null null

Products: Test Broadcast System R&S® SFU
Measuring Bit Error Rate
using the R&S® SFU-K60 Option
Application Note
The measurement of bit error rate (BER) is an important test procedure for determining
the performance of a receiving terminal. This is especially important for set top box and
chipset manufacturers who need to know if their product is able to meet the specification
under a predefined noise condition. This Application Note describes the technique using
an Rohde & Schwarz SFU to measure the BER of digital set top boxes.
Subject to Change – C K Tan 12.2006 – 7BM51_1E
Contents
1 Overview ......................................................................................2
2 Measuring the BER ....................................................................3
2.1 Serial Stream Measurement ..............................................5
2.2 MPEG-2 TS Measurement.................................................5
3 Measurement Setup ...................................................................7
4 Special Cases .............................................................................9
5 Summary................................................................................... 10
6 References ............................................................................... 10
7 Additional Information………………………………………..11
7 Ordering Information ............................................................... 11
1 Overview
Measuring the BER of a set top box requires a test instrument to
inject a predefined test signal, such as a PRBS, and comparing
the output signal from the set top box with that test signal. The
output signal may be extracted at any point in the signal flow of
the set top box. In order to capture the real effects of the noise
on the transmission data, the module responsible for forward
error correction (FEC) is disabled. Otherwise, all bit errors are
either fully corrected or reduced. This is usually carried out by
the Reed Solomon encoder/decoder. In most cases, the signal is
extracted before or after the Reed Solomon(RS) decoder. If the
signal is taken after the RS decoder, the decoder is disabled so
that errors under measurement in the signal is not corrected.
Below drawing shows the basic principle of BER measurement.
Figure 1 The basic principles of BER measurement on set top
boxes. Legend: CI=Common Interface, PRBS=Pseudo-Random Binary
Sequence.
Here the R&S® SFU can inject a PRBS stream or a stream of
MPEG-2[1] packets containing a predefined payload. The BER
tester (BERT) can pick up the output signal from the set top box
2
at 2 locations (BERT1, BERT2). BERT2 will capture the MPEG-2
packets after the RS decoder and BERT1 will extract the PRBS
stream before the RS decoder.
The output signal captured by BERT1 will be compared with the
PRBS signal that was injected in the R&S® SFU just after the RS
encoder. On the other hand, the output signal captured by
BERT2 will be compared with the MPEG-2 packets that were
injected in the R&S® SFU just before the RS encoder.
The choice of the 2 BER tests is left to the user to decide.
In Digital Video Broadcasting (DVB), a BER of 2 x 10-4 before
the RS decoder is specified that will produce a quasi-error free
(QEF) reception at the receiver. The basic idea is to add noise,
preferably AWGN (requires R&S® SFU-K40 installed in the
R&S® SFU), and measure the BER at different noise levels (C/N
settings on the noise generator). Thus, by comparing the output
signal in the receiver just before the RS decoder with the
transmitted signal, it is possible to obtain a set of deviation data
to determine the equivalent noise degradation (END), which is
an important receiver parameter[2].
This application note introduces the measurement option of BER
using the R&S SFU Broadcast Test System which is capable of
generating a set of standard test pattern sequence.
2 Measuring the BER
The aim of the R&S® SFU-K60 Option is to provide a convenient
way to measure the BER performance of a digital receiving
terminal. The R&S® SFU BER application is conveniently
located on the front panel display and Figure 2 shows the BER
window.
3
Figure 2 The BER window of the R&S® SFU
This option is completely independent from other applications
such as Transmitter TX and TS Player/Recorder, which means
that it can use as an independent BERT. It can be integrated
with the other options such as noise and fading to provide a
realistic test environment.
The selection of items in this BER window is shown in Table 1.
Menu Item
MEASUREMENT
Value
ON/OFF
INPUT
BER
SERIAL CLK DATA EN
ASI FRONT
ASI REAR
SPI FRONT
SPI REAR
NORMAL
INVERTED
NORMAL
INVERTED
ALWAYS
ACTIVE HIGH
ACTIVE LOW
e.g. 0.00E-7
EVALUATION
STATE
e.g. 6.3E9 / 1E10
SYNC
ERROR COUNT
e.g. 20
GATING TIME
e.g. 00:15:30
GATING MODE
INFINITE/AUTOMATIC
CLOCK
DATA
ENABLE
Remarks
To turn on/off BER
measurement
The location of the
connection on the R&S
SFU. The signal coming
from
the
DUT
is
connected to this.
Depending
on
the
measured DUT
Depending
on
the
measured DUT
Depending
on
the
measured DUT
Display of the measured
bit error rate
Estimated accuracy
BER measurement is
synchronized
to
the
PRBS
Total number of bit errors
detected
Running time of the
measurement (hh:mm:ss)
Set
the
mode
of
measurement
Table 1 The menu items of the BER measurement on the R&S®
SFU. Please refer to the R&S® SFU Manual for more details.
In Table 1, a unique feature of R&S® SFU-K60 which is worth
mentioning is the gating mode. For any BER measurement, the
longer the measurement (gating) time, the better the confidence
level of the BER reading. At low error rate, if the gating time is
set too short, fewer bit errors will be caught and this gives rise to
a sub-optimal reading . If you are unsure of the BER of the
channel, you can set “AUTOMATIC” for this option which
automatically allows a sufficient amount of gating time for errors
to be captured. Hence, the gating time for a lower error rate will
be longer so that an adequate number of errors can be captured
and vice-versa. The gating time is calculated based on the
reception of 1000 bit errors and the data rate. This feature will
4
ensure a more accurate reading irrespective of the noise
condition.
There are 2 general test streams that are generated by the
R&S® SFU for the BER measurement.
2.1 Serial stream measurements
The basic serial BER stream can be a PRBS, all ones or all
zeros. The PRBS which is in compliance with ITU-T O.151 is
based on the generator polynomial, 223-1 or 215-1. Figure 2
shows the menu on which the payload of the test stream can be
selected. When the input is selected as “SERIAL CLK DATA
EN”, the test signal is a continuous stream of pure PRBS without
any framing structure. The clock and enable signal is taken from
the DUT.
Figure 4 shows the physical connection for this serial test
stream. Apart from the CLK, EN and DATA connectors, the
“BER ERR OUT” interface will output a positive pulse for every
detected bit error.
Figure 4 The interface at the rear of the R&S® SFU for sending
the serial PRBS stream
2.2 MPEG-2 TS measurements
The second test signal is based on the MPEG-2 Transport
Stream standard[1]. It is framed in a 188-byte packet structure.
This test signal is selected when the input is selected as “ASI
FRONT”, “ASI REAR”, “SPI FRONT” or “SPI REAR”. The ASI
5
interface is a 50-ohm female BNC connector and the SPI is a 25pin parallel connector. Figure 5 shows the location of these
interfaces on the R&S® SFU.
Figure 5 The physical interfaces of the BER input for MPEG-2
packets
With this interface, a user can select any of the following 3 types
of test packets to generate.
The first type is a 188-byte packet with the first 4 byte set to a
real TS header. The remaining 184 bytes contain the payload.
The second type is a 188-byte packet with the first byte set to
0x47(hex). The remaining 187 bytes contain the payload.
The last type is a 188-byte packet with a proper 4-byte TS
header but the PID is 0x1FFF. This is the so-called null packets.
To summaries, Table 2 shows the 3 types of packets streams
available for the MPEG-2 TS input. Note that with this test
option, the payload may be PRBS, all zero’s or all one’s.
Selection
Remarks
HEAD 184
PAYLOAD
SYNC 187
PAYLOAD
STUFFING
PACKET
6
Table 2 The 3 type of MPEG-2 TS based packets
The “STUFFING PACKET” can be used when there are other
live MPEG-2 TS streams in the signal. However, a note of
caution is that if any bit in the PID field of the monitored transport
packet is corrupted by noise, the packet will not be evaluated.
This will result in an inaccurate BER measurement.
If PRBS is used in MPEG-2 packets test, the sequence will be
broken into equal segments and filled into the MPEG-2 packets
as payloads before sending out of the R&S® SFU. When the
packets is processed at the return path back to the instrument ,
the PRBS stream will be re-assembled and the associated builtin PRBS counter will pause during the gaps between each
packet. Nevertheless, the BER is calculated and displayed realtime on the BER window.
The type of packets chosen will depend on the test application of
the DUT. Figure 6 shows the types of BER test signals that are
generated by an R&S® SFU-K60 option.
Figure 6 Types of BER test signals available in R&S® SFU-K60
option
3 Measurement Setup
Before a BER measurement is carried out, it is advisable to
know the objective of the test. This will help to determine the
type of packets (e.g . serial stream or MPEG-2 TS) to use in the
test as well as the selected interface on the R&S® SFU. Once
this is known, connect all the required cabling between the
instrument and the DUT.
7
Figure 7 The connection between the R&S® SFU and the DUT.
(Top) Performing a serial BER test (Bottom) Performing a
MPEG-2 TS BER test
Before starting the test, it is a good practice to do a pre-test as
follows:
1. Switch off the noise (requires R&S® SFU-K40 installed in
the R&S® SFU) and fading (requires R&S® SFU-B30
installed in the R&S® SFU).
2. Do a BER measurement after RS with Reed Solomon
ON. The result should be 0 if everything is working well.
Figure 7 The Reed Solomon encoder on the R&S® SFU can
be enabled/disabled.
3. Do a BER measurement before RS with Reed Solomon
OFF. The result should also be 0.
8
4. Switch on the noise and fading (as required in your test)
to see the effect on the BER reading. Figure 8 shows the
BER reading on the BER window.
Figure 8 The real-time BER reading in displayed on the top left
corner of the BER window
With this initial procedure done, proceed to change the C/N of
the noise in steps or change the fading conditions in the channel
simulator, depending on your test requirements and note the
real-time BER reading.
4 Special Cases
There are many new digital transmission standards in the field of
broadcasting.
Currently, the BER measurement option described in this
application note is not available for DVB-S2[3] and DIRECTV[4]
due to the measurement methods used in the R&S® SFU. In
these 2 types of services, it is not possible at the current version
(FW Version 1.30) to insert PRBS at the described points. Also in
DIRECTV, the packets are based on a frame size of 130 bytes
and does not work with an MPEG-2 format of 188 bytes.
9
5 Summary
The R&S® SFU-K60 Option offers great ease and flexibility for a
communications engineer to assess the quality of digital
transmission methods and the components involved. The
combined features of an R&S® SFU to provide a built-in TS
generator, noise generator and an RF output offer a set top box
vendor the ease of testing his products in a short time.
6 References
[1]ISO/IEC 13818-1:”Information technology – Generic coding of
moving pictures and associated audio information”
[2] Franz-Josef Zimmermann: Bit-error-rate measurement on setto boxes, News from Rohde and Schwarz, Number 167 (2000)
[3] ETSI EN 302 307 V1.1.1, Second generation framing
structure, channel coding and modulation systems for
Broadcasting, Interactive Services, News Gathering and other
broadband satellite applications (DVB-S2)
[4] See www.directv.com
7 Additional Information
Our Application Notes are regularly revised and updated. Check for any
changes at http://www.rohde-schwarz.com.
Please send any comments or suggestions about this Application Note to:
10
7 Ordering Information
ROHDE & SCHWARZ GmbH & Co. Mühldorfstraße 15 D-81671 München P.O.B 80 14 69 D-81614 Munchen
Telephone +49 89 4129-0 Fax +49 89 4129 – 13777 Internet: http://www.rohde-schwarz.com
This application note and the supplied programs may only be used subject to the conditions of use set forth in the download area of the
Rohde & Shwarz website.
11

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