Single Frequency Networks (SFNs) are a form of broadcast network design where several transmitters use the same frequency channel to send the same signal. SFNs are usually used for radio and digital TV broadcasting – FM, AM and digital radio can be transmitted over SFNs – while analog television is more problematic because the multiple signal sources tend to produce ghosting.
Why would a broadcaster choose this technique? There are several potential advantages. One is the greater efficiency that SFN can provide: a broadcaster’s available spectrum can be packed with more radio and TV channels than would be possible with a multi-frequency topology. SFNs are also a way of improving coverage area and signal strength, especially in positions that are more or less equidistant between transmitters, where interruptions to reception might otherwise occur. Another particular reason for choosing SFNs is to overcome difficult geographical conditions that interfere with signal delivery. These conditions can occur naturally in the case of sparsely populated areas where transmitters are widely spaced and the terrain is mountainous, but good coverage can also be difficult to obtain in urban areas where densely built-up high-rise development creates a lot of shadowing.
It’s a useful technique for broadcasters to resort to, both when implementing a network infrastructure or – especially in urban applications – when upgrading an existing network to improve coverage and reception quality. But of course there are technical challenges to be met when creating a single frequency network. One of these is the problem of interference between signals reaching the same receiver from more than one transmitter, which can cause fading in DVB-T and DAB networks.
Another key requirement for successfully operating SFNs is highly accurate timing. GPS can be used to arrange phase and frequency co-ordination between the transmitters by providing µs-accurate timing, which is within the parameters for most DVB-T SFN architectures. It’s important that the contribution is timed correctly too, so that the variance in timing between the signal leaving the origination site and reaching each transmitter is taken into account and compensated for. Each transmitter needs to wait until the signal has been received by all of the others before transmission can begin.
Because of these requirements, manufacturers must go to extra lengths to provide technology that can support the required timing accuracy and deliver a good return on investment for the customer. Broadcasters considering the potential scenarios in which SFN may become a desirable option should also note that equipment capable of delivering the accuracy needed for SFN can provide useful benefits in non-SFN applications too. As with most infrastructure purchasing decisions today, flexibility and the means of keeping options open are high on the checklist, so while SFN may not be an immediate priority for some broadcasters, equipment that can deliver accurate timing provides for that possibility in the future, without further investment.
In order both to support SFN and to provide the benefits of accurate synchronization to non-SFN operations, 2wcom has developed its SFN product range. The 2wcom SFN products provide the functionality of a timestamp generator within the transmission feed, inserting the timing information into the MPEG stream. At the transmitter, a FlexDSR02/04+ DVB Satellite Receiver, or MM01, or FMC01 analyzes the signals and ensures the SFN-ready synchronized exact timing of the FM transmission. So the solution delivers precise, micro-second accurate synchronization of the FM cell contribution path, with automatic adjustment of the several contribution runtime differences caused by variable packet delays, jitter and satellite anomalies.
Distribution can be by ASI, E1, IP or satellite, and any one of the feeds can be selected for backup purposes. This allows broadcasters to implement FM SFN networks using existing analog FM technology (just adapted to 1pps synchronisation), with minimal additional hardware: if components such as the MM01 and FMC01 have been installed, they already have the capability for timestamp generation. To upgrade an existing infrastructure from scratch, it just requires one inserter for the uplink, and a 2wcom receiver/decoder at each transmitter site.