A new adaptive blind identification of digital communication channels

Abstract

This paper presents a new adaptive procedure for the blind identification of digital communication channel which is equivalently modeled by the discrete-time finite impulse response (FIR) filter. It is observed that H-1(z) = +/-H-2(z) if the correlations and a 1-D cumulant slice of the received sequence of an FIR channel H-2(z) are the same as those of another FIR channel H-2(z) fora white, i.i.d., and non-Gaussian input sequence. Based on this observation, the adaptive algorithm for the nonminimum phase FIR channel identification is proposed. The performance of the proposed identification algorithm is shown through some computer simulations.