Chapter 3Adaptive transversal filters

We reconsider the Wiener filter introduced in Section 2.1. Given two random processes images and images, we want to determine the coefficients of a finite impulse response (FIR) filter having input images, so that its output images is a replica as accurate as possible of the process images. Adopting, for example, the mean‐square error (MSE) criterion, the knowledge of the autocorrelation matrix images of the filter input vector and the crosscorrelation images between the desired output and the input vector are required. Next, the filter coefficients are determined by solving the system of equations (2.27). When the autocorrelation and crosscorrelation are not known a priori, but only (long) realizations of and are available, two approaches are possible. The direct approach provides that both ...

Get Algorithms for Communications Systems and their Applications, 2nd Edition now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.