An Algorithm for Blind Identification and Equalization Based on Equivalent Transformation of Statistical Property
Zhao Le-jun · Huadong Li-Gong Daxue xuebao · 2008
Stacked system model of SIMO channel is introduced by considering channel output oversampling.The concept of statistical property transformation is discussed according to the situation in which channel input is stationary and non-identically independent distribution(i.i.d.) with a priori known autocorrelation function.In this paper,the closed-form solution of equivalent transformation matrix is con-(cluded) and an algorithm for blind identification suitable to stationary source is proposed.Through(simulations,) the algorithm's performances of estimation error and convergence are analyzed.Simulation(results demonstrate) the proposed approach can effectively equalize the channel driven by non-i.i.d.input.(This algorithm) also has advantages of high accuracy and fast convergence.