Skip to content
January 14, 20153 min read

Comparing PSK and FSK-based digital modulations – Part 2


Cover - Comparing PSK and FSK based modulationsIn the first part of this series, we explained the two main types of digital modulation,  Phase Shift Keying (PSK) and Frequency Shift Keying (FSK) with an example of each – Linear Simulcast Modulation (LSM) – a type of PSK, and Continuous Four Frequency Modulation (C4FM) – a type of FSK. We also compared the spectral efficiencies of LSM and C4FM. In this part, we compare the costs and effect of delay spread of the types of digital modulation.

Effect of delay spread

The diagram here shows typical delay spread resulting from multi-path propagation. The received signal consists of multiple copies of the transmitted signal that have traveled different distances and arrived at the receiver at different times. Delay spread tolerance is the maximum time difference between the arrival of the first and last multi-path signal component before the delivered audio quality degrades to the minimum acceptable level. In a digitally modulated system, this occurs when the BER reduces to about 2%.

Let’s consider the effect of delay spread on the eye diagram of C4FM. We can use this to illustrate how much delay spread can be tolerated before the received signal becomes ‘unreadable’. The inherently elliptical shape of the C4FM eye diagram means that, in the presence of delay spread, the eye ‘closes’ quite quickly to the point where the BER degrades to 2% – when the delay spread reaches about 30µS.

Look at the effect of delay spread on the eye diagram of LSM. The more open eye diagram of LSM means it can tolerate more delay spread before it closes. For LSM, the BER doesn’t degrade to 2% until delay spread reaches around 60µS.

So LSM’s more ‘open’ eye diagram means it can tolerate far greater delay spread before we reach the point where BER degrades to 2%.

So what does this mean?

We have established that LSM has twice the delay spread tolerance of C4FM, but what does this mean in practice?

A challenge in radio system design is to get maximum coverage from the minimum number of sites, using the minimum number of channels.

For example, two base stations are transmitting the same signal and a radio-equipped vehicle is passing between them. When the vehicle is much closer to one base station, the radio picks up a strong signal from that base station and a weak signal from the other. Although there are two signals present, with some delay spread, this isn’t too much of a problem as the stronger signal dominates.

The same will be true once the vehicle has traveled to where the signal from the other base station is stronger. However, in the middle, when the radio receives similar strength signals from both base stations, we have a problem. The radio is effectively receiving two identical signals from two different paths at different times, and communication can be lost, if the time difference between their reception exceeds the delay spread tolerance.

One method of managing this problem is simulcast. Simulcast software ensures all sites transmit the same information, on the same frequency, at the same time. Now, as the vehicle is mid-way between the two base stations, the radio receives similar signals from both, but the signals are synchronized in time (minimal delay spread) so they add coherently within the receiver and communication is maintained.

There may still be areas where delay spread can cause loss of communication, typically at edge of coverage. Although the sites are transmitting simultaneously, the radio receives the signals at different times as it is nearer one site than the other – but not to the extent where the signal strength of one site dominates.

We know that LSM has roughly twice the delay spread tolerance of C4FM. This means PSK-based modulations (like LSM) perform much better in simulcast systems than FSK-based schemes. For simulcast to work using C4FM, the signal from Site 1 must arrive within 30µS of the signal from Site 2. For LSM however, the signal from Site 1 only has to arrive within 60µS of the signal from Site 2.

The diagrams illustrate this.

Comparing costs

To limit delay spread in a simulcast network, we must place the sites closer together. If we compare two simulcast networks, one using C4FM and one using LSM, the LSM network would require fewer sites, as the site separation could be much greater. As network cost is related to site separation, the LSM simulcast network is cheaper, even when accounting for LSM requiring more expensive linear base station transmitters.


Tait Connection - Issue 5 This article is taken from Connection Magazine, Issue 5. Connection is a collection of educational and thought-leading articles focusing on critical communications, wireless and radio technology.

Share your views, comments and suggestions in the Tait Connection Magazine LinkedIn group.


Subscribe to Tait Blog

    Hot Topics