The author is Intraplex product line manager at Harris Broadcast Communications.
Synchronous FM broadcasting is also known as simulcasting. Everybody’s heard about it. Lots of people are doing it. But not everyone understands the subtleties involved in getting it right.
Let’s look at an important but oft-neglected aspect of effective simulcasting: dynamic delay control.
In this context, simulcasting is the use of multiple transmitters with overlapping coverage, operating on the same frequency, to cover a single market.
There are a number of reasons we might want to do this. One is to fill in coverage gaps caused by natural or man-made obstructions such as hills or clusters of high-rise buildings. Another reason is to saturate an urban market without the expense of a high-powered downtown transmitter. Some broadcasters have done well by purchasing relatively inexpensive suburban stations surrounding a city and directing their antennas to cover the metropolis in the center. Multiple simulcast transmitters are used in some places to cover travel corridors such as highways. In general, U.S. FM broadcasters are allowed to license boosters that operate at up to 20 percent of their main transmitters’ ERP levels for whatever purposes they deem appropriate.
Simulcasting clearly offers functional and financial benefits, but getting a simulcast system to work at its best presents a number of technical challenges. This is because of the way that the listener’s radio receiver responds in areas with multiple signals on the same frequency.
Fig. 1: Power vs. Distance The receiver captures a signal cleanly when the listener is close enough to one transmitter or the other. However, the receiver may alternate between the two signals in the overlap area, and any variation between the two in frequency, phase, amplitude or arrival time can cause interference. This renders the receive signal unpleasantly noisy or even unlistenable (Fig. 1).
Frequency and phase alignment can be handled by using GPS clocks to synchronize the inputs to the various transmitters in the system. Amplitude variation can be controlled by using an all-digital air chain. Controlling arrival time can be addressed by using precisely adjustable digital delays, but determining the correct arrival time at each transmitter is much trickier, and is a crucial aspect of simulcasting. To understand why, let’s look at what it takes to optimize a simulcast installation.
No FM simulcast system can provide 100 percent perfect coverage over the entire region between two adjacent transmitters. But a well-designed system can minimize the amount of undesirable interference and also allow the engineers to maximize the performance of the system in the most important coverage areas (busy highways, crowded residential areas) and move the areas of remaining interference to less populated regions.
Fig. 2: Equal ERP Transmitters Fig. 2 shows the area where we need to be concerned about interference between two overlapping transmitters of equal ERP on relatively flat and simple terrain. If we precisely equalize the delay from the studio to each of the two towers, we optimize coverage in the overlap area.
What if the area we’re mainly interested in covering is not dead center between the two transmitters? Fig. 3 shows this scenario, and in this case we would want to delay the signal arriving at transmitter #2 more than that arriving at transmitter #1. This allows both signals to arrive at the same time in the metro area we’re trying to cover.
Another scenario is that which occurs when we’re using transmitters of differing ERP, as is typical when using one transmitter as the main and another as a booster or gap filler (Fig 4). The challenges increase in this situation as the line of equal delay between the two transmitters cannot be made to match that of equal signal level precisely. Add to this the fact that most real-world simulcast projects involve terrain issues as well, and we can see that deciding on the correct amount of delay on each STL to optimize the overall performance of the system is a complex skill — essentially, a combination of art and science.
Fig. 3: Off-Center Coverage Area The bottom line is that it is critical to maintain the exact amount of delay that the system engineers implement on each link.
In a simulcast system, engineers set the desired delay on each STL link by first measuring the actual delay inherent on the STL. The proper amount of additional delay then is added to raise the total to the desired amount.
How precise does this delay control need to be? Radio propagates at 3.33 µs per km, so a system that provides control to an accuracy of 200 ns allows the engineers to adjust coverage areas in as little as 30-meter increments. However, all the accuracy in the world doesn’t help when the absolute delay changes. This is where dynamic delay control comes into play.
The actual delay on an STL circuit can change over time. T1 circuits tend to have fairly stable delay characteristics. Public networks, however, are subject to rerouting. This means that the phone company or service provider can shift the data to a different physical network path in the event of hardware fault or excessive congestion. Rerouting can cause a sudden and dramatic change in the overall circuit delay, and can happen as often as several times each day without warning.
Fig. 4: Unequal ERP Transmitters IP packet networks are even more prone to delay variation than T1, in the form of jitter due to packet routing and buffering. Even point-to-point microwave links can be subject to variable delays as a result of data buffering in modems or other equipment. For example, spread spectrum and other types of radios can be designed to operate cleanly in the face of high levels of interference, but the mechanisms that allow them to do so can affect the throughput delay.
This means that one key factor in the successful operation of a simulcast system is a mechanism that provides dynamic delay control — “dynamic” referring to the fact that the system actively monitors the actual delay on the STL circuit at all times. The amount of delay being added is adjusted whenever it detects a change to maintain the desired total on each link.
Typically, this involves adding a GPS-derived time stamp to the signal leaving the studio, and then comparing this timing indication to a GPS clock at the transmitter site. The system must then have the intelligence to monitor the STL delay for changes and tweak the amount of delay being added as necessary to maintain the total with an acceptable degree of precision. Only systems capable of achieving this highly precise degree of dynamic delay control can provide the kind of reliability that makes them suitable for real-world broadcast applications.
Junius Kim, “RF Simulcasting Over IP Networks,” Proceedings of the National Association of Broadcasters, Broadcast Engineering Conference, April 2008
Bill Gould and Jai Eu, “Using Synchronized Transmitters for Extended Coverage in FM Broadcasting,” Harris Broadcast Communications White Paper