There are many fine technical conferences and broadcast shows in early Fall, but the one I always make it point to attend is the IEEE Broadcast Symposium, held in October in Washington D.C. Traditionally the majority of papers at the conference have focused on transmission or reception and this year was no exception.
Several papers examined the planning factors and assumptions underlying the FCC's methods for determining interference and coverage. Three papers focused on the desired to undesired (D/U) ratios used to determine interference between stations.
Dennis Wallace, in his paper Considerations for Constructing a Low Power DTV Facility, looked at the impact of adjacent channel receiver performance on DTV stations. He explained that the FCC planning factors required compromises to meet policy goals and that to have a reasonable approach to spectrum planning, some assumptions must be made.
The FCC planning factors do not consider the impact of strong signals on DTV receiver performance. Wallace suggested different UHF D/U ratios depending on the received signal strength:
|<p><br><br><b><span style="FONT-SIZE: 9.5pt; LETTER-SPACING: -0.15pt; mso-bidi-font-size: 10.0pt"><span style="mso-spacerun: yes"> </span>Condition <span style="mso-spacerun: yes"> </span><o:p></o:p></span></b></p>||<p><br><br><b><span style="FONT-SIZE: 9.5pt; LETTER-SPACING: -0.15pt; mso-bidi-font-size: 10.0pt">Weak<o:p></o:p></span></b></p>||<p><br><br><b><span style="FONT-SIZE: 9.5pt; LETTER-SPACING: -0.15pt; mso-bidi-font-size: 10.0pt">Medium<o:p></o:p></span></b></p>||<p><br><br><b><span style="FONT-SIZE: 9.5pt; LETTER-SPACING: -0.15pt; mso-bidi-font-size: 10.0pt">Strong<o:p></o:p></span></b></p>|
Weak signal level here is defined as 41 dBu, medium at 61 dBu and strong as 69 dBu, with linear interpolation used to determine the ratios for cases between these levels.
Using these ratios, Dennis Wallace looked at the coverage from a channel 35 low power DTV STA in Washington D.C. This channel is between two high power DTV stations on channels 34 and 36. When measured receiver performance data is used, only one square kilometer in the entire coverage area is predicted to get service. Even though the facility meets the FCC coverage requirement based on its contour, adjacent channel interference results in the station losing 99.9 percent of its coverage!
The WHFT channel 46 DTV in Miami was also studied. In this case, there is an adjacent analog channel 45 and an upper adjacent high power DTV on channel 47. WHFT's 3.7 kW channel 46 DTV is predicted to lose coverage to 132,000 people due to analog interference and over 900,000 people due to adjacent channel DTV interference. Using the modified planning factors, the loss from analog interference climbs to 2,300,000 while the loss from the adjacent channel DTV drops to only 407,414.
Dennis Wallace concluded by recommended changing the rules to require proof not only that the station provides adequate field strength over its city of license, but also that the D/U ratios show the city of license actually receives coverage. He also took note of a problem I mentioned in last month's RF Technology column - high power DTV stations in neighboring markets might limit analog coverage. In this case, he recommended using the DTV signal to feed cable companies. Finally, he emphasized more testing of production DTV receivers is urgently needed.
William Meintel with Techware looked at the impact DTV distributed transmission systems will have on service and interference analysis. He noted that it has been almost a decade since the original DTV plan was created and many of the initial planning factors and assumptions have been forgotten. Receiver performance in the presence of moderate and strong signals and in the presence of multiple interfering signals was ignored.
With distributed transmission systems, more receivers will have strong signals, any advantage from co-locating stations is lost, single interfering signal criteria is no longer valid and antenna elevation patterns do not match those in the planning model.
William Meintel also proposed different D/U ratios for different signal strength conditions. For adjacent channel interference, his ratios were considerably stricter than the current FCC rules and those proposed in Dennis Wallace's paper, although the paper did not specify the signal levels used for the three conditions. He also suggested ratios for determining DTV into NTSC interference:
|<p><br><br><b><span style="FONT-SIZE: 9.5pt; LETTER-SPACING: -0.15pt; mso-bidi-font-size: 10.0pt"><span style="mso-spacerun: yes"> </span>Offset<span style="mso-spacerun: yes"> </span><o:p></o:p></span></b></p>||<p><br><br><b><span style="FONT-SIZE: 9.5pt; LETTER-SPACING: -0.15pt; mso-bidi-font-size: 10.0pt">Condition<o:p></o:p></span></b></p>||<p><br><br><b><span style="FONT-SIZE: 9.5pt; LETTER-SPACING: -0.15pt; mso-bidi-font-size: 10.0pt">Weak<o:p></o:p></span></b></p>||<p><br><br><b><span style="FONT-SIZE: 9.5pt; LETTER-SPACING: -0.15pt; mso-bidi-font-size: 10.0pt">Medium<o:p></o:p></span></b></p>||<p><br><br><b><span style="FONT-SIZE: 9.5pt; LETTER-SPACING: -0.15pt; mso-bidi-font-size: 10.0pt">Strong<o:p></o:p></span></b></p>|
These ratios assume the stations are not co-located. If they are, William Meintel notes that the ratios could be relaxed by 10 dB or more.
Last month I mentioned a case of DTV into analog adjacent channel interference where the two signals were calculated to close to equal. In a medium or strong signal environment, these ratios would have predicted that interference.
In the paper, Meintel points out that since it is a well known fact that the Longley-Rice propagation model significantly over predicts signal levels, especially over flat terrain, the threshold for determining medium and strong signal conditions needs to be adjusted to compensate for it. One option would be to use the traditional FCC propagation curves and the height above average terrain to select the appropriate D/U ratio.
Since the paper focused primarily on distributed transmission systems, I'll have more on it next month when I look at the papers devoted to single frequency networks.
Oded Bendov presented a paper he wrote with Yiyan Wu, Charles Rhodes and John F. X. Browne titled Planning Factors for Fixed and Portable DTTV Reception. The paper looks at the entire DTV link budget, from transmitter to DTV receiver, to determine the real signal level needed for successful DTV reception.
Starting at the transmitter, the paper shows the threshold penalty for a system VSWR of 1.16:1 is 0.8 dB, raising the theoretical threshold level of 15.2 dB to 16 dB. This penalty can be reduced by using a two power amplifiers combined in a hybrid such that reflections from the antenna are dumped into the hybrid reject load instead of the power amplifier.
The current planning factors do not consider the impact of multipath and receive antenna mismatch on the required DTV threshold and received signal levels. Higher signal levels are required to overcome these effects and to overcome man-made noise not included in the planning factors. The paper also criticized the use of "field strength" for specifying DTV coverage and interference. Since digital TV signals are wide band and do not have a recognizable carrier, it is more accurate to specify signal strength in terms of received power in dBm.
Recommended minimum power levels for decoding the DTV signals are -70.5 dBm for low VHF, -73.4 dBm for high-VHF and -71.4 dBm for UHF. Adding a low noise amplifier (LNA) at the antenna reduces the levels to -71.4 dBm, -78.9 dBm, and -79 dBm for low VHF, high VHF and UHF, respectively.
Current U.S. DTV coverage is based on a threshold signal being available at 50 percent of the locations 90 percent of the time (F 50,90). The paper showed the additional signal margins needed to bring this percent up to 75 percent of the locations 99 percent of the time, which ETSI uses for "acceptable" DTV service and 95 percent of the locations 99 percent of the time, defined by ETSI as "good". The added signal level margins needed to reach these service levels at UHF are 10.3 dB for "acceptable" and 21.7 dB for "good". Separate calculations, based on BBC research, indicate approximately 37 dB more signal is needed for indoor reception, resulting in a need for a "field strength" of 90-93 dBu at 30 feet above ground.
I've only been able to cover a small portion of Planning Factors for Fixed and Portable DTTV Reception. The paper has formulas and references to back up its conclusions and detailed charts showing how transmitter performance, system VSWR, noise, and other factors impact DTV reception.
Robert Weller, Merrill Weiss and Sean Driscoll reported on New Measurements and Predictions of UHF Television Receiver Local Oscillator Radiation Interference. In the early days of television, radiation from the local oscillator in a TV set's tuner sometimes caused interference to other receivers. The local oscillator frequency is the sum or difference of the frequency of the channel received and the intermediate frequency (IF) of the TV set. In most cases, the oscillator is set to the sum of the channel and the IF (41 MHz). As a result, radiation from the oscillator would cause interference to nearby sets viewing a station 7 channels (42 Mhz) above the station the first set was tuned to.
There was concern that the risk of interference at UHF would be even greater than at VHF, because shielding was less effective. As a result, the FCC adopted a taboo requiring stations proposing operation +/- 7 channels from another station be located at least 60 miles from that station.
To determine whether local oscillator radiation was problem with current generation TV sets, Merrill Weiss Group LLC and Hammett and Edison, Inc. tested 67 TV tuners in TV sets and VCRs. Refer to the paper for details on how the measurements were made. The measurements showed local oscillator radiation has been steadily decreasing since 1980. Data from an EIA study that showed about 46 percent of TV sets are removed from service after 15 years, over 90 percent of TV sets after 21 years and 99.7 percent after 22 years. Using this data, the age distribution of the current population of TV tuners and the average oscillator radiation level today could be calculated. It was found to be -81.8 dBm.
The paper makes some worst case assumptions about how this radiation would impact other receivers - wall attenuation is ignored, the antennas are assumed to be pointed at each other and the desired signal is assumed to be noise free. Under these conditions, with a Grade A desired signal, the interference margin is 19.3 dB, indicating not only that no visible interference from local oscillator radiation is predicted but even in unusual cases would be highly unlikely. For indoor reception, the margin drops but remains above zero, at 3.3 dB, so no interference is predicted.
Based on this study, the paper's authors concluded it is time to eliminate the N+/-7 local oscillator taboo and base interference analysis at this channel separation on IF beat interference, where the two signals could mix to create an intermodulation product that falls in the IF passband. The same rules used for the N+/-8 taboo, 31.4 km spacing between transmitter sites, could be applied in the N+/-7 case, unless a waiver was obtained.
Next month I'll continue my report on the IEEE Broadcast Symposium papers with a look at the latest news on single frequency networks and distributed transmission systems for DTV. For more information on the IEEE Broadcast Technology Society, visit www.ieee.org/bts. Contact the Society for information on obtaining copies of the papers presented the Symposium.
Drop me a note at email@example.com