基于时频能量谱纹理特征的跳频调制方式识别

doi:10.11959/j.issn.1000-436x.2019191

通信学报 ›› 2019, Vol. 40 ›› Issue (10): 20-29.doi: 10.11959/j.issn.1000-436x.2019191

基于时频能量谱纹理特征的跳频调制方式识别

李红光, 郭英, 眭萍, 齐子森

空军工程大学信息与导航学院，陕西西安 710077

修回日期:2019-07-31 出版日期:2019-10-25 发布日期:2019-11-07
作者简介:李红光（1986- ），男，天津蓟县人，空军工程大学博士生，主要研究方向为信息对抗理论、通信信号处理。|郭英（1961- ），女，山西临汾人，博士，空军工程大学教授、博士生导师，主要研究方向为信息对抗理论、通信信号处理、自适应信号处理。|眭萍（1991- ），女，山西太原人，空军工程大学博士生，主要研究方向为通信信号侦察处理。|齐子森（1982- ），男，河北保定人，博士，空军工程大学讲师，主要研究方向为通信信号侦察处理、阵列信号处理。
基金资助:
国家自然科学基金资助项目(61601500);全军研究生基金资助项目(JY2018C169)

Frequency hopping modulation recognition based on time-frequency energy spectrum texture feature

Hongguang LI, Ying GUO, Ping SUI, Zisen QI

Information and Navigation College,Air Force Engineering University,Xi’an 710077,China

Revised:2019-07-31 Online:2019-10-25 Published:2019-11-07
Supported by:
The National Natural Science Foundation of China(61601500);Military Graduate Student Funding(JY2018C169)

摘要/Abstract

摘要：

针对跳频通信调制方式识别问题，提出了一种基于时频能量谱纹理特征的跳频调制方式识别方法。该方法首先采用平滑伪Wigner-Ville分布算法获取跳频信号时频图，并经过二维维纳滤波去除时频图背景噪声，提高低信噪比条件下时频图清晰度；然后采用连通域检测算法提取每跳信号的时频能量谱并将其转化为时频灰度图，计算其直方图统计特征和灰度共生矩阵特征组成 22 维特征向量；最后通过参数优化后的支持向量机分类器对特征集进行训练、分类和识别。仿真实验表明，所提取的多维特征向量具有较强的表征能力，避免了由单一特征相似性引起的误判问题，在信噪比为-4 dB的条件下，对跳频信号BPSK、QPSK、SDPSK、QASK、64QAM和GMSK共6种调制方式的平均识别正确率达到91.4%。

关键词: 跳频调制识别, 时频能量谱, 时频灰度图, 直方图统计, 灰度共生矩阵

Abstract:

For frequency hopping modulation identification,a novel method based on time-frequency energy spectrum texture feature was proposed.Firstly,the time-frequency diagram of the frequency hopping signal was obtained by smoothed pseudo Wigner-Ville distribution,and the background noise of the time-frequency diagram was removed by two-dimensional Wiener filtering to improve the resolution of the time-frequency diagram under low SNR conditions.Then,the connected-domain detection algorithm was used to extract the time-frequency energy spectrum of each hop signal and convert it into a time-frequency gray-scale image.The histogram statistical features and the gray-scale co-occurrence matrix feature were combined to form a 22-dimensional eigenvector.Finally,the feature set was trained,classified and identified by optimized support vector machine classifier.Simulation experiments show that the multi-dimensional feature vector extracted by the algorithm has strong representation ability and avoids the misjudgment caused by the similarity of single features.The average recognition accuracy of the six modulation methods of frequency hopping signals BPSK,QPSK,SDPSK,QASK,64QAM and GMSK is 91.4% under the condition of -4 dB SNR.

Key words: frequency hopping modulation recognition, time-frequency energy spectrum, time-frequency gray scale image, histogram statistics, gray co-occurrence matrix

中图分类号:

TN971

李红光, 郭英, 眭萍, 齐子森. 基于时频能量谱纹理特征的跳频调制方式识别[J]. 通信学报, 2019, 40(10): 20-29.

Hongguang LI, Ying GUO, Ping SUI, Zisen QI. Frequency hopping modulation recognition based on time-frequency energy spectrum texture feature[J]. Journal on Communications, 2019, 40(10): 20-29.

图/表 9

图1

图2

图3

图4

表1

图5

图6

图7

图8

参考文献 22

[1]	SU W . Feature space analysis of modulation classification using very high-order statistics[J]. IEEE Communications Letters, 2013,17(9): 1688-1691.
[2]	刘明骞, 李兵兵, 曹超凤 . 认知无线电中非高斯噪声下数字调制信号识别方法[J]. 通信学报, 2014,35(1): 82-88.
	LIU M S , LI B B , CAO C F . Recognition method of digital modulation signals over non-Gaussian noise in cognitive radio[J]. Journal on Communications, 2014,35(1): 82-88.
[3]	WANG F , DOBRE O A , CHAN C ,et al. Fold-based kolmogorov-smirnov modulation classifier[J]. IEEE Signal Processing Letters, 2016,23(7): 1003-1007.
[4]	赵雄文, 郭春霞, 李景春 . 基于高阶累积量和循环谱的信号调制方式混合识别算法[J]. 电子与信息学报, 2016,38(3): 674-680.
	ZHAO X W , GUO C X , LI J C . Mixed recognition algorithm for signal modulation schemes by high-order cumulants and cyclic spectrum[J]. Journal of Electronics ＆ Information Technology, 2016,38(3): 674-680.
[5]	HAN L , GAO F , LI Z ,et al. Low complexity automatic modulation classification based on order-statistics[J]. IEEE Transactions on Wireless Communications, 2017,16(1): 400-411.
[6]	DOBTE O A , ONER M , RAJAN S ,et al. Cyclostationarity-based robust algorithms for QAM signal identification[J]. IEEE Communications Letters, 2012,16(1): 12-15.
[7]	XIE L , WAN Q . Cyclic feature based modulation recognition using compressive sensing[J]. IEEE Wireless Communications Letters, 2017,6(3): 402-405.
[8]	RAJENDRAN S , MEERT W , GIUSTINIANO D ,et al. Deep learning models for wireless signal classification with distributed low-cost spectrum sensors[J]. IEEE Transactions on Cognitive Communications and Networking, 2018,4(3): 433-445.
[9]	LI R , LI L , YANG S ,et al. Robust automated VHF modulation recognition based on deep convolutional neural networks[J]. IEEE Communications Letters, 2018,22(5): 946-949.
[10]	ZIEGLER J L , ARN R T , CHAMBERS W ,et al. Modulation recognition with GNU radio,Keras,and HackRF[C]// IEEE International Symposium on Dynamic Spectrum Access Networks. IEEE, 2017: 1-4.
[11]	LI X Q , HUANG Z T , WANG F H ,et al. Towards convolutional neural networks on pulse repetition interval modulation recognition[J]. IEEE Communications Letters, 2018,22(11): 2286-2289.
[12]	QU Z , MAO X , DENG Z ,et al. Radar signal intra-pulse modulation recognition based on convolutional neural network[J]. IEEE Access, 2018,6: 43874-43884.
[13]	WANG C , WANG J , ZHANG X ,et al. Automatic radar waveform recognition based on time-frequency analysis and convolutional neural network[C]// 2017 IEEE International Conference on Acoustics,Speech and Signal Processing. IEEE, 2017: 2437-2441.
[14]	LIM JAE S . Two-dimensional signal and image processing[J]. Englewood Cliffs Nj Prentice Hall P, 1990,34(1): 6-8.
[15]	MISHRA V N , PRASAD R , KUMAR P ,et al. Dual-polarimetric C-band SAR data for land use/land cover classification by incorporating textural information[J]. Environmental Earth Sciences, 2017,76(1):26
[16]	YE F , HAN M . Simultaneous feature with support vector selection and parameters optimization using GA-based SVM solve the binary classification[C]// IEEE International Conference on Computer Communication ＆ the Internet. IEEE, 2016: 426-433.
[17]	ALAM S , . Performance of alzheimer disease classification based on PCA,linear SVM,and multi-kernel SVM[C]// The Eighth International Conference on Ubiquitous and Future Networks. IEEE, 2016: 987-989.
[18]	ZHANG Q , ZHANG P , LONG G ,et al. Online learning from trapezoidal data streams[J]. IEEE Transactions on Knowledge and Data Engineering, 2016,28(10): 2709-2723.
[19]	GRAMA L , TUNS L , RUSU C . On the optimization of SVM kernel parameters for improving audio classification accuracy[C]// International Conference on Engineering of Modern Electric Systems. IEEE, 2017: 224-227.
[20]	GUPTA S , SAURAV S , SINGH S ,et al. VLSI architecture of exponential block for non-linear SVM classification[C]// International Conference on Advances in Computing. IEEE, 2015: 528-532.
[21]	YANG Q , FU H , ZHU T . An optimization method for parameters of SVM in network intrusion detection system[C]// 2016 International Conference on Distributed Computing in Sensor Systems. IEEE Computer Society, 2016: 136-142.
[22]	HESTERMAN J Y , CAUCCI L , KUPINSKI M A ,et al. Maximum-likelihood estimation with a contracting-grid search algorithm[J]. IEEE Transactions on Nuclear Science, 2010,57(3): 1077-1084.

输入	输出
输入	BPSK	QPSK	SDPSK	QASK	64QAM	GMSK
BPSK	89.5%	2.3%	8.2%	0	0	0
QPSK	4.5%	91.8%	3.7%	0	0	0
SDPSK	7.5%	2.1%	90.4%	0	0	0
QASK	0	3.1%	0	91.2%	5.7%	0
64QAM	3.9%	0	0	5.4%	90.7%	0
GMSK	1.8%	3.4%	0	0	0	94.8%

基于时频能量谱纹理特征的跳频调制方式识别

Frequency hopping modulation recognition based on time-frequency energy spectrum texture feature

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 22

相关文章 1

Metrics

推荐阅读 0