基于深度学习的低复杂度LDPC译码器  

Low Complexity LDPC Decoder Based on Deep Learning

作　　者：杨祯琳 YANG Zhen-lin(Hubei Key Laboratory of IntelligentWireless Communication,College of Electronics and Information Engineering,South-Central University for Nationalities,Wuhan 430074,Hubei)

机构地区：[1]中南民族大学电子信息工程学院智能无线通信湖北省重点实验室,湖北武汉430074

出　　处：《电脑与电信》2020年第3期62-65,共4页Computer & Telecommunication

摘　　要：在信道译码结合深度学习技术的研究中,维数限制问题一直是研究者们寻求突破的重点。由于深度神经网络是储存密集型,深度神经网络信道解码器通常需要比传统置信传播(BP)译码大得多的计算和内存开销。为了缓解这个问题,提出了一种应用于LDPC码的改进的神经网络译码器。根据深度神经网络信道解码器中权重参数值分布,有选择性地对新的神经网络解码器添加权重参数,通过限制训练参数数量,降低了深度神经网络信道解码器的规模,并且算法与BP译码相比取得了较大译码增益。In the research of channel decoding combined with deep learning technology, the problem of dimension limitation has always been the focus of researchers. Since the deep neural network is storage intensive, the channel decoder of deep neural network usually needs much more computation consumption and memory than the conventional belief-propagation(BP) decoding. In order to alleviate this problem, an improved neural network decoder for LDPC code is proposed. According to the weight parameter value distribution in the deep neural network channel decoder, weight parameters are selectively added to a new neural network decoder.By limiting the number of training parameters, the scale of the deep neural network channel decoder is reduced. And our algorithm gets a large decoding gain than BP decoding.

关 键 词：深度学习 信道译码 LDPC码 

分 类 号：TN911.22[电子电信—通信与信息系统]