Iterative Tap Pursuit for Channel Shortening Equalizer Design

Abstract

In this work, an iterative tap pursuit algorithm for designing channel shortening equalizers is proposed. Similar to pursuit algorithms, a residual vector is initialized with a desired target impulse response, which is iteratively approximated by one-tap sub-filters. In each iteration, the algorithm selects the location and weight of a one-tap sub-filter. This is proceeded by updating the residual vector by subtracting its already-represented portions by selected sub-filters. The advantage of this algorithm lies in its simplicity in alleviating the need for performing an exhaustive search thus reducing the computational complexity. Convergence of the proposed algorithm is guaranteed by the fact that the energy of the residual decreases with iteration. We show that the proposed algorithm has a significantly reduced computational complexity. Experiments conducted on Rayleigh fading wireless channels validate the effectiveness of the proposed algorithm in designing channel shortening filters in terms of the shortening signal-to-noise ratio measure and complexity.