基于轻量级深度神经网络的电磁信号调制识别技术

doi:10.11959/j.issn.1000-436x.2020237

通信学报 ›› 2020, Vol. 41 ›› Issue (11): 12-21.doi: 10.11959/j.issn.1000-436x.2020237

• 专题：面向6G的智能无线通信关键技术 • 上一篇下一篇

基于轻量级深度神经网络的电磁信号调制识别技术

张思成¹,林云¹,涂涯¹,ShiwenMao²

¹ 哈尔滨工程大学信息与通信工程学院，黑龙江哈尔滨 150001
² 奥本大学电子和计算机工程学院，奥本 36849

修回日期:2020-11-06 出版日期:2020-11-25 发布日期:2020-12-19
作者简介:张思成（1996- ），男，山东临沂人，哈尔滨工程大学博士生，主要研究方向为基于深度学习的智能电磁信号处理|林云（1980- ），男，黑龙江哈尔滨人，哈尔滨工程大学在站博士后，哈尔滨工程大学教授、博士生导师，主要研究方向为人工智能、深度学习、信号识别和智能信息对抗等|涂涯（1994- ），男，湖北十堰人，哈尔滨工程大学博士生，主要研究方向为信号处理、深度学习等|Shiwen Mao（1971- ），男，湖南岳阳人，博士，奥本大学教授，主要研究方向为无线网络、多媒体通信和智能电网
基金资助:
国家自然科学基金资助项目(61771154);中央高校基本科研业务费基金资助项目(3072020CF0813)

Electromagnetic signal modulation recognition technology based on lightweight deep neural network

Sicheng ZHANG¹,Yun LIN¹,Ya TU¹,Shiwen Mao²

¹ College of Information and Communication Engineering,Harbin Engineering University,Harbin 150001,China
² Department of Electrical and Computer Engineering,Auburn University,Auburn 36849,USA

Revised:2020-11-06 Online:2020-11-25 Published:2020-12-19
Supported by:
The National Natural Science Foundation of China(61771154);The Fundamental Research Funds for the Central Universities(3072020CF0813)

摘要/Abstract

摘要：

针对6G时代将会是移动通信与人工智能紧密结合的时代，产生数量庞大的边缘智能信号处理节点的趋势，提出了一种可部署于资源受限的边缘设备上的高效智能电磁信号识别模型。首先，通过绘制电磁信号的星座图将电磁信号具象为二维图像，并根据归一化点密度对星座图上色以实现特征增强；然后，使用二值化深度神经网络对其进行识别，在保证识别准确率的同时明显降低了模型存储开销以及计算开销。采用电磁信号调制识别问题进行验证，实验选取常用的8种数字调制信号，选择加性高斯白噪声为信道环境。实验结果表明，所提方案可以在信噪比为-6～6 dB的噪声条件下获得96.1%的综合识别率，网络模型大小仅为166 KB，部署于树莓派4B的执行时间为290 ms，相比于同规模的全精度网络，准确率提升了0.6%，模型缩减到 $\frac{1}{26.16}$ ，运行时间缩减到 $\frac{1}{2.37}$ 。

关键词: 6G, 边缘智能, 电磁信号调制识别, 图像具象化, 二值化深度神经网络

Abstract:

In response to the trend that in the 6th generation wireless (6G) era,mobile communications and artificial intelligence will be closely integrated,and a huge number of edge intelligent signal processing nodes will be deployed,an efficient and intelligent electromagnetic signal recognition model was proposed,which could be deployed on resource-constrained edge devices.The constellation diagram of electromagnetic signal was firstly drawn to visualize electromagnetic signal as a two-dimensional image,and color the constellation diagram according to the normalized point density to achieve feature enhancement.Then,a binary deep neural network was adopted to recognize the colored constellation diagrams.It was shown that the approach can guarantee a high recognition accuracy,which significantly reduced the model storage and calculation costs.For verification,the proposed approach was applied to the problem of electromagnetic signal modulation recognition.The experiment uses eight commonly used digital modulation signals and selects additive white Gaussian noise as the channel environment.The experimental results show that the scheme can achieve a comprehensive recognition rate of 96.1% under the noise condition of -6～6 dB,while the size of the network model is only 166 KB.Further,the execution time,when executed on a Raspberry Pi 4B,is only 290 ms.Compared to a full-precision network of the same scale,the accuracy is increased by 0.6%,the model is reduced to $\frac{1}{26.16}$ ,and the running time is reduced to $\frac{1}{2.37}$ .

Key words: 6G, edge intelligence, electromagnetic signal modulation recognition, image visualization, binary deep neur-al network

中图分类号:

TN971

张思成,林云,涂涯,ShiwenMao. 基于轻量级深度神经网络的电磁信号调制识别技术[J]. 通信学报, 2020, 41(11): 12-21.

Sicheng ZHANG,Yun LIN,Ya TU,Shiwen Mao. Electromagnetic signal modulation recognition technology based on lightweight deep neural network[J]. Journal on Communications, 2020, 41(11): 12-21.

图/表 12

图1

图2

图3

表1

表2

图4

图5

图6

图7

图8

图9

图10

参考文献 25

[1]	YOU X H , WANG C X , HUANG J ,et al. Towards 6G wireless communication networks:Vision,enabling technologies,and new paradigm shifts[J]. SCIENCE CHINA Information Sciences, 2021,64(1): 1-84.
[2]	张平, 牛凯, 田辉 ,等. 6G 移动通信技术展望[J]. 通信学报, 2019,40(1): 141-148.
	ZHANG P , NIU K , TIAN H ,et al. Technology prospect of 6G mobile communications[J]. Journal on Communications, 2019,40(1): 141-148.
[3]	JAGANNATH A , JAGANNATH J , MELODIA T . Redefining wireless communication for 6G:signal processing meets deep learning[J]. arXiv Preprint,arXiv:2004.10715, 2020
[4]	LI B , LI S , HOU J ,et al. A Bayesian approach for adaptively modulated signals recognition in next-generation communications[J]. IEEE Transactions on Signal Processing, 2015,63(16): 4359-4372.
[5]	WANG D S , ZHANG M , LI Z ,et al. Modulation format recognition and OSNR estimation using CNN-based deep learning[J]. IEEE Photonics Technology Letters, 2017,29(19): 1667-1670.
[6]	查雄, 彭华, 秦鑫 ,等. 基于多端卷积神经网络的调制识别方法[J]. 通信学报, 2019,40(11): 30-37.
	ZHA X , PENG H , QIN X ,et al. Modulation recognition method based on muti-inputs convolution neural network[J]. Journal on Communications, 2019,40(11): 30-37.
[7]	PENG S L , JIANG H Y , WANG H X ,et al. Modulation classification based on signal constellation diagrams and deep learning[J]. IEEE Transactions on Neural Networks and Learning Systems, 2018,30(3): 718-727.
[8]	吴佩军, 侯进, 吕志良 ,等. 基于卷积神经网络的多进制相位调制信号识别算法[J]. 计算机应用与软件, 2019,36(11): 202-209.
	WU P J , HOU J , LYU Z L ,et al. Multi-phase modulation signal recognition algorithm based on convolutional neural network[J]. Computer Applications and Software, 2019,36(11): 202-209.
[9]	JI X F , WANG J , LI Y ,et al. Modulation recognition in maritime multipath channels:a blind equalization-aided deep learning approach[J]. China Communication, 2020,17(3): 12-25.
[10]	XU Y , LI D Z , WANG Z Y ,et al. A deep learning method based on convolutional neural network for automatic modulation classification of wireless signals[J]. Wireless Networks, 2019.25(7): 3735-3746.
[11]	桂冠, 王禹, 黄浩 . 基于深度学习的物理层无线通信技术：机遇与挑战[J]. 通信学报, 2019,40(2): 19-23.
	GUI G , WANG Y , HUANG H . Deep learning based physical layer wireless communication techniques:opportunities and challenges[J]. Journal on Communications, 2019,40(2): 19-23.
[12]	TU Y , LIN Y . Deep neural network compression technique towards efficient digital signal modulation recognition in edge device[J]. IEEE Access, 2019,7: 58113-58119.
[13]	HU H , PENG R , TAI Y W ,et al. Network trimming:a data-driven neuron pruning approach towards efficient deep architectures[J]. arXiv Preprint,arXiv:1607.03250, 2016
[14]	LIN Y , TU Y , DOU Z . An improved neural network pruning technology for automatic modulation classification in edge devices[J]. IEEE Transactions on Vehicular Technology, 2020,69(5): 5703-5706.
[15]	O’SHEA T J , CORGAN J , CLANCY T C . Convolutional radio modulation recognition net-works[C]// International Conference on Engineering Applications of Neural Networks. Piscataway:IEEE Press, 2016: 213-222.
[16]	陶冠宏, 廖开升, 周林 . 一种基于深度学习的电磁信号建模与调制识别新方法[J]. 电子信息对抗技术, 2019,34(5): 10-15.
	TAO G H , LIAO K S , ZHOU L . A novel signal modeling and modulation classification method with deep learning[J]. Electronic Information Countermeasure Technology, 2019,34(5): 10-15.
[17]	周鑫, 何晓新, 郑昌文 . 基于图像深度学习的无线电信号识别[J]. 通信学报, 2019,40(7): 114-125.
	ZHOU X , HE X X , ZHENG C W . Radio signal recognition based on image deep learning[J]. Journal on Communications, 2019,40(7): 114-125.
[18]	DETTMERS T . 8 bit approximations for parallelism in deep learning[J]. arXiv Preprint,arXiv:1511.04561, 2015
[19]	GUPTA S , AGRAWAL A , GOPALAKRISHNAN K ,et al. Deep learning with limited numerical precision[C]// International Conference on Machine Learning. New York:ACM Press, 2015: 1737-1746.
[20]	ZHU C , HAN S , MAO H ,et al. Trained ternary quantization[J]. arXiv Preprint,arXiv:1612.01064, 2016
[21]	COURBARIAUX M , BENGIO Y , DAVID J P . Binaryconnect:training deep neural networks with binary weights during propagations[C]// Advances in Neural Information Processing Systems. New York:ACM Press, 2015: 3123-3131.
[22]	QIN H , GONG R , LIU X ,et al. Binary neural networks:a survey[J]. Pattern Recognition, 2020,105: 1737-1746.
[23]	KRIZHEVSKY A , SUTSKEVER I , HINTON G E . Imagenet classification with deep convolutional neural networks[C]// Proceedings of the 25th International Conference on Neural Information Processing Systems. Massachusetts:MIT Press, 2012: 1097-1105.
[24]	ZHANG J , PAN Y , YAO T ,et al. DABNN:a super fast inference framework for binary neural networks on arm devices[C]// Proceedings of the 27th ACM International Conference on Multimedia. New York:ACM Press, 2019: 2272-2275.
[25]	O'SHEA T J , ROY T , CLANCY T C . Over-the-air deep learning based radio signal classification[J]. IEEE Journal of Selected Topics in Signal Processing, 2018,12(1): 168-179.

基于轻量级深度神经网络的电磁信号调制识别技术

Electromagnetic signal modulation recognition technology based on lightweight deep neural network

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 25

相关文章 14

Metrics

推荐阅读 0

[1]	江沸菠, 彭于波, 董莉. 面向6G的深度图像语义通信模型[J]. 通信学报, 2023, 44(3): 198-208.
[2]	王晓云, 张小舟, 马良, 王亚娟, 楼梦婷, 姜涛, 金婧, 王启星, 刘光毅. 6G通信感知一体化网络的感知算法研究与优化[J]. 通信学报, 2023, 44(2): 219-230.
[3]	杨静雅, 唐晓刚, 周一青, 刘玲, Jiangzhou Wang. 意图抽象与知识联合驱动的6G内生智能网络架构[J]. 通信学报, 2023, 44(2): 12-26.
[4]	张海君, 陈安琪, 李亚博, 隆克平. 6G移动网络关键技术[J]. 通信学报, 2022, 43(7): 189-202.
[5]	廖建新, 付霄元, 戚琦, 王敬宇, 孙海峰. 6G-ADM：基于知识空间的6G网络管控体系[J]. 通信学报, 2022, 43(6): 3-15.
[6]	王志勤, 江甲沫, 刘沛西, 曹晓雯, 李阳, 韩凯峰, 杜滢, 朱光旭. 6G联邦边缘学习新范式：基于任务导向的资源管理策略[J]. 通信学报, 2022, 43(6): 16-27.
[7]	李昂, 陈建新, 魏昕, 周亮. 面向6G的跨模态信号重建技术[J]. 通信学报, 2022, 43(6): 28-40.
[8]	刘传宏, 郭彩丽, 杨洋, 陈九九, 朱美逸, 孙鲁楠. 面向智能任务的语义通信：理论、技术和挑战[J]. 通信学报, 2022, 43(6): 41-57.
[9]	唐盼, 林佳欣, 张建华, 田磊, 常钊玮, 夏亮, 王启星. 面向6G的太赫兹信道反射特性研究[J]. 通信学报, 2022, 43(5): 102-109.
[10]	张晓茜, 徐勇军. 面向零功耗物联网的反向散射通信综述[J]. 通信学报, 2022, 43(11): 199-212.
[11]	陶梅霞, 王栋, 孙瑞, 张乃夫. 联邦学习中基于时分多址接入的用户调度策略[J]. 通信学报, 2021, 42(6): 1-29.
[12]	周振宇, 贾泽晗, 廖海君, 赵雄文, 张磊. 基于上下文学习的电力物联网接入控制方法[J]. 通信学报, 2021, 42(3): 150-159.
[13]	张平, 牛凯, 田辉, 聂高峰, 秦晓琦, 戚琦, 张娇. 6G移动通信技术展望[J]. 通信学报, 2019, 40(1): 141-148.
[14]	耿绥燕,李杏,王琦,王光波,王蒙军,孙韶辉,洪伟,赵雄文. 26GHz室内毫米波人体阻挡衰减特性研究[J]. 通信学报, 2016, 37(11): 68-73.

W_a	W_b	W _aW_b
1	1	1
1	-1	-1
-1	1	-1
-1	-1	1

V_a	V_b	V _a:V_b
1	1	1
1	0	0
0	1	0
0	0	1