Neural net-based continuous phase modulation receivers

Continuous phase modulation (CPM) schemes are generally used in peak-power limited transmission systems such as digital satellite communications. Their major drawback, however, is a prohibitively complex receiver structure, particularly in modulation schemes with high packing densities. In this paper we propose feed-forward neural nets (NN) as receivers for partial response CPM systems. Our approach is to replace the entire receiver structure, excluding timing recovery, with a neural net unit whose inputs are time samples of the incoming baseband signals, and whose outputs are the decoded symbols. We present simulation results for coherent and incoherent NN based receivers, and compare their performance with the optimum maximum-likelihood (ML) receiver. A performance analysis of NN-based receivers at large SNR is presented.