Expanding DTV coverage Part V

Single-frequency networks (SFNs) can be a very efficient way to distribute a DTV signal over a large area with challenging terrain. By its very nature, it can deliver DTV programming without undue interference or using another TV channel. This tutorial continues discussing SFNs and how they are used.

The receiver’s role

As mentioned in the last tutorial, using any type of signal fill method will cause multipath interference in some areas, and it’s up to the receiver to be able to distinguish between two time- and level-displaced signals and decode one of them. First-generation receivers could not handle multipath interference very well and failed to receive any picture in many multipath situations. But as designs improved, so did the receiver’s ability to deal with multipath. The ATSC has a recommended plot of multipath levels and timing displacement showing what a DTV receiver should be able to decode. Over time, receivers have met and far exceeded this recommendation. (See Figure 1.)

This improvement in receiver design has allowed SFNs to come into greater use, because without it, the overlapping areas of coverage would have very limited reception capabilities. And although older DTV receivers perform badly in multipath situations, TV station operators can be grateful that they did not sell in great numbers and, therefore, will not pose too big an obstacle to SFNs.

With that said, purchasing the least expensive DTV set-top box (STB) is not recommended; there can be a huge difference between a $50 STB and an $80 STB as far as multipath goes. Directional receive antennas can also have a large role to play with the ability to reduce the level of the unwanted signals. This may prove to be a tipping point because a less expensive STB may not bring in a given DTV signal without the use of a directional antenna (maybe roof mounted), whereas a somewhat more expensive STB may not need such a big (and difficult to install) antenna. Also, if a given STB does not bring in the desired station, try another one because there are variations between boxes.

How an SFN works

There are two basic topologies for an SFN. The first is where a single full-power transmitter covers a majority of the service area and one or more smaller transmitters are positioned to fill in weak or shadowed areas. All transmitters are fed the exact same transport stream from the same location via microwave or fiber optics. This is unlike a booster, where an off-air signal is merely amplified and broadcast again.

The second is new and uses several relatively small transmitters to cover the entire service area, much like a cell phone network. In this case, each transmitter can cover a much larger area because (unlike the cell phone configuration) this is a one-way system. (See Figure 2.)

The RF part of the design is very important because the areas of overlap should be minimized to reduce interference and maximize efficiency. As mentioned above, all transmitters must be fed the exact same transport stream. If the transport stream is not the same, even by one bit, then that signal will act as a jamming signal if it is received in a multipath situation.

Timing transport streams

The ability of a receiver to distinguish between two identical DTV signals on the same frequency is dependent on two factors: their relative signal strengths and the relative timing of the bit stream contained within the RF signal. There is only so much a station can do to control the levels of its signals arriving at a certain point in space, but it can control the timing of the transport stream it transmits from its own transmitters.

In the ATSC SFN standard (A/110B), the transport stream is fed to a special distributed transmission adapter. This unit passes on the transport stream with a few extra bits to control the timing of the transmitted stream at each transmitter in the network. This distributed transmission adapter can address each transmitter individually and adjust its delay. Each transmitter has its own 8-VSB exciter, GPS receiver and circuitry to receive and strip off the commands from the distributed transmission adapter and adjust its built-in delay accordingly.

To set up the network, you first determine the transit time from the transport stream originating point to each transmitter, taking the longest time and dialing that into the path of all other transmitters so the stream arrives at each at the same moment. In areas where multipath exists, the transmitters’ delay is adjusted so the timing difference between the two signals is within the tolerance of the DTV receivers within the affected area. Field measurement can be made to assist in this calibration, but viewer calls can also help in finding the correct delay settings for each transmitter. All timing adjustments are done at one place, where the distributed transmission adapter is located.

Acknowledgments

Richard Schwartz of Axcera contributed to this tutorial.