Likelihood-Based Demodulation in Multi-User Chaos Shift Keying Communication

This paper deals with the theory of multi-user chaos shift keying communication systems in the coherent case and in particular with bit error performance under near optimal demodulation. The new approximated likelihood and near optimal demodulator is based on the statistical likelihood of the received signal and is a theoretical generalization of the conventional correlation demodulator involving autocorrelations of the spreading. Near optimal theoretical results beginning from an exact analysis are derived for the bit error rate of a typical user; there is emphasis on their structure in regard to separating the effects of channel noise and interference from other users, and in comparison with single-user results. The new demodulator is shown under particular assumptions to be a form of rake demodulator in the sense that it involves autocorrelation-like terms at various lags. The results are illustrated by theoretical calculations and numerical simulations which point out the enhanced performance of the near optimal demodulator relative to the correlation demodulator.