基于SAMME+ResNet的多相码信号识别方法

doi:10.3969/j.issn.1001-506X.2020.10.12

摘要/Abstract

摘要：

针对传统多相码信号识别方法在低信噪比情况下分类精度不高、类识别率不均衡和识别方法不具有通用性的特点,提出了一种利用集成学习中的多类指数损失函数逐步添加模型(stagewise additive modeling using a multi-class exponential loss function, SAMME)算法和残差神经网络(residual neural network, ResNet)的多相码信号识别方法。通过仿真实验对5类多相码信号进行了分类识别,验证了模型的有效性,分析了不同数量基学习器对模型的影响,最后与传统分类方法进行了对比。仿真结果表明,在信噪比低于6 dB的情况下,所提方法相对于单个残差网络提高了约10%的分类精度,同时缩小了类之间识别率的差距,相对于常用的分类方法也有很大的优势。

关键词: 多相码, 信号识别, 集成学习, 残差神经网络

Abstract:

In view of the characteristics of the traditional polyphase code signal recognition methods, such as low classification accuracy, uneven class recognition rate and non-generality of recognition methods in the case of low signal to noise ratio (SNR), a polyphase code signal recognition method based on stagewise additive modeling using a multi-class exponential loss function (SAMME) algorithm in ensemble learning and residual neural network (ResNet) is proposed. Simulation experiments are carried out to classify and identify five kinds of polyphase code signals, and the validity of the model is verified. The influence of different quantity base learners on the model is analyzed. Finally, the proposed method is compared with the traditional classification methods. Simulation results show that when the SNR is lower than 6 dB, the proposed method improves the classification accuracy by about 10% compared with single residual network and reduces the difference of recognition rate between classes and also has great advantages over common classification methods.

Key words: polyphase code, signal recognition, ensemble learning, residual neural network (ResNet)

中图分类号:

TN971.1

孙艺聪, 田润澜, 董会旭, 孙亮. 基于SAMME+ResNet的多相码信号识别方法[J]. 系统工程与电子技术, 2020, 42(10): 2239-2245.

Yicong SUN, Runlan TIAN, Huixu DONG, Liang SUN. Polyphase code signal recognition method based on SAMME+ResNet[J]. Systems Engineering and Electronics, 2020, 42(10): 2239-2245.

图/表 9

图1

图2

表1

图3

表2

图4

图5

图6

表3

参考文献 24

1	WAN J , YU X , GUO Q . LPI radar waveform recognition based on CNN and TPOT[J]. Symmetry, 2019, 11 (5): 725-1- 725-15.
2	KISHORE T R , RAO K D . Automatic intrapulse modulation classification of advanced LPI radar waveforms[J]. IEEE Trans.on Aerospace and Electronic Systems, 2017, 53 (2): 901- 914. doi: 10.1109/TAES.2017.2667142
3	DAVIS R M , FANTE R L , PERRY R P , et al. Phase-coded waveforms for radar[J]. IEEE Trans.on Aerospace and Electronic Systems, 2007, 43 (1): 401- 408. doi: 10.1109/TAES.2007.357142
4	邹梦然, 陈永游, 曾德国, 等. 基于相位差分算法的多相编码信号实时脉内识别技术[J]. 航天电子对抗, 2015, 31 (2): 56- 60, 64.
	ZOU M R , CHEN Y Y , ZENG D G , et al. Real-time intra-pulse identification of multi-phase coded signals based on phase diffe-rence algorithm[J]. Aerospace Electronic Warfare, 2015, 31 (2): 56- 60, 64.
5	王泽众, 曹万平, 刘锋, 等. 基于周期Wigner-Hough变换的多相编码连续波雷达信号检测算法[J]. 海军航空工程学院学报, 2012, 27 (6): 605- 612.
	WANG Z Z , CAO W P , LIU F , et al. Detection algorithm of multiphase coded continuous wave radar signal based on periodic wigner-hough transform[J]. Journal of Naval Aeronautical and Astronautical University, 2012, 27 (6): 605- 612.
6	王冰.多相码雷达信号脉内特征分析方法研究[D].哈尔滨:哈尔滨工程大学, 2019.
	WANG B. Study on the method of intra-pulse characteristic analysis of multi-phase code radar signal[D]. Harbin: Harbin Engineering University, 2019.
7	BRITO C , MACHADO A , SOUSA A , et al. Electrocardiogram beat-classification based on a ResNet network[J]. Studies in Health Technology and Informatics, 2019, 264, 55- 59.
8	YU L , WANG S , LAI K K . Credit risk assessment with a multistage neural network ensemble learning approach[J]. Expert Systems with Applications, 2008, 34 (2): 1434- 1444. doi: 10.1016/j.eswa.2007.01.009
9	HAN X H , WANG L H , XU S J , et al. Recognizing roles of online illegal gambling participants: an ensemble learning approach[J]. Computers & Security, 2019, 87, 101588.
10	ARAQUE O , CORCUEAR-PLATAS I , SANCHEZ-RADA J F , et al. Enhancing deep learning sentiment analysis with ensemble techniques in social applications[J]. Expert Systems with Application, 2017, 77, 236- 246. doi: 10.1016/j.eswa.2017.02.002
11	YU L , ZHOU R T , TANG L , et al. A DBN-based resampling SVM ensemble learning paradigm for credit classification with imbalanced data[J]. Applied Soft Computing, 2018, 69, 192- 202. doi: 10.1016/j.asoc.2018.04.049
12	SADAM A A , EL-SAYED M E A . Using word embedding and ensemble learning for highly imbalanced data sentiment analysis in short arabic text[J]. Procedia Computer Science, 2017, 109, 359- 366. doi: 10.1016/j.procs.2017.05.365
13	黄颖坤, 金炜东, 余志斌, 等. 基于深度学习和集成学习的辐射源信号识别[J]. 系统工程与电子技术, 2018, 40 (11): 2420- 2425.
	HUANG Y K , JIN W D , YU Z B , et al. Radiation source signal recognition based on deep learning and integrated learning[J]. Systems Engineering and Electronics Technology, 2018, 40 (11): 2420- 2425.
14	LIU T , GUAN Y Y , LIN Y . Research on modulation recognition with ensemble learning[J]. EURASIP Journal on Wireless Communications and Networking, 2017, 2017 (1): 179.
15	ZHANG Z , LI Y B , JIN S S , et al. Modulation signal recognition based on information entropy and ensemble learning[J]. Entropy, 2018, 20 (3): 198. doi: 10.3390/e20030198
16	徐继伟, 杨云. 集成学习方法:研究综述[J]. 云南大学学报(自然科学版), 2018, 40 (6): 1082- 1092.
	XU J W , YANG Y . Integrated learning methods: research review[J]. Journal of Yunnan University(Natural Science Edition), 2018, 40 (6): 1082- 1092.
17	杨新武, 马壮, 袁顺. 基于弱分类器调整的多分类Adaboost算法[J]. 电子与信息学报, 2016, 38 (2): 373- 380.
	YANG X W , MA Z , YUAN S . Adaboost algorithm for multi-classification based on weak classifier adjustment[J]. Journal of Electronics & Information Techology, 2016, 38 (2): 373- 380.
18	ZHU J , ZOU H , ROSSET S , et al. Multi-class adaboost[J]. Statistics and Its Interface, 2009, 2 (3): 349- 360. doi: 10.4310/SII.2009.v2.n3.a8
19	GAO Y Z, RONG W G, SHEN Y K, et al. Convolutional neural network based sentiment analysis using Adaboost combination[C]//Proc.of the International Joint Conference on Neural Networks, 2016.
20	YANG S, CHEN L F, YAN T, et al. An ensemble classification algorithm for convolutional neural network based on AdaBoost[C]//Proc.of the IEEE/ACIS 16th International Confe-rence on Computer and Information Science, 2017.
21	HE K W, ZHANG X Y, REN S, et al. Deep residual learning for image recognition[C]//Proc.of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
22	王青媛, 胡婧, 谢智东, 等. 基于形态学预处理的时频检测及识别算法[J]. 军事通信技术, 2015, 36 (3): 77- 82, 98.
	WANG Q Y , HU J , XIE Z D , et al. Time-frequency detection and recognition algorithm based on morphological preprocessing[J]. Journal of Military Communications Technology, 2015, 36 (3): 77- 82, 98.
23	熊坤来, 罗景青, 吴世龙. 基于时频图像和神经网络的LPI雷达信号调制识别[J]. 弹箭与制导学报, 2011, 31 (5): 230- 233.
	XIONG K L , LUO J Q , WU S L . Modulation recognition of LPI radar signal based on time-frequency image and neural network[J]. Journal of Projectile and Arrow Guidance, 2011, 31 (5): 230- 233.
24	VANHOY G , SCHUCKER T , BOSE T . Classification of LPI radar signals using spectral correlation and support vector machines[J]. Analog Integrated Circuits & Signal Processing, 2017, 91 (2): 305- 313.

层名	结构
卷积层	3×3, 16
残差网络层1	$ \left[ \begin{array}{l}3 \times 3, 16\\3 \times 3, 16\end{array} \right] \times 3$
残差网络层2	$ \left[ \begin{array}{l}3 \times 3, 32\\3 \times 3, 32\end{array} \right] \times 3$
残差网络层3	$ \left[ \begin{array}{l}3 \times 3, 64\\3 \times 3, 64\end{array} \right] \times 3$
池化输出层	平均池化, softmax,全连接

模型	集成精度/%	最大基学习器精度/%	精度增加量/%	训练时间/min:s
ResNet 20	87.667	-	-	03:46
SAMME+ResNet 20(2)	89.333	87.933	1.400	07:17
SAMME+ResNet 20(3)	93.111	90.667	2.444	10:51
SAMME+ResNet 20(4)	92.667	91.133	1.534	14:30

分类方法	时间
文献[23]方法	01′32″
文献[24]方法	19′23″
VGG-19	06′59″
GoogleNet	08′24″
本文方法	10′51″

[1]	曹鹏宇, 杨承志, 石礼盟, 吴宏超. 基于PSO-DBSCAN和SCGAN的未知雷达信号处理方法[J]. 系统工程与电子技术, 2022, 44(4): 1158-1165.
[2]	吕勤哲, 全英汇, 沙明辉, 董淑仙, 邢孟道. 基于集成深度学习的有源干扰智能分类[J]. 系统工程与电子技术, 2022, 44(12): 3595-3602.
[3]	徐桂光, 王旭东, 汪飞, 胡国兵, 高涌荇, 罗泽虎. 基于SE-ResNeXt网络的低信噪比LPI雷达辐射源信号识别[J]. 系统工程与电子技术, 2022, 44(12): 3676-3684.
[4]	唐玺博, 张立民, 钟兆根. 基于ADASYN与改进残差网络的入侵流量检测识别[J]. 系统工程与电子技术, 2022, 44(12): 3850-3862.
[5]	朱立为, 黄知涛. 多载波OFDM信号识别方法[J]. 系统工程与电子技术, 2022, 44(11): 3522-3528.
[6]	马骏, 杨镜宇, 邹立岩. 基于Stacking集成元模型的作战体系能力图谱生成方法[J]. 系统工程与电子技术, 2022, 44(1): 154-163.
[7]	丁嘉辉, 汤建龙, 于正洋. 轻量化的增量式集成学习算法设计[J]. 系统工程与电子技术, 2021, 43(4): 861-867.
[8]	谢存祥, 张立民, 钟兆根. 基于时频特征提取和残差神经网络的雷达信号识别[J]. 系统工程与电子技术, 2021, 43(4): 917-926.
[9]	史蕴豪, 许华, 郑万泽, 刘英辉. 基于集成学习与特征降维的小样本调制识别方法[J]. 系统工程与电子技术, 2021, 43(4): 1099-1109.
[10]	高诗飏, 董会旭, 田润澜, 张歆东. 基于SRNN+Attention+CNN的雷达辐射源信号识别方法[J]. 系统工程与电子技术, 2021, 43(12): 3502-3509.
[11]	孙艺聪, 田润澜, 王晓峰, 董会旭, 戴普. 基于改进CLDNN的辐射源信号识别[J]. 系统工程与电子技术, 2021, 43(1): 42-47.
[12]	马愈昭, 王强强, 王瑞松, 熊兴隆. 基于SVD和MPSO-SVM的光纤周界[J]. 系统工程与电子技术, 2020, 42(8): 1652-1661.
[13]	李亚兰, 金炜东, 葛鹏. 基于VMD和特征融合的辐射源信号识别[J]. 系统工程与电子技术, 2020, 42(7): 1499-1503.
[14]	张怡霄, 郭文普, 康凯, 姚云龙, 王攀. 基于聚类和时序相关的重点雷达信号快速识别[J]. 系统工程与电子技术, 2020, 42(3): 597-602.
[15]	弋佳东, 杨洁. 基于IFOA-SA-BP神经网络的雷达信号识别[J]. 系统工程与电子技术, 2020, 42(12): 2735-2741.