Two-stage interference immune blind equaliser which exploits cyclostationary statistics

Since cyclostationarity is inherent in many man-made signals, there is much to be gained in the signal processing and communication areas if this property is properly recognised and exploited. In particular, the authors propose a novel two-stage scheme which exploits second-order cyclostationary statistics for co-channel interference removal and blind equalisation. The first stage adopts the blind adaptive (BA) FRESH filter proposed by Zhang, to remove co-channel interference and channel noise. The second stage is based on a blind equalisation algorithm, proposed by Tong, which operates on the filtered cyclic signal to remove intersymbol interference. Simulation studies have demonstrated that this two-stage blind equalisation scheme has a fast rate of convergence and excellent performance in a high SNR environment