Adaptive blind equalization of simo systems based on canonical correlation analysis

Abstract

In this work the problem of blind equalization of single-input multiple-output (SIMO) systems is formulated as a canonical correlation analysis (CCA) problem. CCA is a classical tool that finds maximally correlated projections of several data sets, and it is typically solved using eigendecomposition techniques. Recently, it has been shown that CCA can be alternatively viewed as a set of coupled least squares regression problems, which can be solved eigendecomposition using a recursive least squares (RLS) algorithm. Similarities and differences between the proposed CCA-RLS algorithm and other blind equalization techniques based on second-order statistics (SOS) are discussed in the paper. In particular, unlike other approaches, the CCA procedure obtains simultaneously the optimal equalizers as well as the best combination of their outputs. Furthermore, the reference signal derived from the regression framework can be readily modified to obtain a soft decision equalizer with improved performance. Some simulation results are presented to demonstrate the potential of the proposed CCA-RLS algorithm.