Estimation of time-varying channels with orthogonal matching pursuit algorithm

Abstract

The adaptive matching pursuit (AMP) algorithm was proposed as a basis selection algorithm for time-varying dictionaries. We propose an adaptive orthogonal matching pursuit (AOMP) algorithm for estimation of time-varying channels with sparse impulse response samples. The performance of the AOMP algorithm is then compared with the AMP algorithm through mean squared identification error and by means of a decision feedback equalizer. It is shown, using simulation results, that the AOMP algorithm, by avoiding the re-selection problem that exists in the AMP algorithm, obtains more accurate channel estimates