Matching Pursuit versus threshold-based approach for channel estimation in OFDM systems

Channels exhibiting a sparse impulse response arise in a number of communication applications. In this paper, we address the problem of sparse channel estimation in Orthogonal Frequency Division Multiplexing (OFDM) systems. Our goal is to compare the accuracy of the Matching Pursuit (MP) [1] approach to an optimized threshold-based approach, named the Probabilistic Framework Estimator (PFE) [2]. The comparison of the two considered approaches is carried in the two scenarios of a priori and no priori knowledge about the channel degree of sparsity (CDS). To this end, the PFE algorithm is extended to the case of known probability of active taps. Also, an adapted version of the MP, to the case of unknown number of paths, is proposed.