INFINITE LENGTH CHANNEL SHORTENING FILTERING BASED ON POLYNOMIAL APPROACH

摘要

Maximum-likelihood estimation and multicarrier modulation techniques are effective methods to mitigate intersymbol interference (ISI) in wireless and wireline communications, but their complexity and performance are limited by the channel length. Channel shortening equalization is an efficient way to shorten the length of the ISI channel while keeping most of the channel energy inside the shortened window. In this paper, we propose an infinite impulse response (IIR) filter for channel shortening and provide polynomial based algorithms to compute the shortening filter. We also considered a very general framework allowing auto-regressive moving average (ARMA) input and noise processes together with an ARMA channel model. An illustrative numerical example shows the superior performance of our proposed scheme as compared to existing techniques.