Efficient design of trellis coded CPFSK

A concept called matched encoding is applied to find the best combinations of rate-1/2 convolutional codes and continuous-phase frequency-shift keying (CPFSK). Binary CPFSK with a modulation index of 1/4 and quaternary CPFSK with modulation indices of 1/2 and 1/4 are considered. Combinations are optimized in the Euclidean distance sense for a given total number of states in the overall maximum likelihood sequence estimation (MLSE) receiver. Results are compared with those from two other encoding techniques: the traditional approach of combining coding techniques: the traditional approach of combining coding and modulation and the decomposition approach recently introduced by B. Rimoldi (1988 and 1989). It is concluded that there is, as yet, no single approach that can always yield the best possible combination of convolutional coding and modulation and that the model representing the CPFSK modulator plays an important role in the optimization process. The three different encoding techniques considered can yield trellis-coded CPFSK schemes with different performances for the same receiver complexity