CN201910041824
CN109905170A
微信扫码查看/分享专利
|
摘要:本发明属于可见光通信技术领域,具体为一种基于K‑DNN的非线性失真补偿算法及可见光通信装置。其提供非线性失真补偿算法,步骤包括:将经过自适应线性均衡器的串行信号转换为并行信号,作为特征向量输入核函数辅助的DNN均衡器;经K‑DNN训练,把特征向量归类于概率最大的类别,获得原始信号;本装置采用上述非线性失真补偿算法,结构包括:发射端和接收端,接收端通过处理芯片对信号进行离线处理,包括:采用S‑MCMMA的自适应后均衡器补偿线性失真;采用K‑DNN均衡和解映射进行非线性失真补偿,获得原始信号。本发明可以方便、高效补偿现有水下无线光通信中复杂的信道环境对信号造成的非线性失真,具有重要的实用意义。
|
Abstract: The invention belongs to the technical field of visible light communication, and particularly relates to a nonlinear distortion compensation algorithm based on K-DNN and a visible light communicationdevice. The method provides a nonlinear distortion compensation algorithm, and comprises the following steps: converting a serial signal passing through an adaptive linear equalizer into a parallel signal, and inputting the parallel signal as a feature vector into a DNN equalizer assisted by a kernel function; through K-DNN training, classifying the feature vectors into a category with the maximumprobability to obtain an original signal; The device adopts the nonlinear distortion compensation algorithm and structurally comprises a transmitting end and a receiving end, and the receiving end carries out offline processing on signals through a processing chip and comprises the steps of using a S-MCMMA's adaptive post-equalizer to compensate linear distortion; using K-DNN equalization and demapping for nonlinear distortion compensation to obtain the original signal.. The method can conveniently and efficiently compensate the nonlinear distortion of the signal caused by the complex channelenvironment in the existing underwater wireless optical communication, and has important practical significance.
|
申请人: 复旦大学
|
Applicant: UNIV FUDAN
|
地址: 200433 上海市杨浦区邯郸路********(隐藏)
|
发明(设计)人: 迟楠 赵一衡 石蒙
|
Inventor: CHI NAN; ZHAO YIHENG; SHI MENG
|
主分类号: H04B10/116(2013.01)I
|
分类号: H04B10/116(2013.01)I H04B10/58(2013.01)I H04B10/61(2013.01)I H04B10/69(2013.01)I
|
|