Initial Probability Adaptation Enhanced Cross-Entropy-Based Tone Injection Scheme for PAPR Reduction in OFDM Systems

Cross entropy (CE) is a method that can be used to obtain the optimal perturbation signal in tone injection (TI) scheme to reduce the peak-to-average power ratio (PAPR) of orthogonal frequency division multiplexing (OFDM) signals. Since CE is an iterative method, the initial probability of CE plays an important role in converging toward the optimal solution. Thus, we focus on finding a more preferable initial probability for CE in this paper. Firstly, we analyze the perturbation signals of TI scheme. It is found that the number of the subcarriers needed to be perturbed is much smaller than the number of the subcarriers of OFDM systems, which demonstrates the perturbed subcarriers of perturbation signals exhibits the property of sparsity. Secondly, we assign different subcarriers with different perturbation weight defined according to the power of clipping noise. Thirdly, we select a candidate subcarrier set with limited size from the whole subcarrier set according to perturbation weight. Finally, by utilizing sparsity and perturbation weight in the determination of initial probability, we propose an enhanced CE method whose initial probability is problem-dependent and adaptive to OFDM symbols. Simulation results verify that the proposed initial probability adaptation enhanced CE-based TI scheme outperforms the traditional CE-based TI scheme and parametric minimum cross entropy (PMCE) based TI scheme on the PAPR, complexity and BER performance.

I. Introduction

High peak-to-average power ratio (PAPR) is a major drawback of orthogonal frequency division multiplexing (OFDM) systems. To amplify OFDM signals without introducing nonlinear distortion, it needs to lower the operating point of high power amplifiers (HPAs), which results in a decrease of the efficiency of HPAs. Therefore, it is necessary to lower the PAPR of OFDM signals.