Research On PAPR Optimization Method For Tone Reservation OFDM System

Orthogonal Frequency Division Multiplexing is a multi-carrier modulation technique suitable for high-speed wireless transmission.It has been used widely in multipath fading conditions,and has advantages of high spectral efficiency and resistance to multipath fading.The OFDM's inherent Peak to Average Power Ratio problem limits its widespread usage.Among these PAPR reduction techniques proposed in many documents,the Tone Reservation technology has received a lot of attention.This technique uses some reserved subcarriers to generate a peak-canceling signal to reduce the PAPR of the OFDM signal.However,obtaining an optimized peak reduction signal requires solving the problem of quadratically constrained quadratic program,which has a very high computational cost.This paper focuses on the above issues.(1)We designed a FISTA-based reduction method,which is compared with previous the AAC algorithm and the AS algorithm.The AS algorithm has two distinct drawbacks:the selection of the optimal clipping level is very difficult and the different clipping levels result in different PAPR reductions.The AAC algorithm addresses these two shortcomings of the AS algorithm,but the AAC algorithm requires multiple FFT-IFFT calculations to achieve the desired PAPR reduction,and the amplitude of the reserved subcarriers is not limited.Unlike the cutting filter technique above,the proposed FISTA algorithm does not require clipping and filtering,and limits the amplitude of reserved subcarriers,it is a simple and effective method whose iteration only includes simple matrix and vector multiplication.Simulations show that the algorithm can achieve better PAPR reduction performance and better BER performance.(2)We designed a TR algorithm based on the alternating direction multiplier method.The algorithm can give an exact solution to the proximal operator of??~2.In the past literature,only the proximal operator of??was considered,and the calculation of the proximal operator is only a truncation solution.In the ADMM algorithm,we consider the proximal operator of??~2 and give its exact solution.At the same time,the algorithm has an explicit iterative formula for each step of the update and is simple to implement.Compared with the AAC algorithm and the AS algorithm,the proposed alternate direction multiplier method does not require clipping and filtering operations,nor does it select the optimal clipping threshold,and is mainly composed of a pair of FFT-IFFT operations.Compared with the FISTA,they have the same computational complexity,but ADMM algorithm obtains better PAPR reduction and BER performance.