Spread Spectrum – Safe haven for wireless consumer applications—Part III

Miss Part I? Click here.
Miss Part II? Click here.

In every asynchronous digital communication, since the receiver generates a local clock independently, it becomes mandatory for the receiver to employ a mechanism to synchronize itself with the transmitter. Without such synchronization, it would be impossible for the receiver to decode the incoming signal. Being asynchronous in nature, Spread spectrum systems have the same requirement as well. These systems need to synchronize the PN code (for DSSS) and frequency hopping pattern (for FHSS).

This synchronization is basically established in two phases – ‘Acquisition’ and ‘Tracking’. In the acquisition phase, the receiver primarily detects whether the incoming signal is from the desired source or not. This can be considered as a coarse tuning, where signals from all undesirable sources are rejected as noise. While in the tracking phase, the receiver does a fine synchronization and tries to track the incoming signal in terms of phase, frequency, or both. This is achieved by using an appropriate locking mechanism.

Since DSSS systems and FHSS systems use entirely different principles for spreading, they employ inherently different techniques for both acquisition and tracking.


Synchronization in DSSS

In a DSSS receiver, if the correlator output is less than a certain minimum threshold, it will discard the incoming sequence, considering it as background noise. Since the auto-correlation of a PN code is minimal, the correlator output would be very low (ideally zero) if the incoming sequence and the locally generated PN code are not phase synchronized (fig. 1). Therefore, without any explicit measures for synchronization, the receiver will not be able to decode the incoming signal reliably.

It is important to note that since the signal spreading in DSSS is basically achieved by the PN code, the carrier frequency of the transmitter remains the same. Hence, there is no need for frequency synchronization between the transmitter and the receiver. The only synchronization required is phase synchronization.  


Acquisition

The acquisition process makes use of the fact that correlation (both auto and cross) between the incoming sequence and the locally generated one will have a peaked maximum for perfect synchronization. It searches for a phase, at which the correlation exceeds a predefined threshold. This search strategy can be either ‘serial’ or ‘parallel’.

In ‘serial search’, a monitoring circuit keeps a check on the correlator output. If the output fails to reach a threshold value, the ‘search control’ block shifts the phase of the generated PN code (fig. 2). This process of monitoring and shifting is carried out until the correlator output reaches the threshold value, which marks the completion of acquisition. The configuration, as a whole, forms a feedback loop which is referred to as a ‘sliding correlator’.


One drawback associated with this approach is that acquisition times are high. Hence, some designs make use of a parallel approach (Figure 3).

Except for the use of multiple correlators, a ‘Parallel search’ strategy is no different from a serial one. This approach provides an advantage in terms of acquisition time because multiple phase comparisons can be carried out simultaneously. The acquisition time is the lowest when the number of correlator equals the number of chips (N) in the PN code because comparison with all possible phases is carried out in a single cycle. For obvious reasons, this advantage comes at the cost of increased hardware resources and complexity.

 

It is important to note here that the acquisition process, both serial and parallel, only achieves a coarse synchronization. The degree of synchronization achieved at the end of this phase is within ±T_c/2, where T_c is the chip duration.