[1]徐光宪,郭若蕾,陶志勇.基于遗传算法改进的LDPC码译码器结构[J].探测与控制学报,2020,42(03):62.[doi:.]
 XU Guangxian,GUO Ruolei,TAO Zhiyong.An Improved LDPC Decoder Based on Genetic Algorithm[J].,2020,42(03):62.[doi:.]
点击复制

基于遗传算法改进的LDPC码译码器结构()
分享到:

《探测与控制学报》[ISSN:1008-1194/CN:61-1316/TJ]

卷:
42
期数:
2020年03
页码:
62
栏目:
出版日期:
2020-06-15

文章信息/Info

Title:
An Improved LDPC Decoder Based on Genetic Algorithm
文章编号:
1008-1194(2020)03-0062-07
作者:
徐光宪郭若蕾陶志勇
辽宁工程技术大学电子与信息工程学院,辽宁 葫芦岛 125105
Author(s):
XU Guangxian GUO Ruolei TAO Zhiyong
School of Electronics & Information Engineering, Liaoning Technical University, Huludao 125105, China
关键词:
低密度奇偶校验码置信传播译码算法卷积神经网络遗传算法
Keywords:
low density parity check code belief propagation decoding convolutional neural network genetic algorithm
分类号:
TP391
DOI:
.
文献标志码:
A
摘要:
针对LDPC码的BP译码器比最大似然译码器译码准确性低,提出了基于遗传算法改进的LDPC码译码器结构。该译码器结构首先引入卷积神经网络来去除传统BP译码器在译码中的估计误差,然后将遗传算法应用到BP译码中,仿照遗传算法的过程,将所有变量节点作为一个种群,每一个变量节点作为一个个体,对应的变量节点传递给校验节点的概率信息作为基因,通过对优势个体的优秀基因进行加强或者对劣势个体的交叉基因进行削弱,以达到整个种群更加适应环境,从而提高译码性能。仿真实验结果表明,GABP-CNN译码器比标准的BP译码器可以获得更好的纠错性能,尤其是在高信噪比环境下译码性能有较大的提升;但是改进的译码器结构在提高译码性能的情况下,系统运行时间上较传统BP译码器略多。
Abstract:
Aiming at the low accuracy of low density parity check codes of the belief propagation decoder comparing with the maximum likelihood decoder, an improved LDPC code decoder structure based on genetic algorithm was proposed. Firstly, the convolutional neural networks was introduced to remove the estimation error in the traditional BP decoder , then genetic algorithm was applied to the BP decoding. It was modeled on the process of genetic algorithm, that all the variable node was taken as a species, each node as an individual, a variable was passed to the corresponding variable node as a gene, the probability of check nodes information based on the individual advantages was used to strengthen the excellent genes, or to weaken disadvantage individuals crossover gene, thus improved the decoding performance. The simulation results showed that the GABP-CNN decoder proposed in this paper could obtain better error correction performance than the standard BP decoder, especially in the high SNR environment. However, in the case of improved decoder structure, the system running time was slightly more than that of the conventional BP decoder.

参考文献/References:

[1] Raj Chandra Bose, Dwijendra K Ray-Chaudhuri. On a class of error correcting binary group codes[J]. Information and control,1960,3(1):68-79.
[2] Claude Berrou, Alain Glavieux, Punya Thitima-jshima. Near shannon limit error orrecting coding and decoding:Turbo-codes.1[C]//in Communications,1993.ICC’93 Geneva. Technical Program,Conference Record,IEEE International Conference on.IEEE,1993:1064-1070.
[3] Robert Gallager. Low-density parity-check codes[J]. IRE Transactions on information theory,1962,8(1):21-28.
[4] Erdal Arikan. Channel polarization:A method for constructing capacity-achieving codes for symmetric binary-input memoryless channels[J]. IEEE Transactions on Information Theory,2009,55(17):3051-3073.
[5] Nachmani E, Be’ery Y, Burshtein D. Learning to decode linear codes using deep learning[C]// 54 th Annual Allerton Conf. On Communication, Control and Computing. US: IEEE, 2016.
[6] Nachmani E, Marciano E, Lugosch L, et al. Deep learnig methods for improved decoding of linear codes[J]. IEEE Journal of Selected Topics in Signal 2Prcessing, 2018,12(1):119-131.
[7] Gruber T, Cammerer S, Hoydis J, et al. On deep learning-based channel decoding[C]// The 51st Annual Conference on Information Sciences and Systems (CISS). US: IEEE, 2017:1-6.
[8] Cammerer S, Gruber T, Hoydis J, et al. Scaling deep learning-based decoding of polar codes via partitioning[EB/OL]. (2017-10-21) https://arxiv. org/abs/1702.06901.
[9] Lyu W,Zhang Z,Jiao C,et al.Performance evalition of channel decoding with deep neural networks[C]//2018 IEEE International Conference on Communications(ICC). US: IEEE,2018:1-6.
[10] Sae-Young Chung,G David Forney Jr.,Thomas J Richardson, et al. On the Design of Low-density Parity-check Codes within 0.0045 dB of the Shannon Limit[J]. IEEE Communications Letters,2001,5(2):58-60.
[11] 李秀花,高永安,马雯.LDPC码译码算法及性能分析[J].现代电子技术,2014,37(1):1-4.
[12] 陈旭灿,刘冬培.改进的LDPC译码算法研究[J].电子科技大学学报,2010,39(2):219-222.
[13] 屈兰,谢显中,张德民.两种LDPC译码算法在OFDM系统应用中的改进[J].计算机工程与应用,2007(7):129-132.
[14] MacKay D J C, Neal R M. Near Shannon limit performance of low density parity check codes[J]. Electron. Lett., 1996,32(18):1645-1646.
[15] Yuen S, Chow C. A genetic algorithm that apaptively mutes and never revisits[J]. IEEE Transactions on Evolutionary Computation,2009,13(2):454-472.
[16] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
[17] Krizhevsky A, Sutskever I, Hinton G E. Imagenet classification with deep convolutional neural networks[C]//Advances in Neural Information Processing Systems (NIPS). US: NIPS, 2012: 1097-1105.
[18] Ren S, He K, Girshick R, et al. Faster R-CNN: Towards real-time object detection with region proposal networks[C]// Advances in Neural Information Processing Systems (NIPS). US: NIPS, 2015:91-99.
[19] Jain V, Seung S. Natural image denoising with convolutional networks[C]// Advances in Neural Information Processing Systems (NIPS). US: NIPS, 2009: 769-776.
[20] Dong C, Loy C C, He K, et al. Image super-resolution using deep convolutional networks[J]. IEEE Trans. Pattern Anal. Mach. Intell., 2016, 38(2): 295-307.

备注/Memo

备注/Memo:
收稿日期:2019-11-27
基金项目:辽宁省高等学校杰出青年学者成长计划项目资助(LJQ2014029);辽宁省博士启动基金项目资助(20170520098)
作者简介:徐光宪(1977—),男,江苏盐城人,博士,教授,博导,研究方向:网络编码、信息论。E-mail: 2078232582@qq.com
更新日期/Last Update: 2020-07-15