I've discussed DTx (distributed transmission systems) in this column before, starting with a two-part introduction in February 2003. (You can read past RF Technology columns on the TV Technology Web site-- www.tvtechnology.com --in the "Doug Lung on RF" section.)
This month, I'll examine the FCC's rulemaking on DTx released Nov. 4; two papers presented at the IEEE Broadcast Technology symposium in October; and results of tests on Samsung's new Gemini VSB chipset at CRC Canada.
FCC PROPOSED DTx RULES
In the Clarification Order and Notice of Proposed Rulemaking in Media Bureau Docket 05-312 released Nov. 4, the FCC clarified interim rules for DTx and outlined proposed rules for future DTx operations. (See my RF Report for Nov. 8, 2005 for information on the clarification part of the FCC release.) I'll focus on the proposed rules here.
One of problems regulators face in dealing with DTx is how to define the allowed service area for DTx facilities. In the NPRM, the FCC proposes limiting a station's DTx coverage to an area comparable to what the station would be authorized to serve with a single transmitter. To simplify determination of allowable DTV service area, the FCC proposed a table (see Table 1) showing the allowable coverage area as a radius from the reference point based on the station's Form 381 certification. The radius varies depending on DTV channel and the zone in which the reference point is located.
The FCC realized that in some areas with irregular terrain, the station's authorized coverage area may not be circular--less in one direction and extending beyond the Table's radius in other directions. In such cases, the FCC proposes allowing stations to use DTx within their authorized coverage area.
In the NPRM, the FCC cautions, "No station is automatically entitled to use the areas described by the parameters set forth in this chart to provide DTx. Rather, DTx stations, like single-transmitter stations, can apply to use these areas to request authorization to maximize after the freeze is lifted."
The commission also asked for comments on the usefulness of the Table and the validity of the underlying assumptions, and sought comment on alternative ways to determine service areas for stations using DTx.
The FCC proposed DTx transmitters be subject to the same power, antenna height and emission mask limits as single full-power DTV transmitters. The FCC asked for comments on whether it should calculate interference based on each DTx transmitter individually, as proposed by the Merrill Weiss Group, or on the combined signals of all the DTx transmitters. Although there is an ATSC standard for DTx--Standard A/110A, "Synchronization Standard for Distributed Transmission"--the FCC proposes not to mandate compliance with this or any other synchronization standard. By not mandating a synchronization standard, adjustments to the existing standard can be made without requiring modification of the FCC rules.
Comments on this NPRM may be filed up to 60 days after publication of the NPRM in the Federal Register. The deadline for reply comments is 90 days after publication.
One session at the 2005 IEEE Broadcast Technology Symposium was devoted to DTx. It would take several RF Technology columns to cover all the topics discussed at the symposium. This month I'll summarize two papers.
Adjacent-channel DTV stations are a factor in many markets. Due to the susceptibility of DTV tuners to interference from strong adjacent channels, this can pose a problem for stations wanting to use DTx where a station on an adjacent channel is using a single transmitter. Merrill Weiss offered some possible solutions to this problem. One is to place a uniform field strength over the coverage area by controlling the elevation pattern of the distributed transmitter.
Interference to a single-stick station is most obvious near the edge of its contour. However, the FCC desired-to-undesired ratios for adjacent channel stations allow the undesired adjacent channel DTV station to be 26 to 28 dB stronger than the desired DTV station. A UHF DTx facility delivering a uniform field strength that never exceeds the noise-limited threshold signal level by more than 26 to 28 dB (depending on whether the undesired signal is lower or upper adjacent), will never interfere with the adjacent channel under the FCC's interference criteria. Merrill Weiss pointed out that even after allowing a 3 dB margin, each distributed transmitter can provide a field strength as high as 64 to 66 dBµV/m, assuming a 41 dBµV/m noise-limited threshold.
Dr. Oded Bendov, in his paper "Areas of Co-channel Interference and Multipath Created by 8-VSB Modulated Distributed Transmitters in Flat Terrain" (available in the symposium proceedings) analyzed the co-channel interference between a high-power, high-height transmission site and a low-power, low-height site near the edge of the service contour of the high-power station. By using contours rather than a terrain-sensitive model such as Longley-Rice, he was able to easily define the maximum coverage area.
He first defined the area where the contours predicted the difference in the amplitude of the two signals would be 21 dB or more, and where co-channel interference would not occur under any timing difference between the two stations. Next, examining the area where the signal from a high-power, high-height site was within 21 dB of the low-power, low-height site, he looked at the equalizer window to determine the area where the receiver's equalizer could correct for the second signal. A fifth-generation 8-VSB receiver has an equalizer range of approximately +/-50 µs. Since the speed of light is 0.3 km per µs, the equalizer window at its narrowest point will be 15 km, as Dr. Bendovs explained in his paper. By adjusting the timing on the distributed transmitter, this window can be adjusted to put the area of interference where there are the fewest receivers.
Dr. Bendov concluded that in one case he studied, he was able to obtain a 41 dBµV/m coverage area extending up to eight miles from a secondary distributed transmitter with an ERP less than or equal to 50 watts and a height of less than or equal to 100 feet. After this analysis, his client decided to use a translator on a different channel to provide a stronger signal to a community at the end of its contour. The major problem is the inability of even the fifth-generation 8-VSB receivers to handle echoes more than 50 µs apart. One possible solution is to include a broadband frequency domain equalizer in the receiver in addition to the more limited time domain equalizer. The time domain equalizer would handle near echoes while the frequency domain equalizer would handle the impulse response for more distant echoes.
Check Dr. Bendov's Web site, www.tvantenna.tv His paper from the symposium may be available there by the time you read this. I'll have more on the IEEE Broadcast Technology symposium in future columns.
GEMENI CHIPSET TESTED
Results from a laboratory evaluation of a prototype receiver using Samsung's new Gemini chip and consumer grade tuners (ALPS or Thomson) are outlined in an Aug. 30, 2005 report by the CRC (Communications Research Centre) in Canada.
The CRC report states, "This Samsung ATSC 8-VSB receiver, based on Samsung Gemini chip and using an ALPS tuner, is the best DTV receiver that CRC ever tested up to now. It has very impressive multipath and C/N performances."
I was encouraged to see that the receiver easily met the ATSC A/74 specification for co-channel and adjacent channel interference. Receiver interference rejection capability will certainly be tested in the crowded TV spectrum when channels 52-69 go away, The Samsung receiver worked with minimum signal levels down to -84.7 dBm, beating the A/74 recommendation of -83.0 dBm and maximum signal levels more than +1.5 dBm, well above the A/74 recommendation of -8.0 dBm. As +1.5 dBm was the highest level available from the test bed, the receiver may work with even higher levels! The interference rejection results were interesting. For moderate signal strengths, the Samsung receiver worked with D/U co-channel interference ratios as low as 14.7 dB, which is surprising because if the interference were pure Gaussian noise, a ratio higher than 14.9 dB would have been required.
Adjacent channel interference rejection was impressive, with -37.1 dB lower adjacent and -35.7 dB upper adjacent D/U ratios required, beating the A/74 recommendation of -33 dB. Note that with a -37.1 dB D/U ratio, the undesired adjacent channel could be 5,000 times stronger that the desired channel! If this performance is duplicated in consumer receivers, it will bode well for LPTV DTV stations and DTx facilities that have to contend with interference from high-power adjacent channel stations.
You can download the CRC report from: http://www.crc.ca/en/html/crc/home/research/broadcast/rtnt.
Questions and comments are always welcome and appreciated. E-mail me at dlung@transmitter.com.
