Blind channel identification based on second order cyclostationary statistics

For finite impulse response (FIR), channels with uncorrelated input sequences, L. Tong et al. (1991, 1992) have proposed an identification method under the special rank condition. The relationship is established between the rank condition and identifiability of FIR channels using second-order cyclostationary statistics. The method is generalized to the FIR channels with correlated input sequences. For general autoregressive moving average (ARMA) channels, a method is proposed that directly identifies the zeros and the poles of channels. Computer simulation results show the effectiveness of the proposed method.<>