Earlier this summer, an interesting debate developed around the FCC Notice of Inquiry on "Technical Standards for Satellite-Delivered Network Signals" over how to determine whether a household can receive a local network affiliate's DTV signal over-the-air. Households that can receive a local network station are not allowed to receive a distant network DTV signal via satellite. The debate pitted EchoStar against broadcasters.
Both sides used engineering exhibits prepared by reputable consulting firms to support their arguments. The exhibits present an interesting picture of how difficult (or easy) it is to receive DTV signals. All filings are available from the FCC's Electronic Comment Filing System at http://gullfoss2.fcc.gov/prod/ecfs/comsrch_v2.cgi (search for Docket 05-182).
Hammett and Edison provided the engineering to support EchoStar's request for much tighter standards for over-the-air DTV reception.
I'm sure many broadcasters were surprised to see H&E's name on the exhibits, as many of H& E's clients are broadcasters and one of their senior engineers is on the board of directors for the Society of Broadcast Engineers. The H&E exhibit raised concerns about the applicability of the FCC planning factors and the performance of DTV receivers and set-top boxes.
H&E's statement supported EchoStar's assertions that: the FCC's signal level for reliable DTV reception was too low; the impact of multipath on DTV reception has to be considered; the FCC's reception standards should consider the use of indoor antennas or mispointed outdoor antennas; current DTV receiver performance falls short of the FCC's DTV planning factors; and because DTV reception does not degrade gracefully, the FCC must give more consideration to time variability when in determining DTV reception.
Engineering provided by the firm Meintel, Sgrignoli and Wallace was used by the National Association of Broadcasters to show that equipment used to receive ATSC DTV signals is improving and in some cases exceeds the parameters used in the FCC planning factors.
William Meintel wrote the coverage and analysis software used by the FCC in creating the FCC Table of Allotments and the software the FCC used in processing DTV applications. Gary Sgrignoli is a recognized expert on ATSC transmission and reception. Dennis Wallace is known for his work on DTV field tests.
The MSW statement supports the NAB argument that the planning factors are a reliable indicator of DTV coverage.
It explains that rotors or multiple antennas can be used to ensure antennas are properly oriented. Where it is difficult to meet some planning factor parameters, readily available mast-mounted low-noise preamplifiers can be used to compensate for the additional losses. They also showed that signal strength is a good proxy for determining the ability to receive a DTV picture.
FACTORS
Let's look closer at engineering statements these two firms filed.
MSW spends a large part of its engineering statement describing the planning factors and the methodology behind them. MSW doesn't dispute that indoor antennas are worse than outdoor antennas, but argues that outdoor antennas are the logical choice since satellite antennas also have to be mounted outdoors.
Antenna pointing is not a major problem, since in crowded markets like the Northeast, the signals coming from another direction are likely to duplicate network programming. MSW also notes that while some antenna systems will not meet the gain criteria in the planning factors, this can be overcome by the use of low-noise mast-mounted amplifiers.
The FCC Notice of Inquiry asked whether DTV signal strength should be used to determine if a location was able to receive a DTV signal. MSW says signal strength is an indicator of DTV reception and that this is supported by field tests--15 separate measurement programs across 12 cities conducted between 1994 and 2001. The Grand Alliance blue rack receiver, which is known to have significantly worse equalizer performance than fourth- and fifth-generation receivers, was used in 11 of the 15 field test programs. Two of the tests used the second-generation receiver and the remaining two used third-generation receivers. None of the 15 tests used newer technology.
MSW states the key statistic to look at in these tests is the System Performance Index--the percentage of sites with signal levels above the FCC minimum that had successful DTV reception. For the 15 test programs, the System Performance Index ranged from 76.8 percent for KING in Seattle to 98.7 percent for KICU in San Jose, both in 1998. The average for all the tests was 90 percent. In most cases where reception wasn't possible, MSW said the problem was multipath or interference. Newer receivers offer much better multipath performance.
For analog TV, the FCC allows the use of Longley-Rice to determine whether a house is able to receive a local analog TV signal. The parameters used for these studies are outlined in OET Bulletin 72. Unlike OET Bulletin 69 studies for DTV coverage, OET Bulletin 72 requires the use of finer terrain data, adds additional losses based on U.S. Geological Survey land-use data and adjusts antenna height based on whether the house at the location studied is one story (20 feet) or two (30 feet). MSW found than when the Longley Rice model was used as outlined in OET Bulletin 72, out of the 2,169 locations measured in the DTV field tests, it correctly predicted whether the signal would be above or below the noise-limited threshold at 2,047 or 94.4 percent, of the locations.
In support of EchoStar's comments, H&E's engineering statement says that its experience shows only a small percentage of households--"perhaps 10 to 15 percent"--with outdoor antennas also utilize a rotor. H&E took a random sample of 4.4 million calculation points covering the continental United States and found that the majority of these sites were expected to receive at least two NTSC signals at Grade B or better. However, at the majority of the sites predicted to receive signals from two or more stations, at least one of the stations is at an angle 25 degrees or more from another station, meaning that without an antenna rotor, reception of at least one station would be degraded by as much as 10 dB (low VHF), 12 db (high VHF) or 14 dB (UHF).
H&E's statement outlines the additional losses encountered with indoor reception, but does not discuss whether indoor antennas should be used for determining whether a household receives a DTV signal. Recognizing that many viewers are likely to use indoor antennas to receive DTV, this information is useful even outside of the satellite distant signal debate.
H&E notes that a PBS study found indoor antennas had, on average, a gain of -1.1 dB, in other words, a loss. This is about 10 dB below the gain assumed for outdoor antennas. A 1979 ITS study found even greater loss, with averages ranging from -2.8 dBd (high VHF) to -4.4 dBd (low VHF). The most recent study of indoor antennas was by done by Kerry Cozad at Dielectric. H&E says this study shows an average gain of 2.4 dBd for indoor antennas, 9.2 dB below the average gain of the measured outdoor antennas.
Of course, antenna gain isn't the only factor to consider. H&E point to a 1963 FCC study that found building penetration loss in less cluttered areas of NYC (outside Manhattan) were about 25 dB at VHF and 21 dB at UHF. In the most cluttered areas, UHF loss increased to 26 dB.
The H&E statement also includes data from a United Kingdom study comparing building penetration loss with the height of the building. In this study, UHF losses ranged from 16.4 dB at ground level to between 2.5 and 4.2 dB at the sixth floor.
In attachments to EchoStar's comments and reply comments, H&E addressed the issue of DTV receiver performance. They compared the sensitivity of six receivers--five consumer model receivers (four purchased in May 2005) and one professional model.
While the consumer receivers had better sensitivity than the professional model, only two met the planning factor sensitivity of -81.2 dBm at Channel 12 and none met the -84.2-dBm UHF sensitivity at any of the UHF channels tested. H&E estimated the margin of error for these measurements at +/-1.5 dB. H&E concluded that when all channels are considered, the typical receiver is 2.4 dB less sensitive than the FCC planning factors.
H&E discusses field tests, but considers only 12 measurement campaigns through 1999. It notes that in these tests, 12 percent of the locations that had the requisite signal strength did not have usable pictures.
For obstructed sites, this increased to 18 percent. For indoor reception, adequate reception was not possible at 26 percent of the sites.
H&E offers that future receivers may provide a picture at some of the sites that didn't work in the 1999 and earlier tests, but states, "these results illustrate that there has been a significant failure rate where consumers cannot receive DTV even though a theoretically-adequate signal level is present."
One of the more interesting portions of the H&E engineering statement concerns temporal variation of DTV signal levels. H&E measured the amplitude of 14 DTV signals over both obstructed and unobstructed paths at its Sonoma, Calif. offices from May 18 until June 1, 2005. Looking at the figures I copied from the H&E engineering statement, the variation in signal strength with time is obvious. The sharp drop in signal on May 19 was caused by what H&E called a "mild storm in the San Francisco Bay Area," where rain rates measured as high as 15.5 millimeters per hour.
Given the variations in signal strength, H&E recommends increasing the probability of reception at a given time from the 90 percent, which is what is currently used for predicting DTV coverage, to 99 percent. H&E states their study shows a UHF Channel 41 signal has to be about 17.5 dB above threshold to achieve 99 percent probability.
MSW's reply comments for NAB disputed the H&E studies and engineering statement. How could two well-respected engineering firms look at the U.S. broadcast DTV system from two different angles and come up with results that are so different? Next month I'll discuss MSW's answer to H&E's findings and offer my observations on these and other engineering statements submitted in the Notice of Inquiry.
Questions and comments are always welcome and appreciated. E-mail me at dlung@transmitter.com.
