PAPR Reduction of OTSM with Random and Orthogonal SLM Phase Sequences and its Recovery in the Presence of EPA, EVA and ETU Channel Models

Abstract

To overcome the high peak-to-average-power ratio (PAPR) in orthogonal time sequency multiplexing (OTSM), this document proposes orthogonal and random selective mapping phase vectors (OSLM and RSLM) after some modifications to be applicable with OTSM constructions at both transmitter and receiver. Consequently, the PAPR (without explicit side information) of the original phase vectors is reduced and restored in the presence of a nonlinear power amplifier, and three different multipath fading channel models: Extended pedestrian A (EPA), Extended vehicular A (EVA) and Extended typical urban (ETU), with different user speed, namely 150 and 500 km/h. Also, complementary cumulative distribution function (CCDF) PAPR comparisons are provided based on the Walsh Hadamard transform (WHT) matrix W_N and the inverse symplectic fast Fourier transform (ISFFT) matrix F_N. Furthermore, three detectors are modified and implemented in this work, namely single tap MMSE (ST-MMSE), Gauss-Seidel iterative matched filter (MFGS) and linear MMSE (LMMSE). As a result, this proposal shows reliable performance in terms of PAPR reduction, bit error rate (BER) and side information error rate (SIER) of the OTSM system at a fraction of extension values (f=0.09 and C=1.15) while maintaining power efficiency.