基于DNN的OFDM系统非线性失真补偿  

DNN⁃based nonlinear distortion compensation of OFDM system

作　　者：刘步花 丁丹 杨柳 LIU Buhua;DING Dan;YANG Liu(Department of Graduate Management,Space Engineering University,Beijing 101416,China;Department of Electronic and Optical Engineering,Space Engineering University,Beijing 101416,China)

机构地区：[1]航天工程大学研究生院,北京101416 [2]航天工程大学电子与光学工程系,北京101416

出　　处：《现代电子技术》2021年第5期1-4,共4页Modern Electronics Technique

基　　金：国家“863”计划(2015AA7026085)。

摘　　要：针对加性高斯白噪声信道下射频发送时高功率放大器导致的非线性失真问题,提出一种基于深度神经网络(DNN)的正交频分复用(OFDM)非线性失真补偿方法。设计一个双输入双输出含两层隐含层的深度神经网络在接收端对非线性失真进行补偿,并讨论了此网络不同的输入输出数据带来的性能差异。仿真结果表明该方法避免了射频发送信号的获取难度和进行输入功率回退时带来的功率损失,具有一定的补偿非线性失真的作用。In view of the nonlinear distortion caused by the high power amplifier during RF transmission in the additive white Gaussian noise channel,a DNN(deep neural network)⁃based nonlinear distortion compensation method of orthogonal frequency division multiplexing(OFDM)system is proposed.A dual input and dual output deep neural network with two hidden layers is designed to compensate the nonlinear distortion at the receiving terminal.The performance differences caused by different input and output data of this network are discussed.The simulation results show this method can avoid the difficulties in the acquisition of the RF transmitted signals and the power loss caused by input power back⁃off,and can compensate nonlinear distortion to a certain extent.

关 键 词：非线性失真补偿 深度神经网络 正交频分复用 高功率放大器 DNN模型训练 仿真分析 

分 类 号：TN914.34-34[电子电信—通信与信息系统]