pDVB ASI Monitoring with the WFM601i/E/M
November ‘97
Monitoring of DVB ASI signals with the WFM601i/E/M
MPEG–2 Transport Streams within DVB systems are increasingly being delivered over the Asynchronous Serial Interface (ASI). In brief, this takes the MPEG-2 Transport Stream, converts the 8 bit words into 10 bit words, and enters these onto a 270Mbit/sec. carrier. Synchronising bytes are inserted as stuffing to fill the 270Mbit/sec. stream.
Within the Broadcast industry, 270Mbit/sec. streams are increasingly evident on the Serial Digital Interface (SDI) for full bit rate video, however the ASI and SDI signals have very little in common.
ASI signals and SDI signals share the same launch amplitude of 800mV, and the same bit rate of 270Mbit/sec, but here the similarity ends. ASI signals are polarity dependent (SDI signals are not), and ASI receivers do not contain automatic equalisers. Clearly, the data content is also totally different, as is the energy dispersal system, such that the power spectrum is more data dependent than with SDI signals.
Equipment such as the Tektronix MTS200 series enables protocol testing of the data contained within the MPEG-2 Transport Stream, but doesn’t enable testing of the ASI physical layer
The following describes the way in which the WFM601i, WFM601E and WFM601M may be used to do basic testing of the ASI physical layer. While the tests are not complete, they enable measurement of launch amplitude, rise and fall time, and jitter, and are therefore felt to be of value.
Setting the WFM601E/M for DVB ASI measurements (Please note that the WFM601i does not require any adjustments)
DVB ASI tests require two very simple adjustments to the WFM601E/M. These are detailed below:
1. Jumper J13 should be moved from pins 2-3 (normal operation) to pins 1-2. The location of the jumper is shown in Figure 1.
2. Set the WFM601E/M to eye diagram display for the A input.
3. Set the eye display to 10 eye mode, 100Hz. Bandwidth.
4. Connect an ASI signal to SDI input A.
5. Adjust potentiometer R269 (located as shown in Figure 1) for a stable eye diagram display as shown in Figure 2. R269 and Jumper J13 are located at the front of the board on the RHS of the instrument, as shown in Figure 1.
It is recommended that a 525/625 line black burst reference is connected to the external reference of the WFM601E/M and that external reference is selected. (Note that the applied black burst reference may be free running, and does not need to be locked to either the 270MBit/sec. ASI stream, or any of the videos contained within the Transport Stream. It is simply used to maintain horizontal sweep calibration of the WFM601E/M display.)
Using the WFM601i/E/M for DVB ASI measurements
DVB ASI signals may be applied to the A and B inputs of the WFM601i/E/M and displayed as eye diagrams.
Figure 2. shows a 10 eye display from an ASI source. The WFM601i/E/M triggers it’s display at one tenth the rate of the 270MBit/second data, so the first, second, third bits etc. in the 10 bit words are always overlaid. The phase of the 10 eye display will be arbitrary but the pattern should be locked. At a Transport Stream rate of 40Mbit/sec., the useful data within the ASI stream is 50Mbits/second, (due to the 8-10 bit conversion used when creating an ASI stream), leaving 220Mbits/second of stuffing, or sychronising bytes.
The predominance of stuffing bytes may be seen in Figure 2 as a bright pattern in the eye diagram. The stuffing byte is K28,5 and has two 10 bit complementary binary patterns, 00 1111 1010 and 11 0000 0101.
Figure 2 shows an eye display that may be expected at the launching point, i.e. at the output of a device. Note that the cursors may be used to measure the launch amplitude, in this case 805mV. Note also that this display removes word correlated jitter.
Compare this with Figure 3, which shows an ASI signal at the end of approximately 30 metres of cable. The effects of cable length on the eye diagram can be clearly seen.
Expanding the eye display to a 3 eye overlay mode as in Figure 4 enables use of the cursors for measurement of jitter. In this example jitter is approximately 550ps p-p..
The standard filters contained within the WFM601i/E/M enable measurement of jitter directly from the eye display with high pass filters of 10 Hz, 100 Hz and 1kHz. (The PLL used to trigger the display in the WFM601i/E/M tracks the jitter at frequencies below the filter cut off point).
An alternative display, shown in Figure 5, is incorporated in the WFM601M. It shows a waveform representing demodulated jitter against time, and is essential for jitter values in excess of 1 UI. In this example, the display is at line rate, i.e. 10micro sec./div., but could be changed to a field rate display. Note that two values of jitter are provided. The first of these is timing jitter, in our example measured at 0.23 UI, representing the total jitter at frequencies above 10Hz. A second measurement is provided for alignment jitter. In our example this is 0.19 UI at frequencies above 1kHz. HighPass filtering may be selected as 10Hz, 1kHz, 10kHz and 100kHz.
DVB recommendations for ASI Interface
The following are extracted from DVB TM1449 “Interfaces for CATV/SMATV Headends and similar Professional Equipment”
The ASI clock is specified to be 270Mbit/sec +/- 100ppm.
Transmitter Output Characteristics / UnitsOutput voltage / mV (p-p) / 800 10%
Deterministic Jitter / % (p-p) / 10
Random Jitter / % (p-p) / 8
Rise/Fall Time (20-80%) / ns (max) / 1,2
Receiver Input Characteristics / Units
Min Sensitivity (D21.5 idle pattern) / MV / 200
Max Input Voltage / mV(p-p) / 880
s11 (range: 0,1 to 1,0 x bit rate) / DB / -17
Min Discrete Connector Return Loss
(5 MHz - 270 MHz) / DB / - 15
Recommended eye display for the launched signal is as below:
Limitations and a word of Caution.
The above application of the WFM601i, WFM601E and WFM601M is of course outside their original intended usage. While operation with DVB ASI signals has been checked, to a limited extent, within the development laboratories, it is offered without any guarantees of measurement accuracy outside those specified for the standard instrument.
It is believed that measurements of jitter, eye amplitude and eye timing are valid, but full validation has not been carried out.
Cable length measurements will not meet the published specification due to the differences in spectral content between SDI and DVB ASI signals.
We would ask that if WFM601E/M are modified in this way, the jumper is returned to its normal operating position prior to returning for service or calibration.
Normal operation of the WFM601E/M for 270Mbit/sec SDI should not be affected. However, if this should be the case, normal operation will be obtained by replacing jumper J13 in its normal operating position.
We hope that the above enables use of existing WFM601i/E/M’s for a new application, and would be pleased to receive any comments on measurements of DVB ASI on the WFM601i/E/M.
Please send any comments to Lionel Durant via Email at the following Email address -
For further information, contact Tektronix:
World Wide Web: Austria +43 (1) 701 77-261; Belgium +32 (2) 725-96-10; Denmark +45 (44) 850700; Finland +358 (9) 4783 400; France +33 (1) 69 86 81 08; Germany +49 (221) 94 77-400; Italy 39 (2) 250861; The Netherlands +31 23 56 95555; Norway +47 (22) 070700; Spain & Portugal +34 (1) 372 6000; Sweden +46 (8) 629 6500; Switzerland +41 (41) 7119192; United Kingdom & Eire +44 (1628) 403300
1 p Nov. ‘97