Abstract

A new algorithm is presented for blind identification of moving average (MA) models using the 1-D slice of its fourth-order cumulants and the autocorrelation functions. The proposed algorithm utilizes the BPSK signals as MA models input that is an independent and identically distributed (i.i.d) process. Our algorithm is compared with Giannakis algorithm. Simulations verify the good performance of the proposed algorithm.

Keywords: Blind identification, higher-order statistics, MA models, BPSK signals