基于高维特征选择的跳频电台细微特征识别

doi:10.3969/j.issn.1001-506X.2020.02.25

摘要/Abstract

摘要：

将高维特征用于跳频电台细微特征个体识别具有很大优势,为了增强对跳频电台的分类识别能力,需要增加特征类型和维数,提高特征集的表征能力,但同时会引入大量冗余特征,导致分类器计算时间过长,分类正确率降低。为了降低高维特征集维数,首先采用相关性快速过滤特征选择算法,删除高维特征集中的不相关冗余特征,得到最优特征集。然后利用经过参数优化的支持向量机(support vector machine, SVM)分类器进行训练分类。实验表明,所提算法能够对高维特征集进行合理的降维,提高了SVM的分类器的分类性能,在保证分类正确率的基础上,降低了运算量,提高了跳频电台细微特征识别的时效性。

关键词: 跳频电台, 细微特征, 特征选择, 支持向量机

Abstract:

The high dimensional feature is used for individual identification of the fine features of the frequency hopping station. In order to enhance the classification and recognition ability of the frequency hopping station, it is usually necessary to increase the feature type and feature dimension of the feature set to improve the representation ability. However, many redundant features are introduced. As a result, the calculation time of the classifier is too long, and the classification correctness rate is lowered. In order to reduce the dimension of high-dimensional feature sets, the feature selection algorithm is firstly used to delete the irrelevant redundant features in the high-dimensional feature set to obtain the optimal feature set. Then, the parameter-optimized support vector machine (SVM) classifier is used for training and classification. Experiments show that the proposed algorithm can reduce the dimensionality of high-dimensional feature sets and improve the classification performance of SVM. On the basis of ensuring the correctness rate of classification, the computational complexity is reduced, and the timeliness of fine feature recognition of frequency hopping stations is improved.

Key words: frequency hopping radio, subtle features, feature selection, support vector machine (SVM)

中图分类号:

TN911.7

李红光, 郭英, 眭萍, 齐子森, 苏令华. 基于高维特征选择的跳频电台细微特征识别[J]. 系统工程与电子技术, 2020, 42(2): 445-451.

Hongguang LI, Ying GUO, Ping SUI, Zisen QI, Linghua SU. Fine feature recognition of frequency hopping radio based on high dimensional feature selection[J]. Systems Engineering and Electronics, 2020, 42(2): 445-451.

图/表 7

图1

图2

图3

表1

表2

图4

图5

参考文献 30

1	徐启华.信号侦察中的跳频信号识别方法[D].西安,西安电子科技大学, 2005.
	XU Q H. Recognition method for frequency hopping in signal reconnaissance[D]. Xi'an: Xidian University, 2005.
2	ERIC M , DUKIC M L , OBRADOVIC M . Frequency hopping signal separation by spatio-frequency analysis based on MUSIC method[J]. Spread Spectrum Techniques and Applications, 2000, 66 (1): 78- 82.
3	王斌, 陈秋华, 王翠柏. 基于聚类的跳频信号分选[J]. 北京邮电大学学报, 2009, 32 (2): 80- 84. doi: 10.3969/j.issn.1007-5321.2009.02.018
	WANG B , CHEN Q H , WANG C B . Identification of frequency hopping signals based on clustering[J]. Journal of Beijing University of Posts and Telecommunications, 2009, 32 (2): 80- 84. doi: 10.3969/j.issn.1007-5321.2009.02.018
4	李文龙,梁涛,许金勇.跳频通信中的射频指纹识别[D].南京:解放军理工大学, 2009.
	LI W L, LIANG T, XU J Y. Radio frequency fingerprinting in frequency hopping communication [D]. Nanjing: PLA University of Science & Technology, 2009.
5	骆振兴, 赵知劲. 基于功放瞬态响应的跳频电台分选方法[J]. 杭州电子科技大学学报, 2009, 29 (1): 29- 32. doi: 10.3969/j.issn.1001-9146.2009.01.008
	LUO Z X , ZHAO Z J . Individual HF transmitter identification based on transient response of power amplifier[J]. Journal of Hangzhou Dianzi University, 2009, 29 (1): 29- 32. doi: 10.3969/j.issn.1001-9146.2009.01.008
6	范虹. 非平稳信号特征提取方法及应用[M]. 科学出版社, 2013.
	FAN H . Non-stationary signal feature extraction and application[M]. Science Press, 2013.
7	VERGARA J R , ESTÉVEZ P A . A review of feature selection methods based on mutual information[J]. Neural Computing & Applications, 2014, 24 (1): 175- 186.
8	CHANDRASHEKAR G , SAHIN F . A survey on feature selection methods[J]. Computers & Electrical Engineering, 2014, 40 (1): 16- 28.
9	HOQUE N , BHATTACHARYYA D K , KALITA J K . MIFS-ND : a mutual information-based feature selection method[J]. Expert Systems with Applications, 2014, 41 (14): 6371- 6385. doi: 10.1016/j.eswa.2014.04.019
10	ESTÉVEZ P A , TESMER M , PEREZ C A , et al. Normalized mutual information feature selection[J]. IEEE Trans.on Neural Networks, 2009, 20 (2): 189- 201. doi: 10.1109/TNN.2008.2005601
11	YU L , LIU H . Efficient feature selection via analysis of relevance and redundancy[J]. Journal of Machine Learning Research, 2004, 5 (12): 1205- 1224.
12	VAPNIK V N . The nature of statistical learning theory, NY Springer[M]. 2000: 25- 314.
13	SMOLA A, BARTLETT P, SCHÖLKOPF B, et al. Bounds on error expectation for SVM[C]//Proc.of the Advances in Large Margin Classifiers, 2008: 2000-2001.
14	DING Z, WANG F, ZHOU P. Fetal ECG extraction based on different kernel functions of SVM[C]//Proc.of the International Conference on Computer Research and Development, 2011: 205-208.
15	ZHANG Q , ZHANG P , LONG G , et al. Online learning from trapezoidal data streams[J]. IEEE Trans.on Knowledge and Data Engineering, 2016, 1- 1.
16	VINH L T , LEE S , PARK Y T , et al. A novel feature selection method based on normalized mutual information[J]. Applied Intelligence, 2012, 37 (1): 100- 120. doi: 10.1007/s10489-011-0315-y
17	LIU Y, DU J.Parameter optimization of the SVM for big data[C]// Proc.of the 8th IEEE International Symposium on Computational Intelligence and Design (ISCID), 2015: 341-344.
18	FEI Y, MIN H.Simultaneous feature with support vector selection and parameters optimization using GA-based SVM solve the binary classification[C]//Proc.of the IEEE International Conference on Computer Communication & the Internet, 2016: 426-433.
19	ZHANG J , CHEN X , XIANG Y , et al. Robust network traffic classification[J]. IEEE/ACM Trans.on Networking, 2015, 23 (4): 1257- 1270. doi: 10.1109/TNET.2014.2320577
20	XIAO Y , WANG H , XU W . Parameter selection of Gaussian Kernel for one-class SVM[J]. IEEE Trans.on Cybernetics, 2015, 45 (5): 927- 939. doi: 10.1109/TCYB.2014.2340032
21	ALAM S. Performance of Alzheimer disease classification based on PCA, linear SVM and multi-Kernel SVM[C]//Proc.of the 8th IEEE International Conference on Ubiquitous and Future Networks, 2016: 987-989.
22	GRAMA L, TUNS L, RUSU C. On the optimization of SVM Kernel parameters for improving audio classification accuracy[C]// Proc.of the IEEE International Conference on Engineering of Modern Electric Systems, 2017: 224-227.
23	GABRIEL G S, BELZARENA P. Statistical traffic classification by boosting support vector machines[C]//Proc.of the Latin American Networking Conference, 2012: 9-18.
24	GUPTA S, SAURAV S, SINGH S, et al. VLSI architecture of exponential block for non-linear SVM classification[C]//Proc.of the IEEE International Conference on Advances in Computing, 2015: 528-532.
25	YANG Q, FU H, ZHU T.An optimization method for parameters of SVM in network intrusion detection system[C]//Proc.of the IEEE International Conference on Distributed Computing in Sensor Systems (DCOSS), 2016: 136-142.
26	HUANG Q, MAO J, LIU Y. An improved grid search algorithm of SVR parameters optimization[C]//Proc.of the IEEE International Conference on Communication Technology, 2013: 1022-1026.
27	MAVROFORAKIS M E , THEODORIDIS S . A geometric approach to support vector machine (SVM) classification[J]. IEEE Trans.on Neural Networks, 2006, 17 (3): 671- 682. doi: 10.1109/TNN.2006.873281
28	HESTERMAN J Y , CAUCCI L , KUPINSKI M A , et al. Maximum-likelihood estimation with a contracting-grid search algorithm[J]. IEEE Trans.on Nuclear Science, 2010, 57 (3): 1077- 1084. doi: 10.1109/TNS.2010.2045898
29	CZARNOWSKI I , JÄDRZEJOWICZ P . Agent-based RBF network classifier with feature selection in a kernel space[J]. Journal of Cybernetics, 2016, 47 (1/2): 17- 31.
30	ZHANG S , LI X , MING Z , et al. Learning K for KNN classification[J]. ACM Trans.on Intelligent Systems & Technology, 2017, 8 (3): 43.

分类特征	特征数	最优子集	正确率/%	时间/s
谱对称性均值f₁	1	{f₁}	43.51	0.008
谱对称性方差f₂	1	{f₂}	43.63	0.008
波形特征f₃	1	{f₃}	51.72	0.013
盒维数f₄	1	{f₄}	64.55	0.013
瑞利熵f₅	1	{f₅}	65.13	0.013
信息维数f₆	1	{f₆}	63.52	0.015
LZC f₇	1	{f₇}	63.43	0.013
常规特征	7	{f₁, f₂, f₃, f₄, f₅, f₆, f₇}	72.32	0.082
FCBF选择出的特征集	4	{f₃, f₄, f₆, f₇}	72.15	0.057
本文算法选择出的特征集	3	{f₄, f₅, f₇}	72.24	0.029

分类特征	特征维数	最优子集	正确率/%	时间/s
高阶谱特征SIB f₈	256	{f₈¹, f₈², …, f₈²⁵⁶}	80.60	1.715
FCBF选择出的特征集	193	{f₈², f₈⁶, …, f₈²⁵³}	82.41	0.774
本文算法选择出的特征集	148	{f₈³, f₈⁷, …, f₈²⁴⁶}	84.63	0.459

[1]	李波, 周家豪, 刘民岷, 朱品朝. 基于改进NSGA3的焊接缺陷评估特征选择[J]. 系统工程与电子技术, 2022, 44(7): 2211-2218.
[2]	薛春岭, 曹菲, 孙庆, 秦建强, 冯晓伟. 基于多特征信息融合的海面微弱目标检测[J]. 系统工程与电子技术, 2022, 44(11): 3338-3345.
[3]	朱峰, 蒋倩倩, 林川, 杨啸. 基于支持向量机的典型宽带电磁干扰源识别[J]. 系统工程与电子技术, 2021, 43(9): 2400-2406.
[4]	曹浪财, 林晓昌, 苏思行. 基于矩阵分解和自适应图的无监督特征选择[J]. 系统工程与电子技术, 2021, 43(8): 2197-2208.
[5]	李睿峰, 许爱强, 孙伟超, 王树友. 基于元学习的航空电子设备特征选择算法推荐方法[J]. 系统工程与电子技术, 2021, 43(7): 2011-2020.
[6]	李昂, 聂党民, 温祥西, 王泽坤, 杨诚修. 基于相依网络和SVM的管制系统运行态势评估[J]. 系统工程与电子技术, 2021, 43(5): 1287-1294.
[7]	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.
[8]	王坤, 侯树贤, 王力. 基于自适应变异PSO-SVM的APU性能参数预测模型[J]. 系统工程与电子技术, 2021, 43(2): 526-536.
[9]	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.
[10]	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.
[11]	王强, 贾希胜, 程中华, 王双川, 马云飞. 基于改进的局部保持投影的战时备件分类[J]. 系统工程与电子技术, 2020, 42(1): 133-140.
[12]	吴青, 臧博研, 祁宗仙, 张昱. 基于压缩感知的多核稀疏最小二乘支持向量机[J]. 系统工程与电子技术, 2019, 41(9): 1930-1936.
[13]	王彩云, 黄盼盼, 李晓飞, 王佳宁, 赵焕玥. 基于AEPSO-SVM算法的雷达HRRP目标识别[J]. 系统工程与电子技术, 2019, 41(9): 1984-1989.
[14]	苏娟, 王延钊, 伍薇. 基于地形起伏和SVM的SAR景象MSA选取[J]. 系统工程与电子技术, 2019, 41(8): 1735-1741.
[15]	刘恒燕, 宋杰, 熊伟, 梁晓军. 大入射余角CFAR检测器[J]. 系统工程与电子技术, 2019, 41(6): 1218-1223.