New fractionally spaced recursive polynomials perceptron model for adaptive M-QAM digital mobile radio reception

A new fractionally spaced recursive polynomial perceptron (FSRPP) model for adaptive M-QAM digital mobile radio reception is described, which can adaptively equalise the multipath fading channels and reject non-Gaussian cochannel interference (CCI) simultaneously. Experimental results obtained are satisfactory, which shows that FSRPP with lower computational burden and faster convergence rate can perform more efficiently than polynomial perceptron (PP) and multilayer perceptron (MLP) in the presence of multipath fading of channels and non-Gaussian interference.