Clipping Noise Recovery of OTFS Signals Using Reliability-Based Compressed Sensing

In this correspondence, an efficient clipping noise recovery method using reliability-based compressed sensing (CS) is proposed for orthogonal time frequency space modulation with iterative clipping and filtering. Concretely, the message passing algorithm is first used to acquire the hard decision of transmit signal, along with the corresponding reliability. Then, an efficient method to determine the reliable observations of clipping noise is presented by utilizing the reliable hard decision signals and the sparsity of the equivalent delay-Doppler channel. Finally, from these reliable observations, the clipping noise can be recovered by CS techniques through utilizing its sparsity in the time-domain. Simulation results show that the proposed method can recover the clipping noise efficiently and thus improve the bit-error-rate performance significantly.

I. Introduction

Orthogonal time frequency space (OTFS) is a novel modulation proposed recently for high-mobility communications [1]–[2]. OTFS modulates information symbols directly onto the delay-Doppler plane and multiplexes each information symbol over two dimensional orthogonal basis functions, thus shows significant advantages over orthogonal frequency division multiplexing (OFDM) in time-varying multipath channels. However, the high peak-to-average power ratio (PAPR) problem still exists in OTFS [3]. To address this problem, in [4], for the pilot-embedded OTFS [5], the iterative clipping and filtering (ICF) [6]–[9] was employed to reduce the PAPR by utilizing the guard interval between data and pilot regions. In general, ICF can reduce the PAPR efficiently while also degrades the bit-error-rate (BER) performance.