Sparse Channel Estimation Including the Impact of the Transceiver Filters with Application to OFDM

摘要

Traditionally, the dictionary matrices used in sparsewireless channel estimation have been based on the discreteFourier transform, following the assumption that the channelfrequency response (CFR) can be approximated as a linearcombination of a small number of multipath components, eachone being contributed by a specific propagation path. In practicalcommunication systems, however, the channel response experiencedby the receiver includes additional effects to those inducedby the propagation channel. This composite channel embodies,in particular, the impact of the transmit (shaping) and receive(demodulation) filters. Hence, the assumption of the CFR beingsparse in the canonical Fourier dictionary may no longer hold.In this work, we derive a signal model and subsequently a noveldictionary matrix for sparse estimation that account for theimpact of transceiver filters. Numerical results obtained in anOFDM transmission scenario demonstrate the superior accuracyof a sparse estimator that uses our proposed dictionary ratherthan the classical Fourier dictionary, and its robustness againsta mismatch in the assumed transmit filter characteristics.