Modulation basics, part 3: Spread spectrum and OFDM

Part 2 explains Phase Shift Keying (PSK) and Quadrature Amplitude Modulation (QAM).

Spread spectrum techniques
In many instances it is necessary to keep transmissions as narrow as possible to conserve the frequency spectrum. However, under some circumstances it is advantageous to use what are known as 'spread spectrum techniques', where the transmission is spread over a wide bandwidth. There are two ways of achieving this: one is to use a technique known as frequency hopping, whilst the other involves spreading the spectrum over a wide band of frequencies so it appears as background noise. This can be done in different ways, and the two most widely used systems for this are DSSS and OFDM.

Frequency hopping
In some instances, particularly in military applications, it is necessary to prevent any people apart from intended listeners from picking up a signal or from jamming it. Frequency hopping may also be used to reduce levels of interference. If interference is present on one channel, the hopping signal will only remain there for a short time and the effects of the interference will be short lived. Frequency hopping is a well-established principle. In this system, the signal is changed many times a second in a pseudo-random sequence from a predefined block of channels. Hop rates vary, and are dependent upon the requirements. Typically the transmission may hop a hundred times a second, although at HF this will be much less.

The transmitter will remain on each frequency for a given amount of time before moving on to the next. There is a small dead time before the signal appears on the next channel, and during this time the transmitter output is muted. This is to enable the frequency synthesizer time to settle, and to prevent interference to other channels as the signal moves.

To receive the signal, the receiver must be able to follow the hop sequence of the transmitter. To achieve this, both transmitter and receiver must know the hop sequence, and the hopping of both transmitter and receiver must be synchronized.

Frequency hopping transmissions usually use a form of digital transmission. When speech is used, this has to be digitized before being sent. The data rate over the air has to be greater than the overall throughput to allow for the dead time whilst the set is changing frequency.

Direct sequence spread spectrum
Direct sequence spread spectrum (DSSS) is a form of spread spectrum modulation that is being used increasingly as it offers improvements over other systems, although this comes at the cost of greater complexity in the receiver and transmitter. It is used for some military applications, where it provides greater levels of security, and it has been chosen for many of the new cellular telecommunications systems, where it can provide an improvement in capacity. In this application it is known as Code Division Multiple Access, because it is a system whereby a number of different users can gain access to a receiver as a result of their different 'codes'. Other systems use different frequencies (Frequency Division Multiple Access – FDMA), or different times or time slots on a transmission (Time Division Multiple Access – TDMA).

Its operation is more complicated than those that have already been described. When selecting the required signal, there has to be a means by which the selection occurs. For signals such as AM and FM different frequencies are used, and the receiver can be set to a given frequency to select the required signal. Other systems use differences in time. For example, using pulse code modulation, pulses from different signals are interleaved in time, and by synchronizing the receiver and transmitter to look at the overall signal at a given time, the required signal can be selected. CDMA uses different codes to distinguish between one signal and another. To illustrate this, take the analogy of a room full of people speaking different languages. Although there is a large level of noise, it is possible to pick out the person speaking English, even when there may be people who are just as loudly (or maybe even louder) speaking a different language you may not be able to understand.

The system enables several sets of data to be placed onto a carrier and transmitted from one base station, as in the case of a cellular telecommunications base station. It also allows for individual units to send data to a receiver that can receive one of more of the signals in the presence of a large number of others. To accomplish this, the signal is spread over a given bandwidth. This is achieved by using a spreading code, which operates at a higher rate than the data. The code is sent repeatedly, each data bit being multiplied by each bit of the spreading code successively. The codes for this can be either random or orthogonal. Orthogonal codes are ones which, when multiplied together and then added up over a period of time, have a sum of zero. To illustrate this, take the example of two codes:

Using orthogonal codes, it is possible to transmit a large number of data channels on the same signal. To achieve this, the data are multiplied with the chip stream (Figure 3-20). This chip stream consists of the codes being sent repeatedly, so that the each data bit is multiplied with the complete code in the chip stream – in other words, if the chip stream code consists of four bits, then each data bit will be successively multiplied by four chip bits. It is also worth noting that the spread rate is the number of data bits in the chip code.

Figure 3-20. Multiplying the data stream with the chip stream.

To produce the final signal that carries several data streams, the outputs from the individual multiplication processes are summed (Figure 3-21). This signal is then converted up to the transmission frequency and transmitted.

Figure 3-21. Generating a signal that carries several sets of data.