Recent developments in blind channel equalization: From cyclostationarity to subspaces

Abstract

Since Tong et al. (1991) demonstrated the feasibility of identifying possibly nonminimum phase channels using second-order statistics, considerable research activity, both in algorithm development and fundamental analysis, has been seen in the area of blind identification of multiple FIR channels. Many of the recently developed approaches invoke, either explicitly or implicitly, the algebraic structure of the data model, while some others resort to the use of cyclic correlation/spectral fitting techniques. The objective of this paper is to establish insightful connections among these studies and present recent developments of blind channel equalization. We also unify various representative algorithms into a common theoretical framework.