Compressive channel estimation using band approximation for OFDM systems

In this paper we propose a compressive channel estimation for fast fading channel in orthogonal frequency division multiplexing (OFDM) systems. We formulate the sparse compressive sensing (CS) problem by exploiting the delay-Doppler sparse structure of the doubly dispersive channel. To combat the severe inter-carrier interference (ICI) caused by the Doppler shift, this estimator is based on the banded matrix model in frequency domain. The band approximation enable the CS technology to obtain a precise estimation of the fast fading channel with feasible complexity. We also utilize the random grouped pilot pattern, which allows a direct construction of measurement matrix for the proposed model. Simulation results demonstrate that our method achieves significant performance gain when the Doppler spread is at high level.