Wireless Network Deployment Guidelines

Wireless Network Deployment Guidelines
Wireless Network Deployment Guidelines
Application Note
JN-AN-1059
Revision 1.2
18 December 2013
Wireless Network Deployment Guidelines
Application Note
2
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
Contents
About this Document
5
Organisation
Conventions
Acronyms and Abbreviations
Related Documents
Support Resources
Trademarks
5
6
6
7
7
7
1. Introduction
9
1.1 Protocol Standards
1.2 Wires vs Wireless
9
10
2. Radio Channel Basics
11
2.1 The 2.4-GHz ISM Band - Living with Your Neighbours
2.2 Transmit Power, Receive Sensitivity and Range
3. Antenna Options
11
12
15
3.1 Modules with Integral Antenna
3.2 Modules with RF Connector
15
16
4. Antenna Properties and Types
19
4.1 Antenna Directivity
4.2 Antenna Issues and Types
4.2.1 Isotropic Antenna
4.2.2 Half-Wave Dipole Antenna
4.2.3 Yagi Antenna
4.3 Antenna Gain
4.4 Antenna Polarisation
4.5 Antenna Commissioning
19
21
21
22
23
24
25
25
5. Reflections, Obstructions and Multipath
27
6. Deployment Guidelines
29
6.1
6.2
6.3
6.4
General Guidelines
Multipath and Fade Margins
Open Space Guidelines
Indoor Environment Guidelines
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
29
29
30
31
3
Chapter 1
6.5 Point-to-Point Link Guidelines
33
7. The Benefits of Mesh Networking
35
8. Sharing Channels with Other Radio Systems
37
9. Deployment Checklist
39
4
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
About this Document
This document provides basic guidelines for the deployment of a wireless network
based on the IEEE 802.15.4 protocol - for example, this could be a ZigBee PRO or
JenNet-IP network which is built on IEEE 802.15.4. It introduces some of the
fundamental concepts and issues related to radio communication in these systems.
As such, it will act as a useful reference point for network installation and maintenance
personnel who are not experts in the radio/wireless domain.
Organisation
This document consists of 9 chapters, as follows:
 Chapter 1 presents a short introduction to wireless networks
 Chapter 2 introduces the basic concepts of radio communication, particularly
those relevant to wireless networks.
 Chapter 3 describes the antenna options for modules featuring NXP JN51xx
wireless microcontrollers.
 Chapter 4 presents the main properties of antennae, as well as different
antenna types.
 Chapter 5 addresses factors that affect the propagation of radio signals.
 Chapter 6 presents guidelines for deploying wireless network nodes in different
environments.
 Chapter 7 introduces the advantages of using a wireless network with a mesh
topology.
 Chapter 8 outlines the issues involved in sharing radio frequency bands with
other radio systems.
 Chapter 9 presents a checklist of the main deployment guidelines.
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
5
About this Document
Conventions
Files, folders, functions and parameter types are represented in bold type.
Function parameters are represented in italics type.
Code fragments are represented in the Courier New typeface.
This is a Tip. It indicates useful or practical information.
This is a Note. It highlights important additional
information.
This is a Caution. It warns of situations that may result
in equipment malfunction or damage.
Acronyms and Abbreviations
IEEE
Institute of Electrical and Electronics Engineers
ISM
Industrial, Scientific and Medical
LOS
Line Of Sight
PAN
Personal Area Network
PER
Packet Error Rate
RF
Radio Frequency
WLAN Wireless Local Area Network
6
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
Related Documents
[1]
ZigBee PRO Stack User Guide (JN-UG-3048)
[2]
Module Development Reference Manual (JN-RM-2006)
[3]
Application Note “Antennae for use with JN51xx” (JN-AN-1030)
[4]
Application Note “Packet Error Rate Testing for JN516x” (JN-AN-1175)
[5]
"The Mobile Radio Propagation Channel", David Parsons, Pentech Press
1992, ISBN0-7273-1316-9
Support Resources
To access online support resources such as SDKs, Application Notes and User
Guides, visit the Wireless Connectivity TechZone:
www.nxp.com/techzones/wireless-connectivity
For JN514x resources, visit the NXP/Jennic web site: www.jennic.com/support
Trademarks
All trademarks are the property of their respective owners.
“JenNet” and “JenNet-IP” are trademarks of NXP B.V..
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
7
About this Document
8
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
1. Introduction
The IEEE 802.15.4 wireless network standard has paved the way for a revolution in
the implementation of PANs (Personal Area Networks), with the traditional wiring used
to connect sensors and switches being replaced with radio links. This wireless solution
is hugely attractive in helping to reduce construction costs of new buildings.
Furthermore, a mesh radio system (such as a ZigBee PRO network) can be installed
in older buildings, permitting improvements that would be impractical with traditional
wiring methods.
For the majority of installations, the siting of individual radio nodes will be done without
expert radio knowledge. This Application Note outlines the basic considerations and
rules to enable successful installations in such cases.
1.1 Protocol Standards
You can develop a wireless network based directly on the IEEE 802.15.4 protocol
standard or based on another standard which is itself based on IEEE 802.15.4 - for
example, ZigBee PRO or JenNet-IP, which both offer additional functionality over
IEEE 802.15.4.
Radio equipment developed to the IEEE 802.15.4 standard ensures interoperability that is, radio components from different sources can operate together. When installing
a radio network, adherence to the required standard is essential for interoperability.
The equipment may be required to meet multiple standards - for example, ZigBee
equipment must meet both the IEEE 802.15.4 and ZigBee standards.
NXP's JN51xx range of single-chip IEEE 802.15.4-based devices allows the
development of network products that meet the following standards:
 IEEE 802.15.4
 JenNet-IP
 ZigBee PRO
 ZigBee Remote Control (RF4CE)
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
9
Chapter 1
Introduction
1.2 Wires vs Wireless
The notion of a wired link is a familiar one - a cable joins two connectors, and a link is
made. The link is normally broken by disconnection at either end (or rarely if the
intervening cable is broken, perhaps by accident). Thus, the physical environment of
a wired link (e.g. the office) usually plays little or no part in determining the state of the
link.
For a radio link, the situation is different. There is no cable to act as a secure and
reliable signal path. The link must be made across free-space, through walls, people
and other obstructions. This environment may be constantly changing, such as in a
busy working area with people moving around. Furthermore, the environment may
already contain other wireless systems that seek to "share" the same airwaves as the
system to be deployed. In short, in wireless system deployment, there is usually little
or no control over the deployment environment, which can vary widely.
So, what can be done to ensure that your network is deployed efficiently? First
consider listening to an FM radio. Here you are enjoying the result of an established
radio link deployment. The engineers who built this system have used high towers on
which they have located their transmit antennae. They have engineered the system to
ensure (as much as is reasonably possible) that the radio signal reaches you in good
enough shape for clear audio reproduction. They have considered walls, trees, hills
and all manner of things that could be located between the transmit tower and you.
The fact that you are listening means they have deployed their network effectively.
The successful deployment of a wireless IEEE 802.15.4-based network requires us to
consider some of the same deployment issues as above, but on a smaller physical
scale. The advice we give here is pragmatic and realistic, acknowledging that a
reliable system must be achieved within limited deployment time.
10
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
2. Radio Channel Basics
This chapter introduces some of the basic concepts of radio communication,
particularly those relevant to wireless networks.
2.1 The 2.4-GHz ISM Band - Living with Your Neighbours
NXP's JN51xx wireless microcontrollers are designed to operate in the 2.4-GHz radio
band, which is available worldwide. This band, also know as an ISM (Industrial,
Scientific and Medical) band, has rules which allow many different systems to use it at
the same time. An IEEE 802.15.4-based network may have to share its frequency
space with systems such as Wi-Fi (e.g. IEEE802.11b/g wireless LAN), video
distribution, Bluetooth and cordless telephones. However, the IEEE 802.15.4 protocol
is well suited to shared-band operation. It has 16 separate channels that allow the
system to ‘choose’ a channel that is not being used by either other IEEE 802.15.4based networks or other 2.4-GHz deployed systems. More information on sharing the
frequency band is provided in Chapter 8.
Once set up to use one of these channels, the IEEE 802.15.4 protocol provides a
reliable radio link. The protocol employs a robust modulation scheme that is not easily
disturbed by other 2.4-GHz band users. Furthermore, the radio modem uses Direct
Sequence Spread Spectrum (DSSS) coding that improves transmitter-to-receiver
range and offers some protection against interference.
We have already raised the issue that the radio link can be interrupted by a variety of
factors. The combination of the IEEE 802.15.4 protocol and a higher-level networking
protocol such as ZigBee PRO or JenNet-IP provides a strong partnership in managing
this issue. Received data is checked for corruption (i.e. errors) - using instantaneous
handshaking (message checking), the receiving device can acknowledge a correct or
incorrect message by transmitting a short return message. With this feature enabled,
any unsuccessful messages can be re-sent until they eventually reach their recipient.
Additionally, the ZigBee PRO protocol provides mesh networking that can
automatically send a data packet via an alternative, reliable route. This is described in
Chapter 7 and in the ZigBee PRO Stack User Guide (JN-UG-3048).
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
11
Chapter 2
Radio Channel Basics
2.2 Transmit Power, Receive Sensitivity and Range
The concepts of "transmit power", "receive sensitivity" and "range" are very familiar to
radio engineers and are fundamental to the operation of a radio link. We will now
explore them and see their importance in a wireless network installation.
We will consider these concepts by taking the analogy to an audio system:
 Transmit Power: This is similar to the power that is delivered to a speaker in a
Hi-Fi system - the more power supplied, the stronger the sound signal (and the
further it will travel and still be heard).
 Receive Sensitivity: This is analogous to how well a person's hearing can
detect the sound signal.
 Range: This is equivalent to the maximum distance the sound signal can travel
and still be intelligibly heard. For example, if the sound signal is a news report,
there is a distance from the speaker at which the sound can still be heard, but
beyond this limit the content of the report becomes indiscernible - this distance
defines the range.
For an IEEE 802.15.4-based radio link, the radio transmit power is limited by the local
regulatory radio regulations. NXP's JN51xx-based modules are designed to comply
with these regulations and are available as "standard" and "high-power" modules. In
radio terms, the transmit power of a high-power module is approximately 100 mW, a
hundred times that of a standard module, which is 1 mW.
At the receive end of a radio link, the minimum power level that can be detected is
approximately 1/1000000000 of 1 mW (10-9 mW or 10-12 W). Thus, radio receivers
require only a tiny amount of radio energy to discern a usable signal. This factor is
used to excellent effect in a radio network.
Due to the very small numbers involved, it is convenient to use a logarithmic scale to
express signal levels. This method involves a calculation based on the ratio of two
signal levels, the result expressed in the unit of the decibel (dB) - the calculation is
given by the expression 10log(Po/Pr), where Po is the power level of interest and Pr is
a reference power level. Therefore, the reference power level must be stated and is
normally incorporated into the unit - dBm means dB referred to 1 mW, while dBw
means dB referred to 1 W. Hence, 0 dBm is 1 mW, while 20 dBm is 100 mW (and
corresponds to -10 dBw).
An important concept in radio networks is "line-of-sight" (LOS). As an example,
consider a radio signal broadcast in Space from one satellite to another satellite. If the
two satellites can visually "see" each other, they have a "line-of-sight". In a wireless
network, a LOS link means that the two nodes can "see" each other, but a non-LOS
link is also possible where the two nodes cannot physically "see" each other but can
still communicate (more on this in Chapter 5).
Taking the ideal case of the two satellites once again, assuming they both have
antennae that radiate equally in all directions with no losses, the range between them
is determined by Equation 1 below.
12
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
 P tx
R = ------ ------4 P rx
Equation 1
where R = Achievable range, in metres
 = Free-space wavelength, in metres
Ptx = Transmitter power, in Watts
Prx = Receiver sensitivity, in Watts
Using approximate values for typical JN51xx module performance measurements, we
can calculate that equipping the satellites with standard modules gives a range of
around 700 metres. Using high-power modules, the increase in transmit power
together with an improvement in receive sensitivity pushes this range up ten-fold to
nearly 7 kilometres.
The above calculations are based on "free-space" radio wave propagation (in a
perfect vacuum) and use antennae that are considered to be "isotropic" (radiate
equally in all directions). In reality, there are many real-world issues that challenge
these assumptions and a different result may be obtained.
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
13
Chapter 2
Radio Channel Basics
14
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
3. Antenna Options
Radio systems must both launch radio energy from their transmitter and capture
energy from the airwaves into their receiver - for this they use an antenna. The correct
choice and connection of the antenna is crucial to maximising performance. Thus this
topic warrants detailed consideration.
On JN51xx modules, two antenna options are available - an integral ceramic antenna
or an external antenna connected to a co-axial RF connector.
3.1 Modules with Integral Antenna
The integrated antenna is a small ceramic chip component that extends from the
screening can that surrounds the chip and associated components. While this antenna
is self-contained, we must ensure that it protrudes from the surrounding ground plane
(see Figure 1 below). The Module Development Reference Manual (JN-RM-2006)
describes the requirements for mounting modules that contain this antenna. If
mounted to these requirements, with the antenna well separated from any metallic
components such as casing or stacked circuit boards, the maximum amount of
transmit power is radiated from the antenna.
Figure 1: JN51xx Module with Integral Ceramic Antenna
As a general rule, try to keep metallic structures over 2 cm from the antenna,
preferably 6 cm away. In addition, any casing that encloses the antenna should be
plastic in order to avoid de-tuning and losses. It is also wise to avoid using plastic
casing with dark fillers or colourants, since these may absorb radio energy and reduce
the range considerably. It is worth noting that the principle of reciprocity applies here
- this means that the antenna works in the same way in transmit and receive modes,
with a simple reversal of the energy flow.
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
15
Chapter 3
Antenna Options
3.2 Modules with RF Connector
The JN51xx range also includes modules with co-axial RF connectors, provided for
connection to either a remote or custom antenna (see Figure 2 below).
Figure 2: JN51xx Module with Standard SMA Connector
The available RF connectors include uFl, SMA and reverse polarity SMA. These
modules offer a huge amount of flexibility in the choice and location of the antennae
in a wireless network, but we must be careful in our choices. Some key points to bear
in mind are:
 Connectors: The SMA connector is a specialist RF connector designed to
work well at the required 2.4-GHz frequency (in fact, special versions of this
connector are available to work well beyond 40 GHz). This connector is
particularly suitable for product evaluation and measurement, as it is widely
available on antennae and test equipment. Modules are also available with the
uFl connector - this is a miniature connector, widely used on Wi-Fi and
Bluetooth equipment, offering a smaller footprint and good performance for
high volume production.
 Antenna frequency range: There are an increasing number of connectable
antennae available for all kinds of wireless systems. It is very important that an
antenna is selected that is designed to operate in the 2400-2500 MHz ISM
band. The antenna is a tuned component and antennae designed to work in
other systems, such as mobile phones at 900 MHz, will not function correctly
and cause severe range problems.
 Cables: There are a number of cables available that are designed for 2.4-GHz
operation. They are co-axial cables designed with low-loss dielectric insulation,
plated inner conductor and high-density or double-braided outer shielding. If
cable must be used to relocate an antenna, as is beneficial in some situations,
then it is important to use suitable cable. In addition, the cable must use the
correct co-axial connector.
16
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
Tip: In order to avoid installation and (later) reliability
issues, NXP recommend that cables are bought as
completed and tested sub-assemblies designed for use
at the relevant frequency.
Tip: NXP recommend a range of antennae for use with
the JN51xx modules fitted with RF connectors - refer to
the Application Note "Antennae for use with JN51xx"
(JN-AN-1030).
The recommended antennae are typically low-loss, dipole types. As a general rule,
this type of antenna is about 2 to 4 times more efficient than the integral ceramic type,
achieving up to double the free-space range of the latter. The dipole antenna is much
bigger than the tiny ceramic antenna. There are many situations in which small size is
important (particularly in hand-held equipment) but, generally speaking, good
performance does not come with small antenna size.
It is possible to attach an antenna directly to the module connector, but this is not
recommended for situations where there could be mechanical stress on the antenna
which could potentially damage the module. As with the ceramic antenna, it is
important to mount the connectable antenna away from other metallic structures,
obeying the same 6-cm clearance rule. It is also unlikely that the larger antennae
would be placed inside the equipment case.
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
17
Chapter 3
Antenna Options
18
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
4. Antenna Properties and Types
This section introduces certain key antenna properties and the main types of antenna.
4.1 Antenna Directivity
Have you ever wondered why a TV satellite dish needs to be pointed very precisely in
the direction of the broadcaster's satellite? The dish is a reflector antenna - radio
waves arriving parallel to the axis of the antenna are collected by the dish (normally
part of a paraboloid) and reflected to a common focus on the axis, where they are
converted into electrical signals. This leads to a particular sensitivity to radio waves
arriving from one specific direction, from one small patch of sky - the antenna is said
to be highly directive. As the diameter of a dish antenna is increased, the antenna
becomes more directive. Thus, a radio telescope, such as the 76-metre diameter
Lovell telescope at Jodrell Bank in England, is much more directive than a satellite
dish - this enables astronomers to isolate an extremely tiny patch of sky for
observation, such as that encompassing a distant galaxy.
Figure 3: A Large Directive Antenna - Lovell Radio Telescope at Jodrell Bank
A consequence of a highly directive antenna is that it has very poor sensitivity in other
directions. The sensitivity or "directivity" of an antenna can be mapped as a function
of the angle of arrival of radio waves - the result is referred to as the "radiation pattern"
or "polar diagram" of the antenna. An antenna with equal sensitivity in all directions
will result in a spherical radiation pattern, while a highly directional antenna will have
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
19
Chapter 4
Antenna Properties and Types
a significant "main beam" or "major lobe" (as well as a number of smaller "side lobes")
- these are illustrated in Figure 4 below.
Isotropic Antenna
Directive Antenna
Figure 4: Radiation Patterns for Isotropic and Directive Antennae
The "directivity" of an antenna in a certain direction can be thought of as its sensitivity
as a receiver in that direction. However, it is easier to think of directivity in terms of the
relative power of the antenna in the given direction when used as a transmitter. As a
numerical quantity, the directivity is expressed relative to that of an isotropic antenna
- that is, an antenna which is equally sensitive (as a receiver) or radiates equally (as
a transmitter) in all directions. The directivity of an isotropic antenna in any direction is
defined as 1. The directivity of another antenna, in a certain direction, is then stated
relative to this isotropic case, with the assumption that both antennae radiate the same
total power (integrated over all directions). The directivity is expressed logarithmically
in units of "dBi" - decibels, with the "i" indicating with reference to the isotropic
antenna. Thus, a directivity of 20 dBi indicates that the antenna radiates 100 times
more power than an isotropic antenna in the given direction.
20
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
4.2 Antenna Issues and Types
Wireless networks do not normally require the use of highly directive antennae, but
there are a number of properties of the antennae that must be considered. These
issues are summarised below:
 A single antenna must function well for all likely angles of arrival of radio energy
from other nodes.
 The antenna must, in most cases, be small enough to be compatible with a
small radio node.
 Local radio regulatory rules may restrict the size of the antenna by setting a
limit on the allowed gain (see Section 4.3).
 Where an antenna can be oriented in a particular direction, it is likely that its
direction setting will not be very precise.
The main types of antenna are considered in the subsections below.
4.2.1 Isotropic Antenna
As already mentioned, an "isotropic" antenna radiates equally in all directions as a
transmitter, and is equally sensitive in all directions as a receiver. The antenna
radiation pattern is simply a sphere with directivity of unity in all directions, as shown
in Figure 4. It is not a real-world antenna and exists only in theory. Its significance is
that it provides a reference point for all other antennae - often, antenna directivity and
gain are specified in units of "dBi", which defines the relative directivity or gain of the
real-world antenna against the theoretical isotropic antenna.
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
21
Chapter 4
Antenna Properties and Types
4.2.2 Half-Wave Dipole Antenna
Figure 5 below shows the radiation pattern of a half-wave dipole antenna (for the
Antenna Factor ANT-2.4-CW-RCT-RP antenna, in this example). Note that the dipole
radiates in a similar way to the isotropic case close to the horizontal, but has virtually
no radiation at the vertical.
Z
Figure 5: Half-Wave Dipole Antenna
22
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
4.2.3 Yagi Antenna
The Yagi antenna is highly directive. Figure 6 below shows its radiation pattern (for the
Telex 2.4-GHz WLAN antenna, in this example). Notice that there is a direction in
which the antenna radiates with almost ten times the power of the isotropic antenna.
However, this directivity comes at the expense of power radiated in other directions
and for considerably higher cost. Antennae of this type will give significantly more
range, but only in one direction, and therefore are only really suitable for point-to-point
links (see Section 6.5). Another potential benefit is that the system is less susceptible
to interfering signals from outside the main beam of the antenna.
Z
Z
Figure 6: Yagi Antenna
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
23
Chapter 4
Antenna Properties and Types
4.3 Antenna Gain
Antenna gain is simply the antenna directivity taking into account any inefficiencies in
the antenna, and is usually expressed in dBi (as for directivity - see above). Antenna
inefficiencies can originate in the antenna structure itself, as well as the RF feed and,
if fitted, the cable to the antenna. This becomes significant with very small antennae,
particularly ceramic types. However, for well engineered antennae with good RF
feeds, it is permissible to consider directivity and gain to be interchangeable.
Note: Most antennae datasheets quote peak gain in the
most favourable direction, so it is worth checking the
radiation patterns to see how much variation in gain
occurs in different directions.
Antennae with positive gains boost the range of the radio system, in the direction that
the gain is quoted. Once again, considering the case of the two satellites, as described
in Equation 1, if we now assume they have antennae with particular gains in the
relevant directions then Equation 1 can be adjusted as shown below:
 P tx G 1 G 2
R = ------ -------------------4
P rx
Equation 2
where R = Achievable range, in metres
 = Free-space wavelength, in metres
Ptx = Transmitter power, in Watts
Prx = Receiver sensitivity, in Watts
G1 and G2 = Linear antenna gains for Satellite 1 and Satellite 2 respectively
Often it is simpler to use decibel (dB) quantities, which are routinely quoted in
datasheets - in this case, Equation 2 is rewritten as:
g 1 + g 2 + p tx – p rx

R = ------  10   -------------------------------------------


4
20
Equation 3
where R = Achievable range, in metres
 = Free-space wavelength, in metres
ptx = Transmitter power, in dBm
prx = Receiver sensitivity, in dBm
g1 and g2 = Antenna gains for Satellite 1 and Satellite 2, in dBi, respectively
Note that range increases with higher antenna gain, higher transmit power and better
receiver sensitivity.
24
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
4.4 Antenna Polarisation
A radio transmission is linearly polarised - that is, the electric and magnetic fields in
the radio waves are in fixed orthogonal planes (each containing the direction of wave
propagation). This is because the transmitting antenna has a dominant polarisation
which, by convention, is defined as horizontal or vertical. The range of the broadcast
is optimised by using a receiving antenna with the same polarisation, particularly when
there is a line-of-sight or directional antennae are used. The use of cross-polarised
antennae will result in reduced range. Normally, an antenna will have guidelines and/
or markings that show the dominant polarisation. Where there is no line-of-sight,
polarisation becomes less important, as signals reflected from different objects arrive
with widely differing polarisations (multipath propagation is described in Chapter 5).
4.5 Antenna Commissioning
If we now look at the concept of directivity in terms of commissioning a wireless
network, there are two scenarios to consider:
 Non-engineered: A node is installed in a changing network in which the node
is required to communicate with other nodes in various directions. These
directions may be known or unknown, but they may also change with time as
nodes are added or moved. This scenario requires that all directions are
covered by the antenna's radiation pattern and, therefore, the use of an
antenna with low directivity, such as the half-wave dipole shown in Figure 5.
 Engineered: A node is installed in a fixed network to communicate with one or
a number of fixed nodes. This situation can benefit from using a directive
antenna, if the antenna can be oriented correctly during installation. However,
subject to a lower range performance, this type of network can still use a nondirective antenna.
Tip: JN51xx modules with integral ceramic antenna are
ideal for non-engineered installations (the module's
supporting data contains installation guidelines, but
these are broad). JN51xx modules with RF connector
can be used in either type of installation, with
appropriate antennae.
Tip: NXP recommend the use of half-wavelength
dipoles which, at 2.45 GHz, are relatively small
antennae but effective performers.
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
25
Chapter 4
Antenna Properties and Types
26
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
5. Reflections, Obstructions and Multipath
In a free-space situation with no nearby obstructions, radio waves have a clear path
between two wireless network nodes. In reality, this situation is only approximately
realised when nodes are very close together, perhaps several metres, and there are
no other structures nearby. More often, there are obstructions in the direct line-of-sight
as well as structures around the nodes giving rise to absorption of radio energy and
multiple reflections. This situation is illustrated in Figure 7 below.
Figure 7: A Simple Multipath Situation
This situation is extremely complex and varies with the type of installation. In fact,
radio waves will propagate through brick walls, concrete floors and plasterboard
partitions, among other materials, but a loss will be incurred (as compared with freespace propagation). Depending on the thickness, moisture content and angle of
incidence, a wall may allow between 1/100 and ¼ of the radio power to pass through.
In radio terms, this loss is significant but not a disaster. However, a metal panel or
metallised glass window will not allow much radio power to pass through - the vast
majority of the incident power will be reflected, as if the panel were a mirror (beware
that plasterboard is often backed with metal foil). This can be a serious issue since
losses can build up very quickly. However, some radio power may propagate through
small holes (apertures) in the panel or around metal edges through the process of
diffraction.
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
27
Chapter 5
Reflections, Obstructions and Multipath
Thick layers of building material absorb a lot of radio energy - for example, very thick
stone walls. Such absorption also poses problems for overground-to-underground
communication.
Tip: In these situations, it is advisable to check the
available performance using a site survey tool, such as
the Packet Error Rate (PER) software available for
NXP's JN516x evaluation kits - refer to the Application
Note “Packet Error Rate Testing for JN516x”
(JN-AN-1175).
For particularly difficult environments, it may be advisable to request advice on
antenna types and placement from NXP or external specialists.
An IEEE 802.15.4-based wireless network will not work underwater. In addition, nodes
operating in wet conditions (i.e. in heavy rainfall) may exhibit some degradation. In
most cases, water droplets on or close to the antenna are more serious than the rain
itself, so care should be taken to ensure that enclosures shed water well.
Apart from a few situations, it is very difficult to exclude ALL radio power from an area.
Thus, a radio link can be made to work under most situations with some careful
planning, but there are many situations to consider and radio path losses vary widely.
For example:
 A large public building may have few obstructions between nodes, but large
amounts of metallic panelling that cause reflections.
 A small domestic installation may have numerous brick walls between nodes in
individual rooms.
 An office building may have metallised dry-wall partitions together with
mezzanine ceilings.
 A system in an open-air car park will enjoy relatively few obstructions and may
benefit from elevated gantry sites.
Given the huge variation in situations, many researchers have studied the effects of
the environment on 2.4-GHz radio wave propagation. The techniques used have
varied from deterministic models using ray-tracing techniques to statistical
approaches using curve fitting to measured data. These are used to determine the
siting of individual radio devices.
For some systems, site surveys are used to perform radio link measurements (for
example, using NXP’s PER test software - see the above Tip) specific to the
installation before individual nodes are placed in their final positions. This technique
has been used widely for wireless PBX equipment (cordless telephones) and
professional proprietary radio systems.
It is assumed that a typical installation of a wireless network is unlikely to have the
benefit of a detailed site survey or propagation modelling techniques. The placing of
individual nodes will be largely non-engineered, as they will normally be sensors or
switches with positions determined by other factors. However, a ZigBee PRO network
will have the benefit of mesh networking, where the important factor becomes the
maximum allowable space between individual nodes.
28
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
6. Deployment Guidelines
The following sections offer some guidelines as a starting point for planning purposes.
6.1 General Guidelines
One of the most important factors when placing any node is its height above its
immediate environment. As a general rule in buildings, head height or above is
preferred, and the higher the better for maximising range. If nodes must be placed in
positions very close to the floor, such as a radiator thermostat, then the range may be
reduced by between 50% and 90%.
It is also important to consider the environment close to the antenna. If possible, avoid
placing the node where there are obscuring objects (such as metal pillars, posts or
signs) near to the antenna - a close object obscures a wider range of solid angle.
6.2 Multipath and Fade Margins
Signal Level (about mean) [dB]
It is a fact of life that radio reflections will be present in most 2.4-GHz radio paths.
When the path between two radio antennae is assessed, the transmitted signal can
follow many different paths to arrive at the receive antenna. One path may be direct,
but other paths may involve multiple reflections from walls or metal obstructions.
These are "multiple paths", normally shortened to "multipath". This is a common
experience in analogue TV systems, where it can cause a "ghost" on a television
image. Where there is no line-of-sight, it is multipath (i.e. the scattered signal) that
propagates the radio energy, and here multipath therefore provides a useful service.
However, multipath gives rise to power fluctuations about an average value depending
on location and, in a busy area, time. This is illustrated in Figure 8 below, which shows
the variation in signal strength over time in a laboratory environment with people
moving around. In basic terms, the strong signals regions (or peaks) are present when
the multipath signals arrive in phase and add up - this is known as constructive
interference. In the low signal regions (nulls), the multipath signals arrive out of phase
and cancel each other out - this is known as destructive interference.
10
5
0
-5
-10
-15
0
2
4
6
8
10
12
14
16
18
20
Elapsed Time [s]
Figure 8: Multipath Fading Over Time
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
29
Chapter 6
Deployment Guidelines
Depending on the strength of the direct signal compared with the reflected signal, this
variation can be serious. In a non-LOS situation at 2.45 GHz, it is possible to move the
receive antenna only 3 to 4 cm and see signal levels vary from 6 dB to –20 dB when
compared with the local average. This can be characterised by using a statistical
multipath model, although this is rarely used in practical situations. Instead, a "fade
margin" can be applied. This works by ensuring that the average signal level is well
above the sensitivity of the system, allowing a "margin" of error when "fades" occur.
The required size of this margin depends on the desired level of confidence. In a mesh
network, it can be argued that when many alternative routes are available, moderate
confidence (perhaps 50% of locations are above a certain power level) is acceptable.
When the network is more sparse, the confidence level should be increased (perhaps
to 75% or even 90%). A word of caution here - a law of diminishing returns applies to
confidence levels - with any wireless system, to achieve a confidence level of 99.9%
would probably require extremely low node spacing.
Tip: As a typical guide, when nodes are installed, there
should be about 20 dB path loss margin allowed to take
fades into account and to give an acceptable level of
confidence.
6.3 Open Space Guidelines
For an open space, a typical installation would be the provision of nodes along a
perimeter fence or across a car park. In radio propagation terms, these situations are
line-of-sight with little or no obstructions, and therefore do not suffer largely from
multipath effects. In these situations, it is acceptable to neglect the traditional fade
margin.
Using a free-space approximation is a good starting point for such situations - a range
of 200-300 metres should be achievable, if each node is 2 metres or higher above the
local ground. A high-power module will achieve a much larger range, but when ranges
are above approximately 500 metres a point-to-point link approximation must be used.
Tip: Another word of caution here - the height at which a
node is installed is crucial to using free-space rules. In
short, if the free-space approximation is applied, the
nodes at transmit and receive ends must be elevated
above local obstructions. This is especially important as
the required range increases.
30
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
6.4 Indoor Environment Guidelines
Following on from Chapter 5, it is certain that when a link is set up inside a building it
will exhibit a shorter range than suggested by the free-space guidelines, even when a
line-of-sight is present. In cluttered environments, the range reduction can be very
significant.
There are many different types of environment for any model that is created to cover
indoor radio propagation. However, this document has used a generalised and
modified version of the models. For a guide to node spacings at different levels of
confidence, refer to the publication "The Mobile Radio Propagation Channel" by David
Parsons (Pentech Press 1992, ISBN0-7273-1316-9). Figure 9 plots node-to-node
spacing versus confidence level for situations where there is no directional antenna
and no line-of-sight present. The plot uses typical values for the JN51xx high-power
and standard modules. Furthermore, in Figure 9 scenarios are considered for
traversing zero, one and two floors.
The node-to-node spacings displayed in Figure 9 are very approximate. However,
they do provide some guidelines for pre-planning an installation, if time permits. If this
is not feasible, a simple approach of pacing out distances between nodes will bring
rewards in terms of mesh reliability (for information on mesh networks, refer to
Chapter 7).
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
31
Chapter 6
Deployment Guidelines
160.0
140.0
120.0
100.0
Indoor
Node
to
80.0
Node
Spacing
(metres)
60.0
40.0
20.0
0.0
60%
65%
70%
75%
80%
85%
90%
95%
100%
Confidence Level
High Power Module Traversing 0 Floors
High Power Module Traversing 1 Floors
Low Power Module Traversing 0 Floors
High Power Module Traversing 2 Floors
Low Power Module Traversing 1 Floors
Figure 9: Node-to-Node Spacing for Various Modules and Confidence Levels
32
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
6.5 Point-to-Point Link Guidelines
A point-to-point (or fixed) link is, by definition, an engineered radio link. It is designed
to link two specific points and is always raised above or placed away from local
obstructions. Examples in everyday use include a DBS satellite dish - in this case, the
dish antenna is carefully aligned with a geostationary satellite found at a specific
bearing.
Antennas are elevated
clear of local
obstructions
Figure 10: Point-to-Point Link
In a wireless network, it is reasonable to assume that the two radio nodes of a pointto-point link will be elevated on a roof or, perhaps, high up on a wall. The nodes will
also have directional antennae with a line-of-sight between them, even though they
may be separated by several kilometres. In this situation, the main concerns will be
any buildings or other large obstructions in the path between the directional antennae,
and the height of the antennae above the ground. Fixed-link planners spend a lot of
time working through these considerations. However, for links that have a range of
less than approximately 10 km, it is possible to obtain some guidance from what is
called the "plane-earth propagation" model, with Egli's empirical terrain factor - refer
to "The Mobile Radio Propagation Channel" by David Parsons (Pentech Press 1992,
ISBN0-7273-1316-9). The assumption here is that the height of the antennae (while
above local obstructions) is much less than the distance between the antennae.
Tip: To achieve optimum range between two nodes in a
wireless network, the nodes should be located as high
as possible off the ground/floor.
Tip: High-gain antennae may not be permitted under
some local radio regulations and guidance should be
sought before using them with ISM band equipment.
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
33
Chapter 6
Deployment Guidelines
34
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
7. The Benefits of Mesh Networking
The previous chapters have given a broad overview of radio links at 2.4 GHz. One key
factor that underpins any link between transmit and receive antennae is that its
performance can vary widely depending on a huge number of factors. There is no
guarantee that a radio link will always be possible. However, to increase the level of
confidence that a link is available, the required level of radio performance increases
rapidly too. In reality, the radio must either transmit more power, using a larger
transmitter or transmit antenna, or exhibit a higher receive sensitivity.
A powerful way of ensuring radio communication between two nodes is provided by
the mesh network topology supported by ZigBee. In a mesh network, each router node
can communicate directly (single hop) with any other node within radio range. This
opens up a variety of multi-hop routes to remote nodes with which communication is
not directly possible. Thus, direct radio contact with the target node is not an issue in
a mesh network, and neither is the availability of any particular multi-hop route, since
an alternative route can normally be found. A simple mesh network is illustrated in
Figure 11 below.
Figure 11: Simple Mesh Network
For further information on Mesh networks, refer to the ZigBee PRO Stack User Guide
(JN-UG-3048).
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
35
Chapter 7
The Benefits of Mesh Networking
36
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
8. Sharing Channels with Other Radio Systems
As mentioned earlier, the ISM band at 2.4 GHz is used by a variety of radio system
types. To illustrate this issue, Figure 12 below shows the situation with respect to the
now prevalent WLAN (Wireless Local Area Network) systems.
1
Channel
6
11
2412
2437
2462
US WLAN
(IEEE 802.11b)
non-overlapping
2400MH
z
2483.5MHz
22MHz
1
Channel
7
13
2412
2442
2472
European WLAN
(IEEE 802.11b)
non-overlapping
2400MH
z
2483.5MHz
2MHz
11
12
13
14
15
16
17
2405
2410
2415
2420
2425
2430
2435
Channel
18
19
20
21
22
23
24
25
26
2450
2455
2460
2465
2470
2475
2480
IEEE 802.15.4
2400 MHz PHY
2400MH
z
Key
2440
2445
Channel does not overlap with WLAN in USA
Channel does not overlap with WLAN in Europe
2483.5MHz
Figure 12: ISM Band Sharing
Any fixed channel system, such as WLAN, that uses the same radio channel as an
IEEE 802.15.4-based network can seriously disrupt the network when handling large
amounts of data. However, in most applications, it is possible to arrange that the
WLAN and IEEE 802.15.4-based systems use different channels. As can be seen in
Figure 12, it is usually possible to find channels that are not used by WLAN systems.
To avoid interference, a channel scanning algorithm can be used in an IEEE 802.15.4based system to ensure that the best channel is chosen.
Another widely used system that shares the ISM band at 2.4 GHz is Bluetooth. Used
mainly for headsets and some peripheral connections, the Bluetooth system rapidly
hops across most of the 2.4 GHz band. This may disrupt an IEEE 802.15.4-based
network, but degradation in performance would be gradual.
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
37
Chapter 8
Sharing Channels with Other Radio Systems
Microwave ovens are present in many IEEE 802.15.4 operating areas and do exhibit
low levels of leakage that can disrupt IEEE 802.15.4-based (or any ISM 2.4 GHz)
equipment. However, the duty cycle of microwave cavity ovens is such that there is
plenty of time between cooking cycles for IEEE 802.15.4 packets to be transmitted
successfully.
As a general rule, for any installation, it is always worth assessing the other systems
that may be installed nearby. When transmitter and receiver are close together, they
couple energy very well indeed. The same rule also applies to any interfering systems.
Therefore, if the situation allows, it is worth locating nodes as far away as possible
from items such as Wi-Fi routers and microwave ovens. A distance of several metres
may suffice, but tens of metres is preferable.
Note that international radio regulations which govern radio equipment ensure that
cellphones use their allocated bands only. It is therefore highly unlikely that cellphones
will interfere with IEEE 802.15.4-based equipment.
Tip: NXP provide tools to help assess the operating
environment of a wireless network. The Site Survey Tool
can be used to detect interference - see the Application
Note “Site Survey Tool for JN516x” (JN-AN-1176). The
PER Tester can be used to assess the quality of radio
links - see the Application Note “Packet Error Rate
Testing for JN516x” (JN-AN-1175).
38
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
9. Deployment Checklist
The following checklist provides a summary of factors from an RF view-point that you
need to consider when designing/installing a wireless network.
 Choose an antenna type which is appropriate for your network:

If a node needs to receive and transmit signals in various directions, use
an antenna with low directivity - NXP recommend a half-wave dipole
antenna.

If a node needs to receive and transmits signals only in a specific direction,
use an antenna with high directivity, such as a Yagi antenna - this solution
is suitable for point-to-point links which require a large range.
Refer to Section 4.2 for details.
 Ensure that the gains, transmit power and receive sensitivity of your chosen
antennae are sufficient to achieve the required range - refer to Section 4.3.
 Be sure to orientate the antennae such that their dominant polarisations are
aligned (particularly for point-to-point links) - see Section 4.4.
 Where possible, try to ensure an uninterrupted line-of-sight between nodes - be
aware that the following will reduce signal strength and quality:

Obstructions (such as furniture and people) absorb, reflect and diffract
radio waves.

Walls, floors and ceilings absorb and reflect radio waves - the degree of
absorption depends on the thickness, structure and construction materials.
Refer to Chapter 5 for details.
 If there is no line-of-sight between nodes, be aware that multipath radio
propagation (through reflection and diffraction) is vital to achieve radio
communication - you may be able to help produce a good multipath signal by
positioning nodes such that their signals reflect off a plane surface.
 When siting nodes, the following factors are useful to bear in mind:

It is normally beneficial to place nodes as high as possible (unless there
are obstructions to be avoided, such as support beams and ceiling lights).

If obstructions are unavoidable, do not place a node near to an obscuring
object, since a close object obscures a larger solid angle.
 You can check the quality of radio communications in an installed network
using a tool such as NXP’s Packet Error Rate (PER) tester - refer to the
Application Note “Packet Error Rate Testing for JN516x” (JN-AN-1175).
 If radio communications between some nodes prove problematic, consider
implementing the network in a Mesh topology (supported by the ZigBee PRO
protocol) - refer to Chapter 7 for further information.
 Wireless networks based on IEEE 802.15.4 can automatically select the
frequency channel with least detected activity when sampled at system startup. If you wish to use a pre-configured channel, you should first investigate the
potential interference present in the given operating environment using a
suitable tool such as NXP’s Site Survey Tool - refer to the Application Note “Site
Survey Tool for JN516x” (JN-AN-1176).
JN-AN-1059 v1.2
© NXP Laboratories UK 2013
39
Chapter 9
Deployment Checklist
40
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
Wireless Network Deployment Guidelines
Application Note
Revision History
Version
JN-AN-1059 v1.2
Date
Comments
1.0
23-July-2007
First release
1.1
12-Feb-2009
ISM band sharing diagram modified
1.2
18-Dec-2013
JN51xx devices and supported protocols updated
© NXP Laboratories UK 2013
41
Wireless Network Deployment Guidelines
Application Note
Important Notice
Limited warranty and liability - Information in this document is believed to be accurate and reliable. However, NXP
Semiconductors does not give any representations or warranties, expressed or implied, as to the accuracy or
completeness of such information and shall have no liability for the consequences of use of such information. NXP
Semiconductors takes no responsibility for the content in this document if provided by an information source outside
of NXP Semiconductors.
In no event shall NXP Semiconductors be liable for any indirect, incidental, punitive, special or consequential damages
(including - without limitation - lost profits, lost savings, business interruption, costs related to the removal or
replacement of any products or rework charges) whether or not such damages are based on tort (including
negligence), warranty, breach of contract or any other legal theory.
Notwithstanding any damages that customer might incur for any reason whatsoever, NXP Semiconductors' aggregate
and cumulative liability towards customer for the products described herein shall be limited in accordance with the
Terms and conditions of commercial sale of NXP Semiconductors.
Right to make changes - NXP Semiconductors reserves the right to make changes to information published in this
document, including without limitation specifications and product descriptions, at any time and without notice. This
document supersedes and replaces all information supplied prior to the publication hereof.
Suitability for use - NXP Semiconductors products are not designed, authorized or warranted to be suitable for use
in life support, life-critical or safety-critical systems or equipment, nor in applications where failure or malfunction of an
NXP Semiconductors product can reasonably be expected to result in personal injury, death or severe property or
environmental damage. NXP Semiconductors and its suppliers accept no liability for inclusion and/or use of NXP
Semiconductors products in such equipment or applications and therefore such inclusion and/or use is at the
customer's own risk.
Applications - Applications that are described herein for any of these products are for illustrative purposes only. NXP
Semiconductors makes no representation or warranty that such applications will be suitable for the specified use
without further testing or modification.
Customers are responsible for the design and operation of their applications and products using NXP Semiconductors
products, and NXP Semiconductors accepts no liability for any assistance with applications or customer product
design. It is customer's sole responsibility to determine whether the NXP Semiconductors product is suitable and fit
for the customer's applications and products planned, as well as for the planned application and use of customer's
third party customer(s). Customers should provide appropriate design and operating safeguards to minimize the risks
associated with their applications and products.
NXP Semiconductors does not accept any liability related to any default, damage, costs or problem which is based on
any weakness or default in the customer's applications or products, or the application or use by customer's third party
customer(s). Customer is responsible for doing all necessary testing for the customer's applications and products
using NXP Semiconductors products in order to avoid a default of the applications and the products or of the
application or use by customer's third party customer(s). NXP does not accept any liability in this respect.
Export control - This document as well as the item(s) described herein may be subject to export control regulations.
Export might require a prior authorization from competent authorities.
NXP Laboratories UK Ltd
(Formerly Jennic Ltd)
Furnival Street
Sheffield
S1 4QT
United Kingdom
Tel: +44 (0)114 281 2655
Fax: +44 (0)114 281 2951
For the contact details of your local NXP office or distributor, refer to:
www.nxp.com
For online support resources, visit the Wireless Connectivity TechZone:
www.nxp.com/techzones/wireless-connectivity
42
© NXP Laboratories UK 2013
JN-AN-1059 v1.2
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

advertising