Robust chaotic communications exploiting waveform diversity. Part 2: Complexity reduction and equalisation

Abstract

The symbolic dynamics method of communication using chaotic systems has previously been shown to offer performance better than conventional binary phase shift keying (BPSK), while also providing similar spectrum efficiency. This is achieved by exploiting diversity in the waveform, through a mechanism similar to partial response signalling. In order to achieve this performance, a correlation method for detection has been proposed, but complexity was high and only performance in additive white Gaussian noise (AWGN) channels has been considered. It is demonstrated that the complexity can be reduced from requiring 1024 correlations per symbol down to only 32, while degrading performance only by 0.6 dB. The ability to further reduce the occupied bandwidth is investigated. Spectrum occupancy equivalent to a root-raised cosine-filtered BPSK signal is demonstrated, and performance is maintained when the reference waveforms are similarly filtered to maintain the signal match. Finally, a novel equalisation technique that incorporates a decision feedback structure into the correlation detector is proposed. Performance in multipath channels is investigated and shown to be effective.