Tracking performance of synchronizers driven by trellis-code modulated signals

The authors investigate the tracking performance of two popular synchronization schemes used for clock and carrier recovery in trellis-code modulated signals. Clock extraction is accomplished by squaring and then filtering the incoming signal with a narrow bandpass filter tuned to the nominal symbol rate. Phase recovery is achieved by means of a decision-directed loop wherein an error voltage, derived from the signal samples and the decoder decisions, is used to update the current estimate of the carrier phase. The standard of comparison is the performance of the same synchronizers when operating with uncoded signals. The main result is that good codes do not affect the tracking operation of the synchronizers. More generally, sufficient conditions are given for the tracking behavior to be code-independent.<>