise (dBc/Hz)
U.S. warfighter and our international coalition
partners.
- 80
- 90
- 100
- 110
- 120
- 130
- 140
- 150
- 160
- 170
- 180
0.1
1
10
100
1,000
10,0000
Offset Frequency (kHz)
400 MHz SAW Oscillator at 10.24 GHz
10.24 GHz Sapphire Oscillator
Figure 2. Sapphire technology performance comparison: -20 dB better phase noise
performance opens new doors for radar performance.
Multifunction Radio Frequency Systems
integration of these technologies with Raytheon
technologies and architectures opens doors to
new commercial and defense opportunities, in
particular in the area of communications where
ultra-high stability (i.e., low phase noise/jitter)
clock oscillators are essential. The importance
of precision timing continues to increase to
support higher speed communications protocols and to support an ever expanding wireless
environment that continues to be constrained
by fixed operational bandwidth allocations.
Figure 3. The Oscillator Development INstrument (ODIN®) phase noise analyzer —
Designing and delivering state-of-the-art sources requires a commensurate state-of-the-art
noise measurement capability.
form factor suitable for radar applications.
This new sapphire-based exciter forms the
foundation of many of our current, highperformance radar systems and continues to
provide Raytheon radar systems with industry
leading levels of performance.
Searls who, early on, saw the potential and
value in transitioning the cryogenic sapphire
resonator technology from his gravity wave
work at the University of Western Australia
(UWA) in Perth, into a product for commercial
and defense applications.
Raytheon Australia’s PSI Sapphire Loaded
Cavity Oscillator (SLCO) and Shoebox Oscillator® (SBO) are shown in Figure 1. The SBO
design represents a groundbreaking productized capability for use in tactical defense
systems. These microwave oscillators are used
today in premier Army, Navy and Air Force
sensor systems and are used as laboratory standards at the National Institute of Standards and
Technology (NIST) and the U.S. Naval Surface
Warfare Center, Crane Division. Figure 2 shows
the superior performance of the Raytheon
Australia sapphire oscillator.
During the development of its sapphire
oscillators, PSI needed measurement tools
that were not limited by conventional
semiconductor performance. To meet this
need, PSI developed ODIN® (Oscillator Development INstrument) – a sophisticated phase
noise analyzer (Figure 3). ODIN was developed
out of necessity to support measurement of
their low-noise sapphire sources, but it evolved
into a standalone, commercially competitive
product line.
For less demanding applications, PSI’s dielectric
resonator oscillator (DRO) product line provides lower-cost solutions. PSI also has a family
of low noise, regenerative frequency dividers
that can be used to derive additional, lowerfrequency signals.
PSI’s sapphire source and noise degeneration
technology has its roots in the field of gravity wave detection, which required a unique,
high-performance measurement system offering many dual use and technology transition
opportunities into radar and sensor programs.
However, it is the innovation and vision of
PSI’s founder and managing director Jesse
Raytheon-PSI Today
PSI technologies are currently being integrated into Raytheon products in the U.S. The
In addition to PSI’s resonator technology, their
RF engineering design and test skills provide
Raytheon Australia with enhanced engineering
expertise. Recently, PSI delivered 16 receiver
systems for the Murchison Widefield Array
(MWA) telescope system. Figure 4 shows portions of this system. The receiver components
were designed at different universities and other
research organizations worldwide as part of an
international scientific consortium that sponsors the MWA telescope project. PSI designed
the receiver node enclosure, integrated all of the
components into their enclosure, tested each
node and delivered them to the MWA team to
support system integration and testing.
During the PSI and Raytheon integration
process, opportunities to grow and expand
PSI’s footprint in commercial and international
markets for low-noise products will continue to
be explored as well as opportunities to enhance
current and future Raytheon products with PSI
technologies. •
Robert E. Desrochers II, Mark Koehnke
and Jesse Searls
Images courtesy of the Murchison Widefield Array via Curtin University
Figure 4. Murchison Widefield Array equipment: 4 x 4 antenna tile (left) and Raytheon
Australia PSI integrated receiver node (right). Each receiver node supports eight tiles for a total
of 128 dual-linear-polarization antenna elements.
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
43
SPECIAL INTEREST
Raytheon’s
Collaboration
With
Worcester
Polytechnic
Institute
Raytheon recruits engineers
from more than 70 universities
around the world. In addition to
recruitment, Raytheon collaborates
with universities in engineering
research, training and sponsorship of
science, technology, engineering and
mathematics (STEM) activities. This
article highlights the collaboration
between Raytheon and Worcester
Polytechnic Institute (WPI) in
these areas.
STEM Sponsored Initiatives
Raytheon and WPI are committed to STEM
education and in fostering and maintaining
excitement among students in STEM disciplines. WPI’s STEM Education Center provides
training for K-12 educators and administrators,
and Raytheon supports several of WPI’s STEMrelated programs, such as RoboNautica, a
major FIRST Robotics event that showcases the
talented young minds of tomorrow’s engineers
and scientists, Camp Reach, a summer camp for
seventh grade girls that introduces the principles
of engineering via hands-on experiences, and
TouchTomorrow, a free interactive festival with
exhibits and activities focused on the future of
science and technology.
These programs support Raytheon’s objective of encouraging students’ enthusiasm
44 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
for engineering and practical applications of
theoretical concepts, ultimately, to enable them
to become future technology leaders.
Raytheon is proud to support these great
events and WPI’s commitment to educating
our future scientists and engineers, as well
as the educators and administrators in our
primary and secondary schools.
Program Protection Planning
Program Protection Planning (PPP) is a
disciplined process that helps ensure adequate protection of a program’s technology,
components, and information from unauthorized collection, exploitation or use. WPI has
been designated as a center of excellence in
cybersecurity research by the National Security
Agency and the Department of Homeland
Security, leading the way in PPP training by
launching the first graduate program aligned
with the new Department of Defense (DoD)
policies. As part
of that program,
WPI developed
a PPP certificate
program based on
the systems security engineering
needs requested by
Raytheon and others in the industry.
The WPI PPP program, launched in 2013,
continues to help expand Raytheon’s
workforce proficiency in systems security engineering. Raytheon plans on leveraging the WPI
program to educate our workforce in PPP implementation as we develop holistic approaches
to vulnerability assessment and secure designs
that integrate across anti-tamper, information
assurance (cyber security), software assurance
and supply chain risk management.
WPI’s PPP program complements Raytheon’s
in-house intensive training for cyber experts
and its awareness training for all employees
by providing an additional training option for
those who want to learn more about PPP but
are not enrolled in the cyber expert training.
The program is accredited and can be used as
part of a master’s degree program.
Technology Development
and Sponsorship
Raytheon is partnering with WPI for research
in digital computing design and architecting.
WPI’s strength in this area is evident in its recent success in the Defense Advanced Research
Projects Agency (DARPA) robotics challenge
and ongoing research in processing intensive
applications, such as communications and
radar.
Raytheon collaborated with WPI in applying
field programmable gate arrays (FPGAs) to
software defined radios (SDR). Traditionally,
applying FPGAs to engineering problems has
presented challenges due to the specialized
knowledge required by the designer. As part
of the Raytheon-sponsored WPI research,
model-based development tools and techniques were investigated to help ameliorate
these challenges, specifically for the development of FPGA-based SDR solutions. The tool
set developed for this research utilized different modulation and demodulation models
that were simulated and analyzed using the
MathWorks Simulink® and MATLAB® tools.
The resulting designs were translated into
very high-speed integrated circuit hardware
SPECIAL INTEREST
description language (VHDL) software using the MathWorks HDL Coder™, and then
implemented on a Xilinx Zynq™ 7000 All
Programmable System on a Chip (SoC)
(Figure 1).
Raytheon and WPI are also collaborating in
other research areas such as distributed beam
forming, where algorithms are being developed
to control the transmission of a common message by multiple distributed antennas with the
goal of maximizing the signal strength at the
receiving location (Figure 2).
WPI professors contributed to Raytheon
research on the modeling and management of
co-located radio frequency emitters (RF) such
as the ones that exist on the Navy’s Zumwaltclass ships, and in the real-time processing of
three-dimensional radar images using general
purpose graphics processing units (GP-GPUs).
Raytheon also sponsored WPI research on
improving the performance of satellite and terrestrial communication links.
ZedBoardTM supplied by Avnet, Inc.
Figure 1. The test hardware for the software defined radio (SDR) implementation
utilizes a Xilinx Zynq™ 7000 All Programmable System on a Chip (SoC).
Figure 2. Algorithms are being developed
to control the transmission of a message
from multiple distributed antennas to
optimize the performance of wireless
communication systems.
across the world to take advantage of their
unique capabilities and leading edge research,
and to provide stimulating and important
learning opportunities for future engineers
and technologists. These partnerships also help
Raytheon identify new technologies and potential employees to help enhance our leadership
position in defense and aerospace systems
through world-class people, innovation and
technology. •
Stephen Freitag, James McGrath
and Elizabeth Wilson, Ph.D.
These technical collaborations enable research
in important areas and offer WPI students and
professors the opportunity to solve challenging
real-world problems.
Raytheon Careers
WPI graduates follow rigorous curricula
with multiple project completion requirements. Their problem-solving skills, developed
through theory and practice, prepare them
to help solve challenging engineering and
technological problems, such as those found
at Raytheon.
Many Raytheon employees are WPI alumni.
Among them are significant contributors
nationwide. Some have risen to leadership
and senior leadership ranks; some are subject
matter experts and are well-respected for
their knowledge and leadership. Among the
WPI alumni are product line vice presidents,
engineering vice presidents and our Chief
Technology Officer.
The WPI PPP
program complements
Raytheon’s in-house
intensive training for
cyber experts and
its awareness training
for all employees by
providing an additional
training option for those
who want to learn more
about PPP but are not
enrolled in the cyber
expert training.
The Raytheon-WPI relationship is an example
of how Raytheon is partnering with universities
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
45
SPECIAL INTEREST
Raytheon’s Partnership
With the Franklin W. Olin
College of Engineering
Gives Students
Real-world,
Project-based
Experience
Raytheon has a legacy of supporting science,
technology, engineering and mathematics
(STEM) programs to foster the creation of the
next generation of university graduates and
researchers. Programs such as MathMovesU®
look to engage middle school students in
math and science by showing them how these
disciplines connect to everyday life.
Raytheon actively supports programs like
FIRST® (For Inspiration and Recognition of
Science and Technology) Robotics and the
Team America Rocketry Challenge (TARC)
to expand STEM initiatives beyond middle
school. Additionally, Raytheon actively
supports research and development (R&D)
activities through university-directed research
projects and contract R&D programs. One
such unique partnership that Raytheon has
supported for the past five years with the
Franklin W. Olin College of Engineering is
the Senior Capstone Program in Engineering
(SCOPE).
Olin College — which opened in 2002
in Needham, Mass. — seeks to redefine
undergraduate engineering education by
emphasizing a project-based curriculum
to prepare students to become engineering innovators who address broad societal
needs. From their first semester at the college,
students take classes that involve hands-on
projects. In SCOPE, which is the culmination of this project-based curriculum, teams
of students work for an academic year with a
corporate partner on a real-world engineering
project that the sponsor values. The program
has been very successful in preparing students to function in the real-world industry
environment when they become practicing
engineers upon graduation.
The SCOPE team. From left: Sasha Sproch, Anton Frolenkov, Vidie Pong, Terry Kirn, Amy
Whitcombe and Rob Leoni.
1
Olin College’s collaboration with Raytheon in
particular gives students experience working on industry research-oriented projects.
While many SCOPE projects involve research
and development, the Raytheon projects —
some of which are described below — are
unique in that they are much more research
than development oriented. Students who
are considering a research-oriented career in
academia or industry gain valuable insight
into how research is executed within a major
defense company. Additionally, the Raytheon
projects have required students to use some
of the more abstract material covered in Olin
College’s curriculum (particularly signal
processing), providing students experience
applying abstract theory to real-world engineering problems.
Raytheon recently began its sixth year supporting the Olin College SCOPE program,
and it has helped students explore some exciting research topics throughout its history.
For example, Raytheon collaborated with
Woods Hole Oceanographic Institution to
create a SCOPE research project on the energy
management of buoys that can be deployed
in remote locations as part of the Ocean
Observatories Initiative1 (OOI). The project
team explored the use of various energyharvesting systems and used this knowledge to
establish a power management simulator for
the remote buoy system. Another Raytheon
SCOPE team looked at the feasibility of using
communications waveforms to connect a set
of networked X-band radars and explored
what a deployment scenario might look like
to realize this capability. A follow-on SCOPE
team took this feasibility study and demonstrated X-band communications waveforms
The Ocean Observatories Initiative (OOI) is a National Science Foundation funded (NSF-funded) program to establish a networked sensor system to monitor the ocean and seafloor.
46 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
SPECIAL INTEREST
Data
Communications
Waveform
Signal to
Transmit
Radar
Non-Linear
Amplifier
Transmitted
Signals
Active Electronically
Steered Antenna
Radar
Waveform
Figure 1. Graphic of communication and radar waveforms transmitted through a single
power amplifier.
between separate sites in collaboration
with the National Science Foundation’s
Collaborative Adaptive Sensing of the
Atmosphere (CASA) program, which is being led by the University of Massachusetts at
Amherst.
The past three Raytheon SCOPE teams
(2011–2013) have focused on investigating
ways of mitigating the negative effects of sending multiple signals simultaneously through a
single radio frequency (RF) transmitter. This
is a vibrant area of research for both civilian
and defense systems as the development of
successful mitigation methods could lead to
significantly reduced system cost and footprint because one system would be capable
of performing many simultaneous functions
that today require multiple systems. In mobile
telecommunication systems, these mitigation techniques will enable more efficient
Figure 2. SCOPE lab set-up for the current
Raytheon project. The team uses a sophisticated set of test equipment to monitor and
analyze the performance of their multiwaveform tests.
use of valuable, limited spectrum resources
as ever higher bandwidth data connections
are required such as those called for in the
long-term evolution (LTE) advanced standard
that requires bandwidths up to 100 MHz. The
defense community has also been pursuing
this research in order to enable the transition
to integrated multifunction, multibeam radio
frequency (RF) arrays, which will reduce the
number of systems and associated antennas
required for each individual function.
These new multifunction RF systems must be
capable of simultaneously supporting communications and radar missions. While radar
transmitters require extremely high RF power,
communication systems are focused on signal
integrity and spectral purity with adequate
RF power. Today’s multifunction transmitter
design approaches require significant compromises in the RF power, prime power, thermal
management and spectral purity in order to
perform both missions. The final stage of RF
amplification in the transmitter part of these
systems determines, to a large extent, the capabilities of the overall system and therefore the
SCOPE teams have been focusing on understanding the relationship between the amplifier
and signal interplay to develop improved design approaches through the use of new classes
of waveforms (Figure 1).
Throughout the projects, the Raytheon SCOPE
teams have evaluated existing multifunction
system designs as well as new designs by developing and validating a unified simulation and
experimental test bench (Figure 2) that is
applicable to Raytheon’s unique and specific
needs. The 2010–2011 team focused on
developing the basic simulation infrastructure
using an industry standard tool, Agilent’s
Advanced Design System. The team used a
generic, somewhat idealized, model for the
power amplifier in order to focus their effort
on the signal-side of the problem. A high-fidelity model for the amplifier and an experimental
test bench for verification and validation of
the simulated performance were developed by
the 2011–2012 team. The current SCOPE team
developed example pairs of radar and communication waveforms, which were exercised
both in the simulator and on the test bench.
Through the cumulative efforts of all three
teams, they have identified, simulated, and
tested advanced signal construction concepts
that may lead to improved multifunction transmitter performance. They also identified future
areas of research that can further advance the
state-of-the-art.
The partnership between Raytheon and Olin
College pairs two advocates of the engineering design process and demonstrates a model
of project-based learning that can be used to
engage the next generation of research engineers to help create future technologies. While
the SCOPE collaboration has yielded positive
and fruitful research, it is the development of
the next generation of engineering innovators
that likely yields the most important benefits.
By providing real-world interesting experience to these college engineers, they are being
encouraged to pursue careers in STEM after
graduation, thus helping to ensure a technically
strong workforce for the future.
Quoting Anton Frolenkov (2012–2013 SCOPE
team member) who is currently pursuing a
master’s degree in electrical engineering at the
University of Michigan, Ann Arbor, “I really
appreciated the lessons I learned and have
since been able to apply. As with anything,
the amount of reward is directly proportional
to the amount of effort exerted. I felt like I
worked hard to push the project along and as
such learned a lot about debugging and research project management along with obvious
technical lessons. I also enjoyed the collaboration aspect of SCOPE. On many levels, I wish
it were possible to devote as much time to the
project as a full-time internship affords.” •
Terry Kirn
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
47
M
United States
Patents
Issued to Raytheon
MARK W. BIGGS, TIMOTHY R. SCHEMPP, GREGORY S. UM
8344946 Single frequency user ionosphere system and techinique
MIKEL J. WHITE
8362849 Broadband balun
RICHARD A. POISEL
8345597 Systems and methods for detecting spread spectrum
signals in an area of interest
HEE KYUNG KIM, CLIFTON QUAN, ALBERTO F. VISCARRA,
FANGCHOU YANG
8362856 RF transition with three-dimensional molded
RF structure
MATTHEW L. SHAW, MATTHEW G. WOODARD
At Raytheon, we encourage people to work on
technological challenges that keep America
strong and develop innovative commercial
products. Part of that process is identifying and
protecting our intellectual property (IP). Once
again, the U.S. Patent Office has recognized our
engineers and technologists for their contributions in their fields of interest. We congratulate
our inventors who were awarded patents
from January through June 2013.
DONALD PRICE, GARY SCHWARTZ, WILLIAM G. WYATT
8341965 Method and system for cooling
MICHAEL BRENNAN, EDWARD DEZELICK, LUIS GIRALDO,
BRETT GOLDSTEIN, MICHAEL MILLSPAUGH, JOHN RYAN,
ROBERT WALLACE
8342069 Device and method for controlled breaching of
reinforced concrete
ROBERT D. TRAVIS
8342070 Methods and apparatus for a control surface restraint
and release system
SCOTT H. ALLEN, JONATHAN T. LONGLEY,
JAMES H. ROONEY III
8342281 Hull robot steering system
TERRY M. SANDERSON, DAVID R. SAR
8342457 Shape-changing structure member with embedded
spring
TIMOTHY A. MURPHY, TIMOTHY R. WERCH
8342867 Free floating connector engagement and retention
system and method for establishing a temporary electrical
connection
MATTHEW R. DEXTER, MARION P. HENSLEY
8345639 Broad propagation pattern antenna
BLAISE ROBITAILLE
8346035 Two stage integrator assembly
DOUGLAS BROWN, GEOFF HARRIS, DANIEL MITCHELL
8347814 Method and apparatus for coating a curved surface
THOMAS W. BASTIAN
8350201 Systems, apparatus and methods to compensate for roll
orientation variations in missile components
NATHAN M. MINTZ, KALIN SPARIOSU
8350223 Quantum dot based radiation source and radiometric
calibrator using the same
MARY K. HERNDON, MATTHEW A. MORTON,
PAYAM SHOGHI
8350777 Metamaterial radome/isolator
MARK A. GLOUDEMANS, DAVID E. MUSSMANN,
THOMAS E. YOUNG
8351534 Distributed maximal ratio combining receiver
architecture
IVANS S. CHOU, CLARA CURIEL, LAWRENCE C. DE PAULA,
FREDERICK C. MERTZ, ROBERT K. PINA,
KARLEEN G. SEYBOLD
8351770 Imaging station and method for repeatable alignment
of images
CINDY W. MA, DEREK PRUDEN, KEVIN C. ROLSTON,
ALBERTO F. VISCARRA
8354595 Adhesive reinforced open hole interconnect
ROBERT W. MARTIN
8354627 Torsion stop deployment system for airborne object
PREMJEET CHAHAL, FRANCIS J. MORRIS
8343806 Hermetic packaging of integrated circuit components
RAYMOND D. EPPICH
8354833 Method for characterizing dielectric loss tangent
JESSE H. BLAKE, CARLOS E. GARCIA,
MATTHEW G. MURPHY
8344302 Optically-coupled communication interface for a laserguided projectile
JACOB KIM, GILBERT M. SHOWS
8354958 Alignment system
CHRIS E. GESWENDER
8344304 Methods and apparatus for missile air inlet
JOHN P. BETTENCOURT, NICHOLAS J. KOLIAS
8344359 Integrated thermoelectric heat pump to cool
GaN transistors
PREMJEET CHAHAL
8344430 Multiple substrate electrical circuit device
JASON M. BAIN, CHARLES M. DE LAIR,
RUDY A. EISENTRAUT, MATTHEW A. OFFOLTER
8344525 Electrical power initiator system and method
PETER R. DRAKE, YUCHOI F. LOK
8344937 Methods and apparatus for integration of distributed
sensors and airport surveillance radar to mitigate blind spots
MARK ACKERMAN, SCOTT R. CHEYNE, JEFFREY PAQUETTE
8355255 Cooling of coplanar active circuits
NATHAN M. MINTZ, MARK R. SKIDMORE
8356775 Space object deployment system and method
EDDIE R. BROCK
8358240 Generating a time deterministic, spectrally noncoherent
signal
STEPHEN JACOBSEN, DAVID MARCEAU, FRASER M. SMITH
8358462 A mini-scope for multi-directional imaging
THOMAS FARLEY, TINA A. OBERAI, JERRY L. PIPPINS JR.,
RICARDO J. RODRIGUEZ, NOAH Z. STAHL, DANIEL TEIJIDO,
JAY J. VISARIA
8359357 Secure e-mail messaging system
ANTHONY J. DELROCCO, DANIEL TEIJIDO
8359641 Multi-level secure information retrieval system
CHI-YUNG CHANG, MICHAEL Y. JIN
8344942 System and method for resolving ambiguity in radar,
lidar, and acoustic systems
JAIME ROBLEDO
8360396 Leverage tool for a crank assembly of a radar system
VERNON R. GOODMAN
8344944 Method and system for continuous wave interference
suppression in pulsed signal processing
WILLIAM B. KING, CHAUNCHY F. MCKEARN
8362410 Source-independent beam director and control system
for a high-energy electromagnetic radiation source
48 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
TIEN M. NGUYEN, JOHN J. WOOTAN
8362945 Systems and methods for detecting and tracking gun
barrels using millimeter waves
PETER ROZITIS, KEVIN WAGNER
8363317 Flourescence microscopy focus drive
SCOTT R. CHEYNE, JOSEPH R. ELLSWORTH,
JEFFREY PAQUETTE
8363413 An assembly to provide thermal cooling
GREGORY V. BURNETT, MICHAEL K. HOLZ,
NATHAN M. MINTZ, BRANDON W. PILLANS,
ANTHONY ROSS
8364439 System and method for detection of concealed cargo
in a vehicle by center of mass measurement
RALPH H. KLESTADT, JAVIER VELEZ
8367993 Aerodynamic flight termination system and method
SCOTT T. JOHNSON, SHADI S. MERHI
8368208 Semiconductor cooling apparatus
FRANCIS J. MORRIS, BRANDON W. PILLANS,
MIKEL J. WHITE
8368491 Systems and methods for providing high-capacitance
RF MEMS switches
MARTT HARDING, MICHAEL D. PIXLEY
8368559 Network of traffic behavior-monitoring unattended
ground sensors
DAVID R. BISHOP, JERRY M. GRIMM,
JAMES F. KVIATKOFSKY, GARY F. WAHLQUIST,
KUANG-YUH WU
8368610 Shaped ballistic radome
KENN S. BATES
8368760 System and method for dynamic infrared targeting
DELMER D. FISHER, BRADY A. PLUMMER,
ROBERT W. PLUMMER
8369062 Detonation control system
MARION P. HENSLEY
8369445 A system and method for improved communication
utilizing velocity related information
ALEKSEY NOGIN, DAVID PAYTON
8370422 Establishing common interest negotiation links between
consumers and suppliers to facilitate solving a resource allocation
problem
JAIME ROBLEDO
8371006 Rotary mechanical vibration mechanism
JAMES H. DUPONT, STEVEN J. ELDER
8371204 Bubble weapon system and methods for inhibiting
movement and disrupting operations of vessels
JAMES F. ASBROCK, BRYAN W. KEAN, KANON LIU
8373458 Read out integrated circuit
KENT P. PFLIBSEN, DARIN S. WILLIAMS
8374404 Iris recognition using hyper-spectral signatures
ERIC P. LAM, CHRISTOPHER A. LEDDY, STEPHEN R. NASH,
HARRISON A. PARKS
8374453 Integrating image frames
MATTHEW L. SHAW, MATTHEW G. WOODARD
8375045 Translating a binary data stream using binary markup
language (BML) schema
JOHN R. STALEY
8375620 Weapon sight having multi-munitions ballistics computer
ERIK T. DALE, RYAN A. EGBERT
8375861 Projectile that includes a gimbal stop
GEORGE L. FIX, JAMES R. STURGES, ROBERT E. WALSH
8378212 Sealed electrical feed-through assembly and methods
of making the same
JOHN M. BRANNING JR., ROBERT A. LEMIRE
8378881 Systems and methods for collision avoidance in
unmanned aerial vehicles
WILLIAM P. HAROKOPUS
8378893 Patch antenna
JAR J. LEE, STAN W. LIVINGSTON, CLIFTON QUAN
8378905 Airship mounted array
MARK S. HAUHE, CLIFTON QUAN, ROHN SAUER
8378916 Systems and methods for providing a
reconfigurable groundplane
PAUL A. MEREMS
8387540 Interceptor projectile and method of use
TIMOTHY D. SMITH
8387914 Solar thermal rotorcraft
STEFAN T. BAUR, ADAM M. KENNEDY, DAVID R. RHIGER
8389947 Method and system for detecting neutron radiation
MATTHEW S. NOELL
8390269 Non-destructive determination of functionality of an
unknown semiconductor device
JAMES CARROLL, JOHN G. HESTON, JOHN R. STANTON
8390395 High power RF switch with active device size tapering
RICHARD A. GILSTRAP, CHUL J. LEE
8390508 Generating radar cross-section signatures
JAR J. LEE, STAN W. LIVINGSTON, JEFFREY B. WEBER,
FANGCHOU YANG
8390520 Dual-patch antenna and array
CASEY T. STREUBER
8390801 Angle measurement for a wide field-of-view (WFOV)
semi-active laser (SAL) seeker
GARY A. FRAZIER
8379293 Method and apparatus for modulating light
DOUGLAS M. ZOUCHA
8391549 Methods and systems for processing data using
product-law symmetry detection
DONALD DENIS, CATHERINE GREENHALGH, STAN SZAPIEL
8379321 Extended depth of field imaging
PAUL H. GROBERT, WILLIAM K. WALLACE
8391772 GPS aided open loop coherent focusing
YVONNE C. LEVENSON, TERRY M. SANDERSON,
DAVID R. SAR
8382042 Structure with reconfigurable polymer material
STEPHEN JACOBSEN, MARC OLIVIER
8392036 Point and go navigation system and method
MICHAEL CRIST, DONALD P. GRAHAM, MARY HEWITT,
HECTOR M. REYES JR.
8384540 Systems and methods for detecting and geo-locating
hazardous refuse
CHARLES T. HANSEN
8384588 Beam stabilization for wideband phase comparison
monopulse angle estimation with electronically steered antennas
GORDON R. SCOTT
8384609 RF aperture coldplate
STEPHEN JACOBSEN, JAMES H. ROONEY III,
FRASER M. SMITH
8386112 Vessel hull robot navigation subsystem
MICHAEL R. JOHNSON, BRUCE E. PEOPLES,
JONATHON P. SMITH
8386489 Applying formal concept analysis to validate expanded
concept types
MARK A. GLOUDEMANS
8386902 Low-complexity method for rational puncturing of
forward error correcting codes
ROBERT CAVALLERI, LLOYD KINSEY JR.
8387360 Integral thrust vector and roll control system
GARRETT L. HALL, MICHAEL R. JOHNSON
8387507 Weapon interceptor projectile with deployable frame
and net
TIMOTHY J. IMHOLT
8387534 Detonation device comprising nanocomposite explosive
material
TERRY M. SANDERSON, DAVID R. SAR,
PHILIP C. THERIAULT
8387536 Interceptor vehicle with extendible arm
STEVEN J. ELDER
8387538 Projectile having casing that includes multiple flachettes
MICHAEL K. BURKLAND, CASEY T. STREUBER,
KRISTOFER E. TVEDT
8392143 Fixed-source array test station for calibration of a
semi-active laser (SAL) seeker
BRIAN RICHARD BOULE, JONATHAN T. LONGLEY,
JAMES H. ROONEY III
8393286 Hull robot garage
JOEL N. HARRIS, JEREMY C. HERMANN,
HOWARD R. KORNSTEIN, JONATHAN T. LONGLEY,
JAMES H. ROONEY III, WEN-TE WU
8393421 Hull robot drive system
RALPH PENSEL
8393422 Serpentine robotic crawler
E. RUSS ALTHOF, SCOTT A. MUSE, WALTER S. POPE,
WAYNE K. WOODALL
8398347 Integrated nutplate and clip for a floating fastener and
method of manufacture and assembly
DOUGLAS BROWN, MARK HANDEREK, GEOFF HARRIS,
ANTHONY LIGHT, DANIEL MITCHELL
8398776 Method and apparatus for supporting workpieces in
a coating apparatus
KENN S. BATES, GENE P. COCHRAN
8399872 System and method for using an optical isolator in laser
testing
DONALD DENIS
8400712 Methods and apparatus for providing a split field of
view in an optical sight
THOMAS B. PEDERSON
8400875 Active sonar system and active sonar method using a
pulse sorting transform
THOMAS BIDIGARE, DANIEL CHANG
8401466 Scalable high speed MIMO-satellite communication
system
NICHOLAS W. BARRETT, CHRISTOPHER MARKLEY,
JOSHUA T. PYLE
8402030 Textual document analysis using word cloud comparison
MICHAEL R. JOHNSON, BRUCE E. PEOPLES,
BRIAN J. SIMPSON
8402046 Conceptual reverse query expander
JAMES H. DUPONT, JEFFREY H. KOESSLER
8402895 Vortice amplified diffuser for buoyancy dissipater and
method for selectable diffusion
JAMES H. BOSTICK
8403106 Man-portable non-lethal pressure shield
THOMAS A. OLDEN
8403267 Ejection system and a method for ejecting a payload
from a payload delivery vehicle
GEORGE F. BARSON, WILLIAM P. HULL JR.,
JAMES IRION II, JAMES S. WILSON
8405548 Multi-orientation phased antenna array
and associated method
RICHARD GOOCH, ALAN ROSCOW
8405550 Near-vertical direction finding and geolocation system
BUU DIEP
8393526 System and method for packaging electronic devices
GENTIAN JAKLLARI, JASON REDI, CESAR A. SANTIVANEZ
8406175 Carrier sense multiple access (CSMA) protocol with
multi-packet reception (MPR) in a wireless ad hoc network
MATTHEW JONAS
8393849 Method and system for adjusting a position of an object
MONTY D. MCDOUGAL
8407324 Dynamic modification of the address of a proxy
ANU AGARWAL, JUEJUN HU, FRANK B. JAWORSKI
8394329 Optical device for detection of agent
MARK W. REDEKOPP
8407639 Systems and methods for mapping state elements
of digital circuits for equivalence verification
MICHAEL D. AMBROSE, BRETT A. ANDERSON,
JAMES ROBERT BETZ, EDWARD G. ROBINSON,
BRYAN J. WALLACE
8395393 Cable test method
ERIC HUANG, DAVID L. II, DEEPAK KHOSLA
8396730 System and method for resource allocation and
management
BRYAN D. GLICK, DONALD R. KRETZ, RODERIC W. PAULK
8396877 Method and apparatus for generating a fused view of
one or more people
DOUGLAS BROWN, GEOFF HARRIS, ALEXANDRE LIFCHITS,
DANIEL MITCHELL
8398251 Method and apparatus for fabricating a precision optical
surface
JAMES BARGER
8408115 Systems and methods for an indicator for a weapon
sight
STEVIE ALEJANDRO, CHRIS E. GESWENDER, PAUL VESTY
8410412 Guidance control for spinning or rolling vehicle
LACY G. COOK, JOHN F. SILNY
8411268 Two material achromatic prism
RYAN A. EGBERT, CHRISTOPHER L. HERNANDEZ
8411362 Optical element retaining system for sensor systems
GREGORY E. LONGERICH, DAVID C. ROBILLARD
8412482 Multi-channel electronic acceleration switch
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
49
BRYAN J. WALASCHEK
8412745 Relational database model optimized for the use and
maintenance of watchlist data in a high demand environment
DELMAR L. BARKER, JOHN WARREN BECK,
WILLIAM RICHARD OWENS
8425735 Fabrication of pillared graphene
JAMES BARGER, SCOTT RITTER
8437223 Systems and methods for detecting shooter location
from an aircraft
SUZANNE P. HASSELL, STEPHEN P. MARRA,
KRISTINA L. STEWART, JASON M. SURPRISE
8413115 Specifying integration points of a system-of-systems
CHRIS E. GESWENDER
8426788 Guidance control for spinning or rolling projectile
ERWIN W. BATHRICK, SUNG I. PARK, DENH T. SY
8437250 Determining paths that satisfy a reliability requirement
WILLIAM P. HULL JR., ROBERT E. LEONI,
JAMES S. WILSON
8427238 Performance optimization of power amplifier
SAMUEL S. BLACKMAN, BRIAN A. CRONIN,
NICHOLAS J. PLOPLYS
8437972 Sensor bias estimation for multiple hypothesis tracker
JOEL E. LAMENDOLA, STANLEY J. POREDA,
MAXIM RAYKIN
8427363 Determining whether a track is a live track or a virtual
track
DAVID M. DORIA
8438128 Empirical modeling of confusion matrices
JOSHUA EDMISON, JOHN-FRANCIS MERGEN
8413154 Energy-aware computing environment scheduler
PAUL B. HAFELI, ELI HOLZMAN, AARON J. STEIN,
MICHAEL VARGAS
8413320 Method of gold removal from electronic components
ROGER W. GRAHAM
8415623 Processing detector array signals using stacked readout
integrated circuits
MOUNGI BAWENDI, SCOTT GEYER, FRANK B. JAWORSKI
8415759 Down-converting and detecting photons
SCOTT E. ADCOOK, CARL D. COOK
8416071 Relative location determination of mobile sensor nodes
SALVATORE BELLOFIORE, DAVID J. KNAPP,
ALPHONSO A. SAMUEL, GLAFKOS K. STRATIS
8416127 Dynamic calibration radar system
LACY G. COOK
8416407 Optical spectrometer with wide field of view fore-optics
DARIN S. WILLIAMS
8416986 Methods and systems for processing data using nonlinear slope compensation
FREDERICK B. KOEHLER, WARD D. LYMAN
8418455 Shape memory alloy separating apparatuses
RICHARD M. LLOYD
8418623 Multi-point time spacing kinetic energy rod warhead
and system
ERIC C. FEST, JIM R. HICKS, JAMES P. MILLS,
NICHOLAS D. TRAIL
8421003 Optical transceiver built-in test (BIT)
JOSEPH F. BORCHARD, WILLIAM H. WELLMAN
8422005 Method and apparatus for multiple field-angle optical
alignment testing
LACY G. COOK, JOHN F. SILNY
8422011 Two material achromatic prism
RAFAEL A. IHLY, SHAH A. NEJAD, CHAD WANGSVICK
8422951 Wireless communication system and method for wireless
signal communication in flight vehicles
DAVID FUCIARELLI, DAVID L. II, JAMES R. ZUBER
8423224 Methods and apparatus for controlling deployment
of systems
ANDREW B. FACCIANO, ROBERT D. TRAVIS,
DALE O. WIDMER, SANDIE S. WORLEY
8424438 Multi-stage rocket, deployable raceway harness
assembly and methods for controlling stages thereof
ROBERT A. BAILEY
8424439 Systems and methods for launching munitions
ALF L. CARROLL III, ERNEST C. FACCINI,
LUCINE KABAKIAN REINBOLD, NATHANIEL J. WARD
8424442 Tile grid substructure for pultruded ballistic screens
MICHAEL R. JOHNSON, BRIAN J. LUKOW
8424444 Countermeasure systems including pyrotechnicallygimbaled targeting units and methods for equipping vehicles with
the same
50 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
JEROME H. POZGAY
8427370 Methods and apparatus for multiple beam aperture
JEROME H. POZGAY
8427371 RF feed network for modular active aperture
electronically steered arrays
DAVID D. CROUCH
8427382 Power combiner/divider for coupling n-coaxial inputs/
outputs to a waveguide via a matching plate to provide minimized
reflection
LACY G. COOK
8427744 All-reflective relayed focal telescope derived from the
first two mirrors of an afocal three-mirror anastigmat
ROBERT D. STULTZ
8427769 Multi-stage Lyot filter and method
ROBERT C. EARL, GEORGE M. HUME JR.
8428204 Recovering distorted digital data
DOUGLAS W. COX, PHILIP J. MILLIS
8428992 Estimating training development hours
BRADLEY M. BIGGS, TIM B. BONBRAKE,
GEORGE D. BUDY
8430028 Shock dampened explosive initiator assembly and
method for dampening shock within a delivery vehicle
PHILIP C. THERIAULT
8430578 Separation of main and secondary inertial
measurements for improved line of sight error of an imaging
vehicle’s isolated detector assembly
JAMES GABURA
8432433 Method and apparatus for colorizing a monochrome
infrared image
CHRISTOPHER FLETCHER, FRANK B. JAWORSKI
8432467 Integrated detection and display imaging system
and method
BRIEN ROSS, STAN SZAPIEL
8432610 Multi-magnification viewing and aiming scope
JONATHAN HABIF
8433070 Systems and methods for stabilization of interferometers
for quantum key distribution
STEPHEN JACOBSEN, DAVID MARCEAU
8434208 Method for manufacturing a complex structure
THOMAS R. BERGER, SAMI DAOUD,
MICHAEL J. VILLEBURN
8434411 Cluster explosively-formed penetrator warheads
JOHN CORRADINI
8436285 Projectile that includes a fin adjustment mechanism
with changing backlash
LACY G. COOK, IAN S. ROBINSON, YIFAL J. SHAHAM
8436992 Low distortion spectrometer
JERRY HINSON
8438201 Digital fractional integrator
ROBERT HARROVER, JOHN S. LEAR, JOHN E. STEM,
KENNETH W. WRIGHT, JULIAN A. ZOTTL
8438296 Playback communications using a unified
communications protocol
WALTER C. MILLIKEN
8438401 Device and method for securely storing data
KENT P. PFLIBSEN, CASEY T. STREUBER
8440972 Radiation detector with microstructured silicon
EDWARD P. SMITH
8441087 Direct readout focal plane array
MIKEL J. WHITE
8441322 Broadband linearization by elimination of harmonics
and intermodulation in amplifiers
EMMANUEL NEGATU, THEODORE VORNBROCK,
JOHN GEORGE WITZEL
8441360 Search and rescue using ultraviolet radiation
MICHAEL G. ADLERSTEIN
8441385 Power digital to analog converter
THOMAS R. BERGER, SAMI DAOUD,
MICHAEL J. VILLEBURN
8444784 Insensitive munition-type BNCP explosive material
and methods for forming the same
WAYNE L. SUNNE
8445822 One-piece nano/nano class nanocomposite optical
ceramic (NNOC) extended dome having seamless
non-complementary geometries for electro-optic sensors
DARIN S. WILLIAMS, RICHARD J. WRIGHT
8445823 Guided munition systems including combustive dome
covers and methods for equipping guided munitions with the
same
VICTOR D. KRUPPA, RICHARD A. SCHMIDT
8445836 System for maintaining an even temperature distribution
across a laser detector
ANDREW M. PIPER, MATTHEW D. THOREN
8445864 Method and apparatus for anti-biofouling of a protected
surface in liquid environments
TERRY C. CISCO, MARY A. TESHIBA
8446230 Microwave directional coupler
CHARLES A. CORWIN, DAVID C. FISHER, SARAH E. LAW
8446310 Method and system for locating signal jammers
MICHAEL K. BURKLAND
8447550 Compact fixed-source array test station for calibration
of a semi-active laser (SAL) seeker
JOE H. LINDLEY, GARY THOMAS
8448127 Software forecasting system
RICHARD GUERRERO, NICHOLAS L. SELLIER
8448246 Protecting sensitive email
ALF L. CARROLL III, DONALD L. CHALOUPKA,
KATHERINE E. HOFFMAN, NATHANIEL J. WARD
8448561 Weapon mount
MAREK GAJEWSKI
8456134 System and method for applying a plurality of energy
pulses to a cathode for rapid depolarization of batteries
BRIAN M. WHITE, COLIN WHITE
8448718 Modified hand-held tool
THOMAS E. WOOD
8456350 Methods and apparatus for sea-state measurement
via radar sea-clutter eccentricity
JAYSON KAHLE BOPP, JAMES A. NEGRO
8449009 Adapter for attachment of a display unit to
a console of a vehicle
MARY K. HERNDON, RALPH KORENSTEIN, CHAE DEOK LEE
8450185 Semiconductor strutures having directly bonded heat
sinks and methods for making such structures
LARRY A. BOOKER, JOHN W. GERSTENBERG
8450951 System and method for gyrotron power regulation
THOMAS E. WOOD
8456352 Methods and apparatus for automatic STC from
sea-state measurement via radar sea-clutter eccentricity
DOUGLAS BROWN, GEOFF HARRIS, DANIEL MITCHELL
8456740 Method and apparatus for pathlength adjustment
in an optical system
CHARLES S. KUEHL, WILLIAM T. STIFFLER
8457034 Airborne communication network
MICHAEL L. FORSMAN, JAMES J. MAYS,
MICHAEL L. WILLIAMS
8468246 System and method for allocating resources in
a distributed computing system
RANDALL S. BROOKS, DANIEL TEIJIDO
8468344 Enabling multi-level security in a single-level
security computing system
BRADLEY T. FORD, RANDY S. JENNINGS,
KEVEN K. KALKBRENNER, MONTY D. MCDOUGAL,
BRIAN N. SMITH, WILLIAM E. STERNS
8468602 System and method for host-level malware detection
TIMOTHY E. ADAMS, CHRISTOPHER MOSHENROSE,
JAMES A. PRUETT
8469330 Systems and methods for securing objects to vehicles
GARY A. FRAZIER, CODY B. MOODY,
BRANDON W. PILLANS
8451070 Self-powered microelectromechanical oscillator
DMITRI JERDEV
8457418 Local area contract enhancement
ROBERT R. CLARKSON
8471567 Circuit for detection of failed solder-joints on array
packages
BENJAMIN J. ANNINO, ROGER L. CLARK,
WILLIAM W. COOPER, MARK J. GUGLIUZZA
8451071 Low noise oscillators
JOHN J. COOGAN, KEVIN PETERSON
8457437 System and method for enhancing registered images
using edge overlays
KENNETH A. ESSENWANGER
8471646 Wideband, differential signal balun for rejecting
common mode electromagnetic fields
STEVEN T. CUMMINGS, JOSEPH LICCIARDELLO,
PETER D. MORICO, STEPHEN J. PEREIRA,
JEROME H. POZGAY, ANGELO M. PUZELLA,
JAMES A. ROCHE JR., MICHAEL G. SARCIONE
8451165 Mobile radar system
JOHN GRIFFITH III, JOHN-FRANCIS MERGEN, CARL POWELL
8458491 Cryptographically scrubbable storage device
NATHAN GOODMAN, ROBERT MICHAEL PAWLOSKI,
ALPHONSO A. SAMUEL
8471758 Virtual aperture radar (VAR) imaging
CHARLES T. HANSEN
8451173 Maximum likelihood angle estimation of wideband
signals using phased array antennas
ALEXANDRA L. BLAKE, DAVID B. HATFIELD,
NICHOLAS B. SACCKETTI, LAWRENCE A. WESTHOVEN JR.,
DARIN S. WILLIAMS, RICHARD J. WRIGHT
8461501 Guided munitions including self-deploying dome covers
and methods for equipping guided munitions with the same
STEVE E. HUETTNER
8451186 System and method for passive protection of an
antenna feed network
JONATHAN COMEAU, MATTHEW A. MORTON,
EDWARD WADE THOENES
8461901 Harmonic reject mixer with active phase mismatch
compensation in the local oscillator path
LAWRENCE P. STRICKLAND
8452254 Selecting and routing sub-signals from a common
signal path
BARBARA J. BLYTH, RICHARD DELONG
8462042 Generating a kinematic indicator for combat
identification classification
JERRY HINSON
8452826 Digital frequency channelizer
JAMES E. HARDIN, BENJAMIN PIERCE III,
THOMAS P. ZAHM
8462861 Methods and apparatus for multipath mitigation
MARIA ANTONIA ANDREWS,ROBERT B. BATIE,
ALEN CRUZ, LUISITO D. ESPIRITU, STEPHAN GONZALEZ,
SYLVIA A. TRAXLER
8453212 Accessing resources of a secure computing network
MELVIN CAMPBELL, ETHAN S. HEINRICH,
KEVIN C. ROLSTON, ROSALIO S. VIDAURRI,
ALBERTO F. VISCARRA, DAVID T. WINSLOW
8453314 Process for forming channels in a flexible circuit
substrate using an elongated wedge and a channel shaped
receptacle
JIMMY CLARK, MICHAEL DOAN,
WILFRIED KRONE-SCHMIDT, DAVID LA KOMSKI,
ALBERTO E. SCHROTH, CHRIS SHAFFER, STEPHEN SMITH,
SHANNON WHITE
8453393 Encapsulated and vented particulate thermal insulation
JAMES D. KUENEMAN, ANTON VANDERWYST
8453426 Current controlled field emission thruster
RICHARD DRYER, CHRIS E. GESWENDER
8453572 Gun-fired propellant support assemblies and
methods for same
BUU DIEP, ROLAND GOOCH
8454789 Disposable bond gap control structures
MICHAEL L. BREST, KENNETH L. MCALLISTER,
RICHARD N. MULLINS
8455826 Variable aperture mechanism retention device
GREG S. NEATH, JOHN W. ROSENVALL
8463612 Monitoring and collection of audio events
MICHAEL R. JOHNSON, BRUCE E. PEOPLES,
JONATHON P. SMITH
8463808 Expanding concept types in conceptual graphs
JAMES F. KVIATKOFSKY, MARK A. NAMEY,
JAMES R. TOPLICAR
8464949 Method and system for countering an incoming threat
TIMOTHY J. IMHOLT
8465201 Electro-magnetic radiation detector
RICHARD A. FUNK, DAVID J. KNAPP,
CHADWICK B. MARTIN
8466407 Stray light baffles for a conformal dome with arch
corrector optics
MARTIN S. DENHAM
8466816 Method and apparatus for serializing bits
JOSEPH J. ICHKHAN, DAVID A. ROCKWELL,
JOHN H. SCHROEDER
8467426 Method and apparatus for cooling a fiber laser
or amplifier
GEORGE C. ADAMS, ELIZABETH M. BEGIN,
SEANNA J. GITTLER, ARTHUR B. JOHNSON,
KENRIC P. NELSON, SAMUEL H. ROSENTHAL,
BRIAN J. SCANNELL, ERIC SCHEID, NORA T. TGAVALEKOS
8468111 Determining confidence of object identification
DAVID L. JOHANSEN, DIANA J. KENNEDY, TOBY REED
8471905 Methods and apparatus for imaging
MARK T. BUSCH, LACY G. COOK, IAN S. ROBINSON
8471915 Self-correcting adaptive long-stare electro-optical
system
HOWARD C. CHOE, DEWEY R. MYERS,
CLIFFORD C. SAMMONS, LARRY R. SPLITTER,
8473929 Architecture tailoring system
JAMES F. ASBROCK, JOHN E. CLEMENT,
WILLIAM O. MCKEAG
8477291 System and method for ranging of targets
JOHN BEDINGER, ROBERT B. HALLOCK,
THOMAS E. KAZIOR, MICHAEL A. MOORE,
KAMAL TABATABAIE
RE44303 Passivation layer for a circuit device and method
of manufacture
International Patents
Issued to Raytheon
Titles are those on the U.S.-filed patents; actual titles
on foreign counterparts are sometimes modified and
not recorded. While we strive to list current international patents, many foreign patents issue much later
than corresponding U.S. patents and may not yet be
reflected.
AUSTRALIA
MARK E. BEHRENS, DANIEL A. COLICA,
KENNETH W. VIRGIL
2007323898 Automated logistics support system incorporating
a product integrity analysis system
CHRISTOPHER J. GRAHAM, JOHN A. WHEELER,
MATTHEW R. YEAGER
2008318929 Unmanned vehicle route management system
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
51
BRIAN A. ADAMS, CHRISTOPHER HECHT,
JOHN A. WHEELER, MATTHEW R. YEAGER
2008318930 Unmanned vehicle simulation system
JAMES W. CASALEGNO, MICHAEL F. JANIK,
THOMAS MCHALE, KENNETH J. MCPHILLIPS,
ARNOLD W. NOVICK, ILYA ROZENFELD, JOHN R. SHORT
2009222991 Autonomous sonar system and method
CHARLES A. HALL, THEODORE N. TAHMISIAN JR.
2009241388 Small aperture interrogator antenna system
employing sum difference azimuth discrimination techniques
PETER R. DRAKE, YUCHOI F. LOK
2009244465 Methods and apparatus for detection/classification
of radar targets including birds and other hazards
K. BUELL, JIYUN C. IMHOLT, MATTHEW A. MORTON
2009300419 Multilayer metamaterial isolator
CANADA
SHANNON DAVIDSON, ROBERT J. PETERSON
2503773 System and method for computer cluster virtualization
using dynamic boot images and virtual disk
FRITZ STEUDEL
2532328 Process for phase-derived range measurements
MARWAN KRUNZ, PHILLIP ROSENGARD
2539080 Encapsulating packets into a frame for a network
QUENTON JONES, MARTIN STEVENS
2561774 Secondary radar message decoding
VERNON R. GOODMAN, DAVID M. SHIFRIN,
TIMOTHY R. HOLZHEIMER
2565775 Generating three-dimensional images using
impulsive radio frequency signals
GIB LEWIS
2580935 Overlapping subarray architecture
JESSE GRATKE, MICHAEL F. JANIK, RYAN LEWIS,
JAMES MILLER, THOMAS B. PEDERSON,
JAMES H. ROONEY III, WILLIAM C. ZURAWSKI
2625683 Sonar system and method providing low probability
of impact on marine mammals
ANDREW B. FACCIANO, RICHARD A. MCCLAIN JR.,
ROBERT T. MOORE, CRAIG SEASLY, RAYMOND J. SPALL
2670325 Detachable aerodynamic missile stabilizing system
CHINA
CHRISTOPHER HIRSCHI, STEPHEN JACOBSEN,
BRIAN MACLEAN, RALPH PENSEL
2007800461684 Conformable track assembly for a robotic
crawler
DAVID R. BISHOP, JERRY M. GRIMM,
JAMES F. KVIATKOFSKY, GARY F. WAHLQUIST,
KUANG YUH WU
2007800360753 Shaped ballistic radome
STEPHEN JACOBSEN, MARC OLIVIER, RALPH PENSEL,
FRASER M. SMITH
2008801029156 Serpentine robotic crawler having a
continuous track
EGYPT
DANIEL FLOYD, DOUGLAS HALL
26080 Method and apparatuses for squelch break signaling
device to provide session initiation protocol
FRANCE
CLARK DAVIS, STEPHEN JACOBSEN,
MARC OLIVIER
1488087 Controllable combustion method and device
GARY A. FRAZIER
1536562 Method and apparatus for generating a pulse of very
narrow width
ALEXANDER A. BETIN, KALIN SPARIOSU
1585202 Scalable laser with robust phase locking
ROBERT P. ENZMANN, FRITZ STEUDEL, GEORGE THOME
1869492 System and method for coherently combining a
plurality of radars
MICHAEL B. SCHOBER
1902329 System and method for passively estimating angle and
range of a source using signal samples collected simultaneously
from a multi-aperture antenna
JOHN BEDINGER, ROBERT B. HALLOCK,
THOMAS E. KAZIOR, MICHAEL A. MOORE,
KAMAL TABATABAIE
1956872 Environmental protection coating system and method
EDWARD KITCHEN, DARIN S. WILLIAMS
1983485 FLIR-to-missile boresight correlation and non-uniformity
compensation of the missile seeker
ROBERT S. BRINKERHOFF, ROBERT CAVALLERI,
JAMES M. COOK, RICHARD D. LOEHR,
MICHAEL J. MAHNKEN
1991825 System and method for attitude control of a flight
vehicle using pitch-over thrusters
DEREK L. BUDISALICH, GEORGE D. BUDY,
ERIK A. FJERSTAD
2002198 Methods and apparatus for integrated locked thruster
mechanism
CHARLES M. DE LAIR, CHRISTOPHER OWAN
2005036 Implicitly timed gear bearings
JONATHAN LYNCH
2064777 Variable cross-coupling partial reflector and method
MICHAEL BRENNAN, EDWARD DEZELICK, LUIS GIRALDO,
BRETT GOLDSTEIN, MICHAEL MILLSPAUGH, JOHN RYAN,
ROBERT WALLACE
2072190 Device and method for controlled breaching of
reinforced concrete
SOLOMON DE PICCIOTTO, BRADLEY D. KELLY
2074006 Method of determining a collision avoidance
MAURICE J. HALMOS
2078968 Synthetic aperture ladar with chirped mode
locked waveform
STEPHEN JACOBSEN
2082159 Serpentine robotic crawler
STEPHEN JACOBSEN
2008801044423 Pressure control valve having an asymetric
valving structure
STEPHEN JACOBSEN, MARC OLIVIER, RALPH PENSEL
2092265 Unmanned ground robotic vehicle having an
alternatively extendible and retractable sensing appendage
DENMARK
KENNETH W. BROWN, JAMES R. GALLIVAN
2115711 Safeguard system for ensuring device operation
in conformance with governing laws
JONATHAN D. GORDON, REZA TAYRANI
2182632 Broadband microwave amplifier
RODNEY J. CALLISON
2145206 Spotlight synthetic aperture radar (SAR) system and
method for generating a SAR map in real-time using a modified
polar format algorithm (PFA) approach
LACY G. COOK, JOSHUA THORNES
2153498 Optical pulse-width modifier structure
VETIS B. DAVIS, JOSE I. RODRIGUEZ
2167902 Method and apparatus for rapid mounting and
dismounting of a firearm accessory
JONATHAN D. GORDON, REZA TAYRANI
2182632 Broadband microwave amplifier
WILLIAM P. HAROKOPUS, DARRELL W. MILLER
2214256 Composite radome and radiator structure
PATRICK HOGAN, RALPH KORENSTEIN, JOHN MCCLOY,
CHARLES WILLINGHAM JR.
2234936 Treatment method for optically transmissive bodies
ROBERT CAVALLERI, THOMAS A. OLDEN
2245291 Pellet loaded attitude control rocket motor
CRAIG BRADFORD, MARC A. BROWN, FRANK HITZKE,
WILLIAM E. KOMM, MICHAEL W. LITTLE,
DOMENIC F. NAPOLITANO, DAVID A. SHARP,
DOUGLAS VEILLEUX II
2262676 Autonomous data relay buoy
JAMES W. CASALEGNO, MICHAEL F. JANIK,
THOMAS MCHALE, KENNETH J. MCPHILLIPS,
ARNOLD W. NOVICK, ILYA ROZENFELD, JOHN R. SHORT
2263097 Autonomous sonar system and method
RICHARD M. WEBER, WILLIAM G. WYATT
2274965 Cooling system for a computing rack
CHARLES A. HALL, THEODORE N. TAHMISIAN JR.
2281324 Small aperture interrogator antenna system employing
sum difference azimuth discrimination techniques
INUKA D. DISSANAYAKE, DONALD M. HUGHES
2298039 Method and apparatus for an ionizer
BRANDON H. ALLEN, KEVIN W. CHEN,
WILLIAM P. HAROKOPUS, KERRIN A. RUMMEL,
GARY L. SEIFERMAN, RICHARD M. WEBER
2313946 Heat removal system for a radome
PAUL H. BARTON, RAYMOND R. BESHEARS,
BERNARD D. HEER, CARL KIRKCONNELL,
ROBERT R. OGDEN, BRADLEY A. ROSS
2326893 Monitoring the health of a cryocooler
CHRIS E. GESWENDER, SHAWN B. HARLINE,
NICHOLAS E. KOSINSKI
2335007 Projectile with filler material between fins and fuselage
THOMAS K. DOUGHERTY, STEVEN E. LAU, CINDY W. MA,
CHRISTOPHER T. SNIVELY, WILLIAM J. WOLFGONG
2356681 X-ray opaque coatings and application
CARY C. KYHL
2388786 Temperature tolerant cover layer construction
TERRY C. CISCO
2390954 Microwave directional coupler
TERRY M. SANDERSON
2411673 Shape-change material and method
LACY G. COOK
2418528 Pointable optical system with coude optics having a
short on-gimbal path length
LACY G. COOK, ERIC M. MOSKUN
2423726 Wide field of view LWIR high speed imager
52 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
DANIEL J. MOSIER, DAVID J. PARK
2427719 Method and system of aligning a track beam
and a high energy laser beam
JAMES T. SCHLEINING, MICHAEL P. UNGER,
STEPHEN D. WITHERSPOON
2433084 Guided missile
JOHN BEDINGER, ROBERT B. HALLOCK,
THOMAS E. KAZIOR, MICHAEL A. MOORE,
KAMAL TABATABAIE
2469993 Environmental protection coating system and method
GERMANY
ROBERT P. ENZMANN, FRITZ STEUDEL, GEORGE THOME
1869492 System and method for coherently combining a plurality
of radars
MICHAEL B. SCHOBER
1902329 System and method for passively estimating angle and
range of a source using signal samples collected simultaneously
from a multi-aperture antenna
JOHN BEDINGER, ROBERT B. HALLOCK,
THOMAS E. KAZIOR, MICHAEL A. MOORE,
KAMAL TABATABAIE
1956872 Environmental protection coating system and method
EDWARD KITCHEN, DARIN S. WILLIAMS
1983485 FLIR-to-missile boresight correlation and non-uniformity
compensation of the missile seeker
ROBERT S. BRINKERHOFF, ROBERT CAVALLERI,
JAMES M. COOK, RICHARD D. LOEHR,
MICHAEL J. MAHNKEN
1991825 System and method for attitude control of a flight
vehicle using pitch-over thrusters
DEREK L. BUDISALICH, GEORGE D. BUDY,
ERIK A. FJERSTAD
2002198 Methods and apparatus for integrated locked
thruster mechanism
CHARLES M. DE LAIR, CHRISTOPHER OWAN
2005036 Implicitly timed gear bearings
ROBERT CAVALLERI, THOMAS A. OLDEN
2245291 Pellet loaded attitude control rocket motor
CAREY C. KYHL
6020110021079 Temperature tolerant cover layer construction
CRAIG BRADFORD, MARC A. BROWN, FRANK HITZKE,
WILLIAM E. KOMM, MICHAEL W. LITTLE,
DOMENIC F. NAPOLITANO, DAVID A. SHARP,
DOUGLAS VEILLEUX II
2262676 Autonomous data relay buoy
CLARK DAVIS, STEPHEN JACOBSEN, MARC OLIVIER
603441165 Controllable combustion method and device
CHARLES A. HALL, THEODORE N. TAHMISIAN JR.
2281324 Small aperture interrogator antenna system employing
sum difference azimuth discrimination techniques
INUKA D. DISSANAYAKE, DONALD M. HUGHES
2298039 Method and apparatus for an ionizer
BRANDON H. ALLEN, KEVIN W. CHEN,
WILLIAM P. HAROKOPUS, KERRIN A. RUMMEL,
GARY L. SEIFERMAN, RICHARD M. WEBER
2313946 Heat removal system for a radome
PAUL H. BARTON, RAYMOND R. BESHEARS,
BERNARD D. HEER, CARL KIRKCONNELL,
ROBERT R. OGDEN, BRADLEY A. ROSS
2326893 Monitoring the health of a cryocooler
CHRIS E. GESWENDER, SHAWN B. HARLINE,
NICHOLAS E. KOSINSKI
2335007 Projectile with filler material between fins and fuselage
TERRY C. CISCO
2390954 Microwave directional coupler
TERRY M. SANDERSON
2411673 Shape-change material and method
LACY G. COOK
2418528 Pointable optical system with coude optics having
a short on-gimbal path length
JAMES T. SCHLEINING, MICHAEL P. UNGER,
STEPHEN D. WITHERSPOON
2433084 Guided missile
GREECE
MICHAEL BRENNAN, EDWARD DEZELICK, LUIS GIRALDO,
BRETT GOLDSTEIN, MICHAEL MILLSPAUGH, JOHN RYAN,
ROBERT WALLACE
3080102 Device and method for controlled breaching of
reinforced concrete
HUNGARY
DOUGLAS M. KAVNER
228601 System and method for reading license plates
ISRAEL
MICHAEL K. BURKLAND, DAVID B. HATFIELD,
ELAINE E. SEASLY
174557 Molecular containment film modeling tool
GERALD L. EHLERS, CHARLES LEPPLE, AARON WATTS
177660 Personal authentication device
STEVEN COTTEN, BENJAMIN DOLGIN, MICHAEL SHORE
191589 Positioning system and method
JAR J. LEE, STAN W. LIVINGSTON, CLIFTON QUAN
193821 Airship mounted array
KENNETH GERBER, ROBERT GINN
195265 Method of construction of CTE matching structure
with wafer processing and resulting structure
FRANK N. CHEUNG
195990 Data translation system and method
PATRICK M. KILGORE
196759 System and method for adaptive non-uniformity
compensation for a focal plane array
JAR J. LEE, STAN W. LIVINGSTON, CLIFTON QUAN
196879 Dual band space-fed array
JOHN BEDINGER, ROBERT B. HALLOCK,
THOMAS E. KAZIOR, MICHAEL A. MOORE,
KAMAL TABATABAIE
2469993 Environmental protection coating system and method
STEPHEN JACOBSEN
198710 Serpentine robotic crawler
ALEXANDER A. BETIN, KALIN SPARIOSU
6020040411140 Scalable laser with robust phase locking
CHRISTIAN HEMMI, JAMES MASON
199299 Polarization control method for phased arrays
GARY A. FRAZIER
6020040421723 Method and apparatus for generating
a pulse of very narrow width
JOHN S. ANDERSON, CHUNGTE CHEN
199967 Common aperture optical system incorporating
a light sensor and a light source
JONATHAN D. GORDON, REZA TAYRANI
6020060340516 Broadband microwave amplifier
LACY G. COOK, JOSHUA THORNES
201467 Optical pulse-width modifier structure
MAURICE J. HALMOS
2078968 Synthetic aperture ladar with chirped mode
locked waveform
STEPHEN JACOBSEN
6020070297583 Serpentine robotic crawler
PATRIC M. MCGUIRE
201560 Methods and apparatus for selecting a target from
radar tracking data
KENNETH W. BROWN, JAMES R. GALLIVAN
2115711 Safeguard system for ensuring device operation
in conformance with governing laws
STEPHEN JACOBSEN, MARC OLIVIER, RALPH PENSEL
6020070297621 Unmanned ground robotic vehicle having an
alternatively extendible and retractable sensing appendage
ITALY
LACY G. COOK, JOSHUA THORNES
2153498 Optical pulse-width modifier structure
THOMAS K. DOUGHERTY, STEVEN E. LAU, CINDY W. MA,
CHRISTOPHER T. SNIVELY, WILLIAM J. WOLFGONG
6020090130323 X-ray opaque coatings and application
JONATHAN LYNCH
2064777 Variable cross-coupling partial reflector and method
MICHAEL BRENNAN, EDWARD DEZELICK, LUIS GIRALDO,
BRETT GOLDSTEIN, MICHAEL MILLSPAUGH, JOHN RYAN,
ROBERT WALLACE
2072190 Device and method for controlled breaching of
reinforced concrete
SOLOMON DE PICCIOTTO, BRADLEY D. KELLY
2074006 Method of determining a collision avoidance
VETIS B. DAVIS, JOSE I. RODRIGUEZ
2167902 Method and apparatus for rapid mounting
and dismounting of a firearm accessory
RICHARD M. WEBER, WILLIAM G. WYATT
6020090138839 Cooling system for a computing rack
WILLIAM P. HAROKOPUS, DARRELL W. MILLER
2214256 Composite radome and radiator structure
DANIEL J. MOSIER, DAVID J. PARK
6020100066916 Method and system of aligning a track beam
and a high energy laser beam
PATRICK HOGAN, RALPH KORENSTEIN, JOHN MCCLOY,
CHARLES WILLINGHAM JR.
2234936 Treatment method for optically transmissive bodies
LACY G. COOK, ERIC M. MOSKUN
6020110013572 Wide field of view LWIR high speed imager
ROBERT P. ENZMANN, FRITZ STEUDEL, GEORGE THOME
1869492 System and method for coherently combining a plurality
of radars
JOHN BEDINGER, ROBERT B. HALLOCK,
THOMAS E. KAZIOR, MICHAEL A. MOORE,
KAMAL TABATABAIE
1956872 Environmental protection coating system and method
ROBERT S. BRINKERHOFF, ROBERT CAVALLERI,
JAMES M. COOK, RICHARD D. LOEHR,
MICHAEL J. MAHNKEN
1991825 System and method for attitude control of a flight
vehicle using pitch-over thrusters
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
53
MICHAEL BRENNAN, EDWARD DEZELICK, LUIS GIRALDO,
BRETT GOLDSTEIN, MICHAEL MILLSPAUGH, JOHN RYAN,
ROBERT WALLACE
2072190 Device and method for controlled breaching
of reinforced concrete
JONATHAN D. GORDON, REZA TAYRANI
2182632 Broadband microwave amplifier
CRAIG BRADFORD, MARC A. BROWN, FRANK HITZKE,
WILLIAM E. KOMM, MICHAEL W. LITTLE,
DOMENIC F. NAPOLITANO, DAVID A. SHARP,
DOUGLAS VEILLEUX II
2262676 Autonomous data relay buoy
CHARLES A. HALL, THEODORE N. TAHMISIAN JR.
2281324 Small aperture interrogator antenna system employing
sum difference azimuth discrimination techniques
JOHN BEDINGER, ROBERT B. HALLOCK,
THOMAS E. KAZIOR, MICHAEL A. MOORE,
KAMAL TABATABAIE
2469993 Environmental protection coating system and method
JAPAN
REGINA ESTKOWSKI, PETER TINKER
5172326 System and method for adaptive path planning
SCOTT R. CHEYNE, JEFFREY PAQUETTE
5181079 An electrical connector to connect circuit cards
DAVID D. HESTON, JON MOONEY
5183199 Integrated circuit resistor
JAMES BALLEW, SHANNON DAVIDSON
5184552 Computer storage system
MICHAEL G. ADLERSTEIN, JOHN C. TREMBLAY
5192691 Radio frequency limiter circuit
JAMES H. DUPONT, RICHARD D. LOEHR,
WILLIAM N. PATTERSON
5242809 Buoyancy dissipator and method to deter an
errant vessel
PREMJEET CHAHAL, ATHANASIOS SYLLAIOS
5248771 Integrated spectroscopic microbolometer with
microfilter arrays
RANDY C. BARNHART, CRAIG S. KLOOSTERMAN,
MELINDA C. MILANI, DONALD V. SCHNAIDT,
STEVEN TALCOTT
5256019 Data handling in a distributed communication network
JOHN SELIN
5259182 Quadrature offset power amplifier
DAVID D. CROUCH, WILLIAM E. DOLASH,
MICHAEL J. SOTELO
5259184 Multiple-port patch antenna
ROBERT C. GIBBONS
5265332 Imaging system
ROLAND TORRES
5272083 +28V aircraft transient suppression
FRANCIS J. MORRIS
5279731 Method for fabricating electrical circuitry on
ultra-thin plastic films
ROBERT W. BYREN, DAVID SUMIDA, MICHAEL USHINSKY
5280998 Solid-state laser with spatially-tailored active ion
concentration using valence conversion with surface masking
and method
STEPHEN JACOBSEN
5285701 Modular robotic crawler
ARYEH PLATZKER, DOUGLAS TEETER
5198700 Transistor amplifier having reduced parasitic oscillations
CHARLES A. HALL, THEODORE N. TAHMISIAN JR.
5290403 Small aperture interrogator antenna system employing
sum difference azimuth discrimination techniques
FRITZ STEUDEL
5202844 Process for phase-derived range measurements
STEPHEN JACOBSEN, TOMASZ J. PETELENZ
5295223 Digital wound detection system
JAMES FLORENCE, CLAY E. TOWERY
5203710 Method and apparatus for safe operation of an
electronic firearm sight depending upon the detection
of a selected color
MICHAEL K. HOLZ, IRL W. SMITH
5296158 Wide-angle beam steering system
ANDREW B. FACCIANO, CHIN SHIAU
5215322 Digital interface unit (DIU) and method for controlling
stages of a multi-stage missile
EDWARD M. JACKSON, HEE KYUNG KIM, CLIFTON QUAN,
KEVIN C. ROLSTON, FANGCHOU YANG
5221606 Multi-layer microwave corrugated printed circuit board
and method
STEPHEN JACOBSEN, MARC OLIVIER
5232177 Antagonistic fluid control system for active and passive
actuator operation
STEPHEN JACOBSEN
5232227 Pressure control valve having an assymetric valving
structure
JAMES BARGER, MARSHALL BRINN,
STEPHEN D. MILLIGAN, RICHARD MULLEN
5232847 Systems and methods for disambiguating shooter
locations
DAVID CORKUM, MICHAEL K. HOLZ
5236853 Electronically variable light attenuator
SOLOMON DE PICCIOTTO, BRADLEY D. KELLY
5237278 Method of determining a collision avoidance
54 2014 ISSUE 1 RAYTHEON TECHNOLOGY TODAY
ARYEH PLATZKER, DOUGLAS TEETER
5296181 Transistor amplifier having reduced parasitic oscillations
NETHERLANDS
JONATHAN D. GORDON, REZA TAYRANI
2182632 Broadband microwave amplifier
SOUTH KOREA (REPUBLIC OF KOREA)
WILLIAM J. DAVIS, WARD G. FILLMORE,
SCOTT MACDONALD
101219619 Method for packaging semiconductors at
a wafer level
KARL G. DAXLAND, FREDERICK FRODYMA,
JOHN R. GUARINO, NAMIR W. HABBOOSH,
WILLIAM HORAN, RAYMOND JANSSEN,
LEONARD V. LIVERNOIS, DAVID A. SHARP
101231829 Method and apparatus for acoustic system having
a transceiver module
MICHAEL G. ADLERSTEIN
101236454 Integrated thermal sensor for microwave transistors
BORIS S. JACOBSON
101236455 Integrated smart power switch
VINH ADAMS, DENNIS BRAUNREITER, WESLEY DWELLY
101240876 Pseudo-orthogonal waveforms radar system,
quadratic polyphase waveforms radar, and methods for locating
targets
JAMES FLORENCE, CLAY E. TOWERY
101252871 Method and apparatus for safe operation of an
electronic firearm sight
JAMES FLORENCE, CLAY E. TOWERY
101252929 Method and apparatus for safe operation of an
electronic firearm sight depending upon the detection of a selected
color
BORIS S. JACOBSON
101262919 Integrated smart power switch
FRANCIS J. MORRIS, BRANDON W. PILLANS,
MIKEL J. WHITE
101268208 Systems and methods for providing high-capacitance
RF MEMS switches
SPAIN
MICHAEL BRENNAN, EDWARD DEZELICK, LUIS GIRALDO,
BRETT GOLDSTEIN, MICHAEL MILLSPAUGH, JOHN RYAN,
ROBERT WALLACE
2072190 Device and method for controlled breaching of
reinforced concrete
JAMES W. CASALEGNO, MICHAEL F. JANIK,
THOMAS MCHALE, KENNETH J. MCPHILLIPS,
ARNOLD W. NOVICK, ILYA ROZENFELD, JOHN R. SHORT
2263097 Autonomous sonar system and method
CHARLES A. HALL, THEODORE N. TAHMISIAN JR.
2281324 Small aperture interrogator antenna system employing
sum difference azimuth discrimination techniques
JOHN BEDINGER, ROBERT B. HALLOCK,
THOMAS E. KAZIOR, MICHAEL A. MOORE,
KAMAL TABATABAIE
2469993 Environmental protection coating system and method
SWEDEN
WESLEY T. DULL, JEROME H. POZGAY
101220952 System and technique for calibrating radar array
STEVEN D. BERNSTEIN, RALPH KORENSTEIN,
STEPHEN J. PEREIRA
101227923 Fabricating a gallium nitride layer with diamond layer
STEVEN D. BERNSTEIN, RALPH KORENSTEIN,
STEPHEN J. PEREIRA
101227925 Fabricating a gallium nitride device
with a diamond layer
JONATHAN D. GORDON, REZA TAYRANI
2182632 Broadband microwave amplifier
SWITZERLAND
MICHAEL BRENNAN, EDWARD DEZELICK, LUIS GIRALDO,
BRETT GOLDSTEIN, MICHAEL MILLSPAUGH, JOHN RYAN,
ROBERT WALLACE
2072190 Device and method for controlled breaching of
reinforced concrete
TAIWAN
SCOTT M. HESTON
382652 Amplifying a signal using a control modulator that
provides a bias resistance
KARL G. DAXLAND, FREDERICK FRODYMA,
JOHN R. GUARINO, NAMIR W. HABBOOSH,
WILLIAM HORAN, RAYMOND JANSSEN,
LEONARD V. LIVERNOIS, DAVID A. SHARP
384251 Method and apparatus for acoustic system having a
transceiver module
SCOTT H. ALLEN, JOEL N. HARRIS,
HOWARD R. KORNSTEIN, JONATHAN T. LONGLEY,
TIMOTHY I. MANN, JAMES H. ROONEY III
389820 Hull robot with rotatable turret
JOEL N. HARRIS, STEPHEN JACOBSEN,
JONATHAN T. LONGLEY, JAMES H. ROONEY III,
FRASER M. SMITH
391295 Hull robot
BORIS S. JACOBSON
392230 Integrated smart power switch
BRIAN RICHARD BOULE, JONATHAN T. LONGLEY,
JAMES H. ROONEY III
395689 Hull robot garage
JESSE GRATKE, MICHAEL F. JANIK, RYAN LEWIS,
JAMES MILLER, THOMAS B. PEDERSON,
JAMES H. ROONEY III, WILLIAM C. ZURAWSKI
398662 Sonar system and method providing low probability of
impact on marine mammals
TURKEY
MICHAEL BRENNAN, EDWARD DEZELICK, LUIS GIRALDO,
BRETT GOLDSTEIN, MICHAEL MILLSPAUGH, JOHN RYAN,
ROBERT WALLACE
2072190 Device and method for controlled breaching of
reinforced concrete
JAMES W. CASALEGNO, MICHAEL F. JANIK,
THOMAS MCHALE, KENNETH J. MCPHILLIPS,
ARNOLD W. NOVICK, ILYA ROZENFELD, JOHN R. SHORT
2263097 Autonomous sonar system and method
CHARLES A. HALL, THEODORE TAHMISIAN JR.
2281324 Small aperture interrogator antenna system employing
sum difference azimuth discrimination techniques
UNITED KINGDOM
CLARK DAVIS, STEPHEN JACOBSEN, MARC OLIVIER
1488087 Controllable combustion method and device
GARY A. FRAZIER
1536562 Method and apparatus for generating a pulse of very
narrow width
ALEXANDER A. BETIN, KALIN SPARIOSU
1585202 Scalable laser with robust phase locking
ROBERT P. ENZMANN, FRITZ STEUDEL, GEORGE THOME
1869492 System and method for coherently combining a plurality
of radars
MICHAEL B. SCHOBER
1902329 System and method for passively estimating angle and
range of a source using signal samples collected simultaneously
from a multi-aperture antenna
JOHN BEDINGER, ROBERT B. HALLOCK,
THOMAS E. KAZIOR, MICHAEL A. MOORE,
KAMAL TABATABAIE
1956872 Environmental protection coating system and method
EDWARD KITCHEN, DARIN S. WILLIAMS
1983485 FLIR-to-missile boresight correlation and non-uniformity
compensation of the missile seeker
ROBERT S. BRINKERHOFF, ROBERT CAVALLERI,
JAMES M. COOK, RICHARD D. LOEHR,
MICHAEL J. MAHNKEN
1991825 System and method for attitude control of a flight
vehicle using pitch-over thrusters
DEREK L. BUDISALICH, GEORGE D. BUDY,
ERIK A. FJERSTAD
2002198 Methods and apparatus for integrated locked thruster
mechanism
CHARLES DE LAIR, CHRISTOPHER OWAN
2005036 Implicitly timed gear bearings
JONATHAN LYNCH
2064777 Variable cross-coupling partial reflector and method
MICHAEL BRENNAN, EDWARD DEZELICK, LUIS GIRALDO,
BRETT GOLDSTEIN, MICHAEL MILLSPAUGH, JOHN RYAN,
ROBERT WALLACE
2072190 Device and method for controlled breaching of
reinforced concrete
SOLOMON DE PICCIOTTO, BRADLEY D. KELLY
2074006 Method of determining a collision avoidance
MAURICE J. HALMOS
2078968 Synthetic aperture ladar with chirped mode locked
waveform
STEPHEN JACOBSEN
2082159 Serpentine robotic crawler
STEPHEN JACOBSEN, MARC OLIVIER, RALPH PENSEL
2092265 Unmanned ground robotic vehicle having an
alternatively extendible and retractable sensing appendage
KENNETH W. BROWN, JAMES R. GALLIVAN
2115711 Safeguard system for ensuring device operation in
conformance with governing laws
RODNEY J. CALLISON
2145206 Spotlight synthetic aperture radar (SAR) system and
method for generating a SAR map in real-time using a modified
polar format algorithm (PFA) approach
LACY G. COOK, JOSHUA THORNES
2153498 Optical pulse-width modifier structure
VETIS B. DAVIS, JOSE I. RODRIGUEZ
2167902 Method and apparatus for rapid mounting and
dismounting of a firearm accessory
WILLIAM P. HAROKOPUS, DARRELL W. MILLER
2214256 Composite radome and radiator structure
PATRICK HOGAN, RALPH KORENSTEIN, JOHN MCCLOY,
CHARLES WILLINGHAM JR.
2234936 Treatment method for optically transmissive bodies
ROBERT CAVALLERI, THOMAS A. OLDEN
2245291 Pellet loaded attitude control rocket motor
CRAIG BRADFORD, MARC A. BROWN, FRANK HITZKE,
WILLIAM E. KOMM, MICHAEL W. LITTLE,
DOMENIC F. NAPOLITANO, DAVID A. SHARP,
DOUGLAS VEILLEUX II
2262676 Autonomous data relay buoy
JAMES W. CASALEGNO, MICHAEL F. JANIK,
THOMAS MCHALE, KENNETH J. MCPHILLIPS,
ARNOLD W. NOVICK, ILYA ROZENFELD, JOHN R. SHORT
2263097 Autonomous sonar system and method
STEPHEN E. BENNETT, CHRIS E. GESWENDER
2268996 Methods and apparatus for guidance of ordnance
delivery device
RICHARD M. WEBER, WILLIAM G. WYATT
2274965 Cooling system for a computing rack
CHARLES A. HALL, THEODORE TAHMISIAN JR.
2281324 Small aperture interrogator antenna system employing
sum difference azimuth discrimination techniques
INUKA D. DISSANAYAKE, DONALD M. HUGHES
2298039 Method and apparatus for an ionizer
BRANDON H. ALLEN, KEVIN W. CHEN,
WILLIAM P. HAROKOPUS, KERRIN A. RUMMEL,
GARY L. SEIFERMAN, RICHARD M. WEBER
2313946 Heat removal system for a radome
PAUL H. BARTON, RAYMOND R. BESHEARS,
BERNARD D. HEER, CARL KIRKCONNELL,
ROBERT R. OGDEN, BRADLEY A. ROSS
2326893 Monitoring the health of a cryocooler
CHRIS E. GESWENDER, SHAWN B. HARLINE,
NICHOLAS E. KOSINSKI
2335007 Projectile with filler material between fins and fuselage
THOMAS K. DOUGHERTY, STEVEN E. LAU, CINDY W. MA,
CHRISTOPHER T. SNIVELY, WILLIAM J. WOLFGONG
2356681 X-ray opaque coatings and application
CARY C. KYHL
2388786 Temperature tolerant cover layer construction
TERRY C. CISCO
2390954 Microwave directional coupler
TERRY M. SANDERSON
2411673 Shape-change material and method
LACY G. COOK
2418528 Pointable optical system with coude optics having
a short on-gimbal path length
LACY G. COOK, ERIC M. MOSKUN
2423726 Wide field of view LWIR high speed imager
DANIEL J. MOSIER, DAVID J. PARK
2427719 Method and system of aligning a track beam
and a high energy laser beam
JAMES T. SCHLEINING, MICHAEL P. UNGER,
STEPHEN D. WITHERSPOON
2433084 Guided missile
JOHN BEDINGER, ROBERT B. HALLOCK,
THOMAS E. KAZIOR, MICHAEL A. MOORE,
KAMAL TABATABAIE
2469993 Environmental protection coating system and method
DELMER D. FISHER, BRADY A. PLUMMER,
ROBERT W. PLUMMER
2485742 Detonation control device
Raytheon’s Intellectual Property (IP) is valuable. If you become aware of any
entity that may be using any of Raytheon’s proprietary inventions, patents,
trademarks, software, data or designs, or would like to license any of the
foregoing, please contact your Raytheon IP counsel: David Rikkers (IDS),
Craig J. Bristol (IIS), John Horn (MS), Robin R. Loporchio (Corporate) and
Saint St. Julian (SAS).
RAYTHEON TECHNOLOGY TODAY 2014 ISSUE 1
55
Copyright © 2014 Raytheon Company. All rights reserved.
Approved for public release. Printed in U.S.A.
“Customer Success Is Our Mission,” R6s, TracShare and MathMovesU are registered trademarks of Raytheon Company.
TransTalk is a registered trademark of Raytheon BBN Technologies.
ODIN and Shoebox Oscillator are registered trademarks in Australia of Poseidon Scientific Instruments Pty Ltd.
FIRST is a registered trademark of the United States Foundation for Inspiration and Recognition of Science and Technology.
ARM is a registered trademark of ARM Holding, PLC.
Intel is a registered trademark of Intel Corporation.
MatLab and Simulink are registered trademarks of MathWorks, Inc.
ZedBoard is a trademark of Avnet, Inc.
GBS_AM_4341185
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement