The Tide is Turning

The tide is indeed changing. I refer to the fact that many broadcasters sought to minimize their initial investment in DTV by operating a low-power transmitter, perhaps not certain that DTV would catch on, or doubting the FCC would ever shut down analog transmission in the United States.

Now that the FCC has established its timetable for broadcasters to elect which channel to return, I believe these same broadcasters will seek to maximize their DTV facilities.

For UHF broadcasters with a DTV channel in the 14-51 core set of UHF channels, this would appear to be simple. But readers of this column know of my concern about interference between DTV signals, an unresolved issue largely hinging upon whether manufacturers provide enough linear dynamic RF signal power range.

Intermodulation products generated in a receiver front-end are what cause interference from signals on adjacent and taboo channels. I understand that the ATSC recommends receivers be designed for maximum RF power input level of -8 dBm. But is this the highest level at which third-order intermodulation products are below some yet-to-be-defined level; or what is the significance of this number?

What if more than one powerful signal is received? How do you sum the power in multiple undesired signals? We will examine here what the maximum DTV received power can be and analyze the interference from one or both adjacent channels and from UHF taboo channels. Oh, by the way, there are no D/U limits for DTV-DTV interference from taboo channels. The limits on D/U from adjacent channels are -28 and -26 dB. These were changed in 1998.

Broadcasters seeking an ERP for their DTV signal of 1,000 kW (30 dBK above 1 kW) are required to hold their field strength at the horizon One dB below the field strength at the horizon that would have resulted from their present allocated ERP. In most cases, the allocated power is 50 kW (17 dBK). So in order to increase their ERP by 13 dB to 30 dB, they must lower it at the horizon by 1 dB. The FCC specifically said this could be done by employing sufficient beam-tilt. The required beam-tilt depends on the vertical beam profile of the transmitting antenna. Examples of this are shown in Fig. 1 for three antennas to make the point that antenna gain determines the required beam-tilt to reduce field strength by 14 dB at the horizon.

(click thumbnail)Fig. 1
(click thumbnail)Fig. 2
A reduction in field strength of 14 dB corresponds to a relative amplitude of 0.2 in Fig. 1. This figure gives the vertical beam profiles for UHF antennas with gains of 9.7 (low), 21.8 (medium) and 43.8 (high). These antennas have a symmetrical beam profile; they do not provide null-fill. Most UHF antennas provide null-fill. Three examples are provided in Fig. 2. These figures and the supporting data files were provided by Dr. Oded Bendov and used with his permission and my appreciation.

He noted that low-gain antennas are favored for LPTV, but full-power UHF TV broadcasters use moderate- or high-gain antennas. Gain of transmitting antennas is a number; not expressed in dBlogarithmic units (dB). I think this stems from the fact that the numerical gain of an antenna is related to its physical size in terms of wavelengths.

Dr. Bendov's plots are normalized to a relative gain of one. In these normalized plots, a gain of 0.2 is -14 dB, which is the reduction of field strength required in seeking to increase ERP from 17 dBk to 30 dBK. You give up 1 dB at the horizon to get much stronger signals inside your coverage area, which may benefit users with indoor antennas.


How much beam-tilt is necessary to protect the horizon depends on your antenna gain as shown in Figs. 1 and 2. As null-fill antennas are much more generally used, we will look at Fig. 2, which gives us three examples of vertical beam profiles with null-fill. For an antenna with G=8, Fig. 2 shows the beam-tilt would have to be increased from 1 degree, as shown, to 5.5 degrees to reduce the field strength by 14 dB at the horizon from its peak.

(click thumbnail)
For G=16, the required beam-tilt would be increased from 0.5 degree to 4 degrees and for G=30, the beam-tilt would have to be increased from 0.5 degree to about 1.8 degrees. Higher-gain antennas need less beam-tilt.

Fig. 3 demonstrates. The "nose" of the beam (where relative amplitude is 1.0) for the G=16 antenna will reach 30 feet above ground at a distance from the tower base for an HAAT of 1,200 feet with 4 degrees of beam-tilt.


The maximum field strength for an ERP of 0 dBK (1 kW) at one mile is 102.8 dB above 1 microvolt/meter. It decreases by 6 dB for each doubling of the path length over line-of-sight paths.

I am assuming that there is a line-of-sight path from the roof of this home to the center of radiation of the transmitting antenna, simply because it is only 3.2 miles from the tall tower.

So we can expect the field strength at 3.2 miles to be about 92.5 dB uV/m for an ERP of 0 dBK. For 30 dBK, the estimated field strength at this distance would be 122.5 dB uV/m. Using the well-known dipole factor, which at 615 MHz is 130.8 dB uV/m

-dBm, the received power would be -8.3 dBm for a resonant-dipole-receiving antenna.

With a a high-gain rooftop antenna, the signal power (dBm) available to the receivers will be about the same as the line loss, plus the signal splitter attenuation is about equal to the gain of a good rooftop antenna over a dipole. The effect of beam-tilt is to increase signal levels for many viewers by up to 14 dB. The effect of maximizing ERP is to increase these signal levels by another 13 dB. By some coincidence, the D/U ratio for adjacent DTV channel interference is now -27 dB.

Consider two stations on adjacent channels sharing the same transmitting antenna, an extreme case of being co-sited One maximizes to 30 dBK with beam-tilt, the other stays with 17 dBK ERP. Their field strengths over a line-of-sight path can now differ by 27 dB.

Note that I've assumed co-siting and the same antenna, both of which tend to reduce signal differences. Without co-siting, the D/U can have a greater range. With different antennas, there will be greater signal-level differences. This is a real problem.

What if there are three stations and the third maximizes, too? Is not the D/U ratio now -30 dB? It sure is for me. But as receiver overload (the real cause of adjacent-channel interference) is a nonlinear process, the intermodulation products (the interference) will go up 6 dB for a 3 dB increase in the undesired signal, not 3 dB. With two undesired signals, their combined effect is somewhat stronger, but this has never been investigated.

These unintended consequences will also apply to undesired signals on UHF taboo channels. At present, there are no D/U limits for DTV-DTV taboo channel signal interference.This column has shown that there is a "disconnect" between the levels of signals that will exist due to maximization and the receiver performance recommended recently as a "voluntary standard."

Designing a receiver front-end to remain linear for multiple signals, each of which may exceed -8 dBm, is no simple matter but then, fitting an HDTV signal into 6 MHz was no simple matter either, but we did it.