基于循环谱截面智能分析的混合信号调制识别方法

doi:10.16182/j.issn1004731x.joss.22-0239

摘要/Abstract

摘要：

针对已有混合信号识别方法存在智能化程度低、适应性差等问题，提出了一种基于循环谱截面和深度学习相结合的智能识别方法。理论推导分析了常见混合通信信号的循环谱零谱频率截面特征；利用提出的非线性分段映射和指向性伪聚类新方法对上述截面图进行预处理特征增强，提高了截面特征的适应性和一致性；并将预处理后的特征图与经典残差网络相结合，利用深度学习网络对特征图中调制信息的深层次细节挖掘分析能力，实现了混合信号的有效识别。仿真结果表明，该方法对噪声不敏感，当信噪比不低于-2 dB时，平均识别率大于90%；且该方法对信号参数及信号间能量比变化有较好的适应能力。

关键词: 通信信号处理, 深度学习, 调制识别, 循环谱, 残差网络

Abstract:

Aiming at the problems of low intelligence and poor adaptability for the existing mixed signal recognition methods, an intelligent recognition method based on cyclic spectral cross section and deep learning is proposed. For common mixed communication signals, the characteristics of zero frequency cross section of cyclic spectrum are theoretically deduced and analyzed. Two new pre-processing methods, nonlinear segmental mapping and directional pseudo-clustering are proposed, which can effectively improve the adaptability and consistency of cross section features. The pre-processed feature graph is combined with the residual network (ResNet), and the deep learning network is used to mine and analyze the deep details of modulation information in the feature graph, and the effective recognition of mixed signals is realized. Simulation results show that the method is insensitive to noise, and the average recognition rate is more than 90% when the SNR is no less than -2dB. The proposed method has good adaptability to the variation of signal parameters and energy ratio between signals.

Key words: communication signal processing, deep learning, modulation recognition, cyclic spectrum, residual network

中图分类号:

TN911.72

杜宇, 杨新权, 张建华, 袁素春, 肖化超, 袁晶晶. 基于循环谱截面智能分析的混合信号调制识别方法[J]. 系统仿真学报, 2023, 35(1): 146-157.

Yu Du, Xinquan Yang, Jianhua Zhang, Suchun Yuan, Huachao Xiao, Jingjing Yuan. Modulation Recognition Method of Mixed Signal Based on Intelligent Analysis of Cyclic Spectrum Section[J]. Journal of System Simulation, 2023, 35(1): 146-157.

图/表 12

图1

图2

表1

混合信号不同频点循环谱理论值

序号	循环频率	类型	BPSK+QPSK			BPSK+OQPSK			QPSK+OQPSK
1	$α = n T$	理论值	0			0			0
		最大值归一化幅值	$E B : E Q$	1 $:$ 1	0	$E B : E O$	1:1	0	$E Q : E O$	1:1	0
				1:4	0		1:4	0		1:4	0
				4:1	0		4:1	0		4:1	0
2	$α = 2 f c$	理论值	$T E B / 4$			$T E B / 4$			0
		最大值归一化幅值	$E B : E Q$	1:1	1.00	$E B : E O$	1:1	1.00	$E Q : E O$	1:1	0
				1:4	1.00		1:4	0.49		1:4	0
				4:1	1.00		4:1	1.00		4:1	0
3	$α = 2 f c ± 1 T$	理论值	$T E B / π 2$			$T E B / π 2 + T E O / π 2$			$T E O / π 2$
		最大值归一化幅值	$E B : E Q$	1:1	0.41	$E B : E O$	1:1	0.81	$E Q : E O$	1:1	1.00
				1:4	0.41		1:4	1.00		1:4	1.00
				4:1	0.41		4:1	0.51		4:1	1.00
4	$α = 2 f c ± 2 T$	理论值	0			0			—
		最大值归一化幅值	$E B : E Q$	1:1	0	$E B : E O$	1:1	0	$E Q : E O$	1:1	—
				1:4	0		1:4	0		1:4	—
				4:1	0		4:1	0		4:1	—
5	$α = 2 f c ± 3 T$	理论值	$T E B / 9 π 2$			$T E B / 9 π 2 + T E O / 9 π 2$			$T E O / 9 π 2$
		最大值归一化幅值	$E B : E Q$	1:1	0.05	$E B : E O$	1:1	0.10	$E Q : E O$	1:1	0.11
				1:4	0.10		1:4	0.14		1:4	0.10
				4:1	0.05		4:1	0.06		4:1	0.11
6	$α ≥ 2 f c + 4 T$ $α ≤ 2 f c - 4 T$	理论值	0			0			0
		最大值归一化幅值	$E B : E Q$	1:1	0	$E B : E O$	1:1	0	$E Q : E O$	1:1	0
				1:4	0		1:4	0		1:4	0
				4:1	0		4:1	0		4:1	0

表1

图3

图4

图5

图6

图 7

图8

图9

图10

图11

参考文献 18

1	林冲, 闫文君, 张立民, 等. 通信信号调制识别综述[J]. 中国电子科学研究院学报, 2021, 16(11): 1074-1085.
	Lin Chong, Yan Wenjun, Zhang Limin, et al. An Overview of Communication Signals Modulation Recognition[J]. Journal of CAEIT, 2021, 16(11): 1074-1085.
2	赵宇峰, 曹玉健, 纪勇, 等. 基于循环频率特征的单信道混合通信信号的调制识别[J]. 电子与信息学报, 2014, 36(5): 1202-1208.
	Zhao Yufeng, Cao Yujian, Ji Yong, et al. Modulation Identification for Single-Channel Mixed Communication Signals Based on Cyclic Frequency Features[J]. Journal of Electronics & Information Technology, 2014, 36(5): 1202-1208.
3	张洋, 彭华. 单通道混合信号调制识别[J]. 信息工程大学学报, 2016, 17(6): 662-668, 712.
	Zhang Yang, Peng hua. Modulation Recognition for Mixed Signals in Single Channniel[J]. Journal of Information Engineering University, 2016, 17(6): 662-668, 712.
4	曾创展, 贾鑫. 一种单信道时频重叠多进制数字相位调制信号调制识别方法[J]. 科学技术与工程, 2016, 16(5): 172-176.
	Zeng Chuangzhan, Jia Xin. A Modulation Recognition Method for Time-Frequency Overlapped MPSK Signals in the Single Channel[J]. Science Technology and Engineering, 2016, 16(5): 172-176.
5	陈思南, 龚晓峰, 安冬冬. 利用高阶累积量的时频重叠MPSK信号调制识别[J]. 科学技术与工程, 2017, 17(36): 195-199.
	Chen Sinan, Gong Xiaofeng, An Dongdong. Modulation Identification of Time-Frequency Overlapped MPSK Signal Using High Order Cumulant[J]. Science Technology and Engineering, 2017, 17(36): 195-199.
6	Huang S, Yao Y, Wei Z, et al. Automatic Modulation Classification of Overlapped Sources Using Multiple Cumulants[J]. IEEE Transactions on Vehicular Technology (S0018-9545), 2017, 66(7): 6089-6101.
7	陈思南, 龚晓峰, 陈利波. 共信道时频重叠的数字信号调制识别方法[J]. 计算机工程与设计, 2018, 39(3): 612-616, 622.
	Chen Sinan, Gong Xiaofeng, Chen Libo. Method of Digital Signal Modulation Recognition Based on Co-channel Time-Frequency Overlap[J]. Computer Engineering and Design, 2018, 39(3): 612-616, 622.
8	吴娟, 谢跃雷. 基于分数阶小波变换的OFDM混合信号识别方法[J]. 电视技术, 2020, 44(11): 29-37.
	Wu Juan, Xie Yuelei.Recognition Method of OFDM Mixed Signal Based on Fractional Wavelet Transform[J]. Video Engineering, 2020, 44(11): 29-37.
9	张承畅, 余洒, 徐余, 等. 人工神经网络在调制识别中的应用综述[J]. 重庆邮电大学学报(自然科学版), 2022, 34(2): 181-192.
	Zhang Chengchang, Yu Sa, Xu Yu, et al. Survey of Artificial Neural Network Applied to Modulation Recognition[J]. Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition), 2022, 34(2): 181-192 .
10	郭业才, 姚文强. 基于信噪比分类网络的调制信号分类识别算法[J]. 电子与信息学报, 2022, 44(10): 3507-3515.
	Guo Yecai, Yao Wenqiang. Modulation Signal Classification and Recognition Algorithm Based on Signal to Noise Ratio Classification Network[J]. Journal of Electronics & Information Technology, 2022, 44(10): 3507-3515.
11	Ahmed K A, Ergun E. Automatic Modulation Classification Using Different Neural Network and PCA Combinations[EB/OL]. [2021-12-28]. .
12	Peng Shengliang, Sun Shujun, Yao Yudong. A Survey of Modulation Classification Using Deep Learning: Signal Representation and Data Preprocessing[J]. IEEE Transactions on Neural Networks and Learning Systems (S1045-9227), 2021, 99: 1-19.
13	郭业才, 张浩然. 基于改进LDA和自编码器的调制识别算法[J]. 系统仿真学报, 2021, 33(2): 494-500.
	GuoYecai, Zhang Haoran. Modulation Recognition Algorithm Based on Improved LDA and Autoencoders[J]. Journal of System Simulation, 2021, 33(2): 494-500.
14	郭业才, 王庆伟. 基于截断迁移与并行残差网络的调制识别算法[J/OL]. 系统仿真学报[2021-12-28]. .
	Guo Yecai, Wang Qingwei. Modulation Recognition Algorithm Based on Truncated Migration and Parallel ResNet[J/OL]. Journal of System Simulation[2021-12-28]. .
15	Zhou S Y, Wu Z L, Yin Z D, et al. Blind Modulation Classification for Overlapped Co-channel Signals Using Capsule Networks[J]. IEEE Communications Letters (S1089-7798), 2019, 23(10): 1849-1852.
16	董聪, 张传武, 高勇. 基于残差神经网络的通信混合信号识别[J]. 无线电工程, 2020, 50(9): 727-731.
	Dong Cong, Zhang Chuanwu, Gao Yong. Mixed Communication Signal Recognition Based on Residual Neural Network[J]. Radio Engineering, 2020, 50(9): 727-731.
17	Wang L J, Han Y, Ge H F, et al. Modulation Recognition of Composite Signal Based on ResNet and Frequency Domain Graph[C]// 2021 3rd International Conference on Artificial Intelligence and Computer Science (AICS). Beijing, 2021: 1-11.
18	Yin Z D, Zhang R, Wu Z L, et al. Co-channel Multi-signal Modulation Classification Based on Convolution Neural Network[C]// 2019 IEEE 89th Vehicular Technology Conference. Kuala Lumpur, Malaysia: IEEE, 2019: 1-5.

[1]	张大永, 杨镜宇, 吴曦. 兵棋推演空中任务智能预测方法研究[J]. 系统仿真学报, 2023, 35(1): 212-220.
[2]	柴慧敏, 张勇, 李欣粤, 宋雅楠. 基于深度学习的空中目标威胁评估方法[J]. 系统仿真学报, 2022, 34(7): 1459-1467.
[3]	康旭, 张晓峰. 基于生成对抗神经网络的雷达遥感数据增广方法[J]. 系统仿真学报, 2022, 34(4): 920-927.
[4]	刘家瑞, 杨国田, 王孝伟. 基于孪生深度神经网络的风电机组故障诊断方法[J]. 系统仿真学报, 2022, 34(11): 2348-2358.
[5]	陆荣秀, 张笔豪, 莫振龙. 基于脸部特征和头部姿态的疲劳检测方法[J]. 系统仿真学报, 2022, 34(10): 2279-2292.
[6]	仝卫国, 曾世超, 张立峰, 侯哲, 郭佳跃. 基于深度残差神经网络的电阻层析成像及流型辨识方法[J]. 系统仿真学报, 2022, 34(09): 2028-2036.
[7]	林硕, 安磊, 高治军, 单丹, 尚文利. 结合栈式自编码及长短时记忆的入侵检测研究[J]. 系统仿真学报, 2021, 33(6): 1288-1296.
[8]	程文聪, 史小康, 王志刚. 基于生成对抗网络的仿真卫星云图生成方法[J]. 系统仿真学报, 2021, 33(6): 1297-1306.
[9]	董书琴, 张斌. 面向不平衡数据的网络流量异常检测方法[J]. 系统仿真学报, 2021, 33(3): 679-689.
[10]	王霄汉, 张霖, 任磊, 谢堃钰, 王昆玉, 叶飞, 陈真. 基于强化学习的车间调度问题研究简述[J]. 系统仿真学报, 2021, 33(12): 2782-2791.
[11]	王步维, 王敏, 范谦, 王雅男, 章涵文, 乐云亮. 基于深度学习的晶体性质预测研究[J]. 系统仿真学报, 2021, 33(12): 2854-2863.
[12]	冯晓, 张辉, 周蕊, 乔璐, 魏东, 李丹丹, 张玉尧, 郑国清. 基于深度学习和籽粒双面特征的玉米品种识别[J]. 系统仿真学报, 2021, 33(12): 2983-2991.
[13]	杜金莲, 李淑飞, 金雪云. 三维烟雾流场超分辨率数据生成网络模型的研究[J]. 系统仿真学报, 2021, 33(10): 2381-2389.
[14]	刘瑞军, 王向上, 张晨, 章博华. 基于深度学习的视觉SLAM综述[J]. 系统仿真学报, 2020, 32(7): 1244-1256.
[15]	阴敬方, 朱登明, 石敏, 王兆其. 基于引导对抗网络的人体深度图像修补方法[J]. 系统仿真学报, 2020, 32(7): 1312-1321.