A Variable Step-Size Backtracking SAMP Channel Estimation Method for OTFS System

Orthogonal Time Frequency Space (OTFS) modulation has been proposed to provide seamless and reliable communication in high-mobility environments. In particular, OTFS modulation enables the delay-Doppler (DD) domain model to be represented sparsely so that the compressed sensing (CS) algorithms can be applied in channel estimation to get more accurate channel state information (CSI). However, conventional CS algorithms in OTFS channel estimation assume that channel sparsity K is known, which is not available in some practical applications. In this paper, we propose the sparsity adaptive matching pursuit (SAMP) algorithm for OTFS channel estimation without prior information of the channel sparsity K, and we further propose a variable step-size backtracking sparsity adaptive matching pursuit (VSB-SAMP) to improve both accuracy and reconstruction speed. We first formulate the channel estimation problem as a sparse signal recovery problem. Then, the SAMP algorithm is introduced to solve the problem that channel sparsity is unknown. Furthermore, a VSB-SAMP is proposed to accelerate reconstruction speed. Simulation results show that the proposed algorithm can achieve accurate channel state information with less iterations.