一种基于循环神经网络的极化码BP译码算法

doi:10.19678/j.issn.1000-3428.0060159

计算机工程 ›› 2022, Vol. 48 ›› Issue (1): 197-203. doi: 10.19678/j.issn.1000-3428.0060159

一种基于循环神经网络的极化码BP译码算法

何彦琦^1,2, 彭大芹^1,2, 赵雪志^1,2

1. 重庆邮电大学通信与信息工程学院, 重庆 400065;
2. 重庆邮电大学电子信息与网络工程研究院, 重庆 400065

收稿日期:2020-12-01 修回日期:2021-01-05 发布日期:2021-01-11
作者简介:何彦琦(1996-),男,硕士研究生,主研方向为极化码译码、SCMA多用户检测;彭大芹(通信作者),正高级工程师;赵雪志,硕士研究生。
基金资助:
国家自然科学基金“基于SIW圆极化赋形阵列的毫米波可重构切换波束天线研究”（E020B2018023）。

A Recurrent Neural Network Based BP Decoding Algorithm for Polar Codes

HE Yanqi^1,2, PENG Daqin^1,2, ZHAO Xuezhi^1,2

1. School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications, Chongqing 400065, China;
2. Institute of Electronic Information and Network Engineering, Chongqing University of Posts and Telecommunications, Chongqing 400065, China

Received:2020-12-01 Revised:2021-01-05 Published:2021-01-11

摘要/Abstract

摘要： 置信传播（BP）算法作为极化码最常用的软判决输出译码算法之一，具有并行传输、高吞吐量等优点，但其存在收敛较慢、运算复杂度高等缺陷。提出一种基于循环神经网络的偏移最小和近似置信传播译码算法。通过偏移最小和近似算法替代乘法运算，修改迭代过程中的消息更新策略，并运用改进的循环神经网络架构实现参数共享。仿真结果表明，相比传统BP译码算法，该译码算法在提升误码率（BER）性能的前提下，减少约75%的加法运算且收敛速度大幅提升，相比基于深度神经网络的BP译码算法，该算法在确保BER性能无显著下降的前提下，使用加法运算替代乘法运算，节省了约80%的存储空间开销。

关键词: 极化码, 置信传播, 循环神经网络, 偏移最小和, 运算复杂度

Abstract: Belief Propagation(BP) is one of the most commonly used soft decision decoding algorithms for polar codes, which enables parallel transport and displays high throughput.However, BP suffers from slow convergence and high computational complexity.To address the problem, this paper proposes a Recurrent Neural Network(RNN)-based approximate BP decoding algorithm for polar codes with and Offset Min-Sum(OMS).The message update strategy is modified in the iterative process by replacing multiplication operations with the minimum offset and approximation algorithm, and an improved RNN architecture is used to realize parameter sharing.Simulation results show that compared with traditional BP algorithms, the proposed decoding algorithm can reduce addition operations by about 75% and greatly improve the convergence speed and the Bit Error Ratio(BER) performance. Compared with the DNN-BP decoding algorithm, the proposed algorithm uses addition operation to replace multiplication operation, and saves about 80% storage space overhead with no significant decline in BER performance.

Key words: polar codes, Belief Propagation(BP), Recurrent Neural Network(RNN), Offset Min-Sum(OMS), computational complexity

中图分类号:

TN919.3

何彦琦, 彭大芹, 赵雪志. 一种基于循环神经网络的极化码BP译码算法[J]. 计算机工程, 2022, 48(1): 197-203.

HE Yanqi, PENG Daqin, ZHAO Xuezhi. A Recurrent Neural Network Based BP Decoding Algorithm for Polar Codes[J]. Computer Engineering, 2022, 48(1): 197-203.

https://www.ecice06.com/CN/Y2022/V48/I1/197

图/表 8

20220108123111

20220108123115

20220108123119

20220108123123

20220108123127

20220108123131

20220108123135

20220108123138

参考文献

[1] ARIKAN E.Channel polarization:a method for constructing capacity-achieving codes[C]//Proceedings of 2008 IEEE International Symposium on Information Theory.Washington D.S., USA:IEEE Press, 2008:143-156.
[2] ARIKAN E.Channel polarization:a method for constructing Capacity-achieving codes for symmetric binary-input memoryless channels[J].IEEE Transactions on Information Theory, 2009, 55(7):3051-3073.
[3] ALAMDAR-YAZDI A, KSCHISCHANG F R.A simplified successive-cancellation decoder for polar codes[J].IEEE Communications Letters, 2013, 17(12):2360-2363.
[4] 刘亚军, 李世宝, 刘建航, 等.一种低时延极化码列表连续删除译码算法[J].计算机工程, 2018, 44(3):78-81. LIU Y J, LI S B, LIU J H, et al.A low-latency successive cancellation list decoding algorithm for polar codes[J]. Computer Engineering, 2018, 44(3):78-81.(in Chinese)
[5] FAYYAZ U U, BARRY J R.Polar codes for partial response channels[C]//Proceedings of IEEE International Conference on Communications.Washington D. S., USA:IEEE Press, 2013:4337-4341.
[6] YANG J, ZHANG C, ZHOU H, et al.Pipelined belief propagation polar decoders[C]//Proceedings of 2016 IEEE International Symposium on Circuits and Systems. Washington D. S., USA:IEEE Press, 2016:352-367.
[7] YUAN B, PARHI K K.Architecture optimizations for BP polar decoders[C]//Proceedings of IEEE International Conference on Acoustics.Washington D.S., USA:IEEE Press, 2013:653-669.
[8] YUAN B, PARHI K K.Early stopping criteria for energy-efficient low-latency belief-propagation polar code decoders[J].IEEE Transactions on Signal Processing, 2014, 62(24):6496-6506.
[9] NACHMANI E, MARCIANO E, LUGOSCH L, et al.Deep learning methods for improved decoding of linear codes[J].IEEE Journal of Selected Topics in Signal Processing, 2018, 12(1):585-597.
[10] XU W, WU Z, UENG Y L, et al.Improved polar decoder based on deep learning[C]//Proceedings of 2017 IEEE International Workshop on Signal Processing Systems. Washington D.S., USA:IEEE Press, 2017:235-246.
[11] DOAN N, HASHEMI S A, MAMBOU E N, et al.Neural belief propagation decoding of CRC-Polar concatenated codes[EB/OL].[2020-11-01].https://arxiv.org/pdf/1811.00124.pdf.
[12] WANG X, ZHENG Z, LI J, et al.Belief propagation bit-strengthening decoder for polar codes[J].IEEE Communications Letters, 2019, 23(11):1958-1961.
[13] CHEN C H, TENG C F, WU A Y.Low-complexity LSTM-assisted bit-flipping algorithm for successive cancellation list polar decoder[C]//Proceedings of 2020 IEEE International Conference on Acoustics, Speech and Signal Processing.Washington D.S., USA:IEEE Press, 2020:1451-1467.
[14] WEN C, XIONG J, GUI L, et al.A BP-NN decoding algorithm for polar codes[C]//Proceedings of the 11th International Conference on Wireless Communications and Signal Processing.Washington D.S., USA:IEEE Press, 2019:325-331.
[15] TENG C F, HO K S, WU C H, et al.Convolutional neural network-aided bit-flipping for belief propagation decoding of polar codes[C]//Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing.Washington D.S., USA:IEEE Press, 2019:656-667.
[16] CAO Z, ZHU H, ZHAO Y, et al.Learning to denoise and decode:a novel residual neural network decoder for polar codes[C]//Proceedings of the 92nd IEEE Vehicular Technology Conference.Washington D.S., USA:IEEE Press, 2019:764-778.
[17] XU W, TAN X, BEERY Y, et al.Deep learning-aided belief propagation decoder for polar codes[J].IEEE Journal on Emerging and Selected Topics in Circuits and Systems, 2020, 55(9):145-156.
[18] LUGOSCH L, GROSS W J.Neural offset min-sum decoding[C]//Proceedings of IEEE International Symposium on Information Theory.Washington D.S., USA:IEEE Press, 2017:1361-1365.
[19] TENG C F, CHEN H, HO K S, et al.Low-complexity recurrent neural network-based polar decoder with weight quantization Mechanism[C]//Proceedings of 2019 IEEE International Conference on Acoustics, Speech and Signal Processing.Washington D.S., USA:IEEE Press, 2019:667-675
[20] WODIANY I, POP A.Low-precision neural network decoding of polar codes[C]//Proceedings of the 20th IEEE International Workshop on Signal Processing Advances in Wireless Communications.Washington D.S., USA:IEEE Press, 2019:256-268.
[21] PAN Z, LI E, ZHANG L, et al.Design and optimization of joint iterative detection and decoding receiver for uplink polar coded SCMA system[J].IEEE Access, 2018, 7:11365-11376.
[22] SUN F, NIU K, DONG C.Deep learning based joint detection and decoding of non-orthogonal multiple access systems[C]//Proceedings of 2018 IEEE GLOBECOM Workshops.Washington D.S., USA:IEEE Press, 2018:367-378.

选择文件类型/文献管理软件名称

选择包含的内容

一种基于循环神经网络的极化码BP译码算法

A Recurrent Neural Network Based BP Decoding Algorithm for Polar Codes

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	更藏措毛, 黄鹤鸣, 杨毅杰. 融合多尺度特征与上下文信息的语音增强方法[J]. 计算机工程, 2024, 50(6): 138-147.
[2]	赵季红, 张富, 崔曌铭. 基于自适应动态预测的网络切片资源冲突优化[J]. 计算机工程, 2024, 50(1): 183-190.
[3]	杨静, 陆铭华, 马洁琼, 吴金平, 刘星璇. 基于交替循环神经网络的水下防御态势预测方法[J]. 计算机工程, 2023, 49(9): 69-78.
[4]	张会云, 黄鹤鸣. 基于异构并行神经网络的语音情感识别[J]. 计算机工程, 2022, 48(4): 113-118.
[5]	李小剑, 谢晓尧, 徐洋, 张思聪. 基于CNN-SIndRNN的恶意TLS流量快速识别方法[J]. 计算机工程, 2022, 48(4): 148-157,164.
[6]	施荣华, 金鑫, 胡超. 基于图注意力网络的方面级别文本情感分析[J]. 计算机工程, 2022, 48(2): 34-39.
[7]	吴铁洲, 邹智, 姜奔, 张晓星. 基于TLBGA-GRU神经网络的短期负荷预测[J]. 计算机工程, 2022, 48(11): 69-76.
[8]	李江和, 王玫. 一种用于因果式语音增强的门控循环神经网络[J]. 计算机工程, 2022, 48(11): 77-82.
[9]	朱思猛, 杜瑞颖, 陈晶, 何琨. 基于循环神经网络的Web应用防火墙加固方案[J]. 计算机工程, 2022, 48(11): 120-126.
[10]	张稣荣, 卜佑军, 陈博, 孙重鑫, 王涵, 胡先君. 基于多层双向SRU与注意力模型的加密流量分类方法[J]. 计算机工程, 2022, 48(11): 127-136.
[11]	徐兴荣, 张帅鹏, 李婷, 郭娜, 董乐乐, 刘聪, 任崇广. 基于轨迹聚类的业务流程剩余时间预测方法[J]. 计算机工程, 2022, 48(11): 247-256,265.
[12]	黄胜, 郑秀凤, 曹志雄. 基于对数似然比与极化信道可靠度的SCF译码算法[J]. 计算机工程, 2022, 48(1): 170-174,181.
[13]	苗佳, 段跃兴, 张月琴, 张泽华. 基于CNN-BiGRU模型的事件触发词抽取方法[J]. 计算机工程, 2021, 47(9): 69-74,83.
[14]	钱声攀, 于洋, 翟天一, 张徐东, 常彦博. 基于深度学习的主机负载在线预测模型研究[J]. 计算机工程, 2021, 47(9): 84-89.
[15]	王会勇, 卢超, 张晓明. 基于小样本学习和语义信息的图像描述模型[J]. 计算机工程, 2021, 47(8): 260-270.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

一种基于循环神经网络的极化码BP译码算法

A Recurrent Neural Network Based BP Decoding Algorithm for Polar Codes

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献

相关文章 15

编辑推荐

Metrics

本文评价