基于改进的半监督主动学习的雷达信号识别
|
摘要
为解决在雷达信号分类识别过程中训练样本较少的问题,本文提出了联合主动学习和半监督学习,并对其伪标记样本进行迭代验证改进的分类算法。针对复杂的电磁环境下雷达信号识别率低的问题,本文将径向高斯核时频分析应用于雷达信号,并对时频分布进行奇异值分解,提取出奇异向量作为雷达信号识别的特征参数。针对传统的半监督主动学习算法的不足,利用改进的半监督主动学习算法构建分类器,该算法通过对伪标记样本进行迭代验证来提高伪标记信息的准确性,从而改善了最终的分类性能,实现了在可获取的有标签样本数量较少的条件下对雷达信号的高概率识别。仿真结果表明,本文提出的特征识别方法可以获得较高的识别率。
In the process of radar signal recognition,fewer training samples is a common and challenging problem. A novel algorithm named improved semi-supervised active learning is proposed for signal classification,which is based on pseudolabels verification procedure. For the problem of low radar signal recognition in complex electromagnetic environment,the time-frequency analysis of radially Gaussian kernel is applied to radar signals. Through singular value decomposition of time-frequency distribution,it extracts its singular values as feature parameters for radar signal recognition. In order to overcome the shortcomings of the traditional semi-supervised active learning algorithm,a classifier is constructed using an improved semi-supervised active learning algorithm. The proposed algorithm enables a collaborative labeling procedure by both human experts and classifiers to acquire more confidently labeled samples to improve the final classification performance and realize the high probability of radar signal recognition when the number of available labeled samples is small. Simulation results show that the proposed feature recognition method can achieve higher radar signal recognition at low SNR.
In the process of radar signal recognition,fewer training samples is a common and challenging problem. A novel algorithm named improved semi-supervised active learning is proposed for signal classification,which is based on pseudolabels verification procedure. For the problem of low radar signal recognition in complex electromagnetic environment,the time-frequency analysis of radially Gaussian kernel is applied to radar signals. Through singular value decomposition of time-frequency distribution,it extracts its singular values as feature parameters for radar signal recognition. In order to overcome the shortcomings of the traditional semi-supervised active learning algorithm,a classifier is constructed using an improved semi-supervised active learning algorithm. The proposed algorithm enables a collaborative labeling procedure by both human experts and classifiers to acquire more confidently labeled samples to improve the final classification performance and realize the high probability of radar signal recognition when the number of available labeled samples is small. Simulation results show that the proposed feature recognition method can achieve higher radar signal recognition at low SNR.
引文
[1]Baraniuk R G,Jones D L.Signal-dependent time-frequency analysis using a radially Gaussian kernel[J].Signal Processing,1993,32(3):263-284.
[2]Rong Li,Wang Huaning,Cui Yanmei,et al.Solar flare forecasting using learning vector quantity and unsupervised clustering techniques[J].Science China,2011,54(8):1546-1552.
[3]Guo Qiang,Nan Pulong,Zhang Xiaoyu,et al.Recognition of radar emitter signals based on SVD and AF main ridge slice[J].Journal of Communications&Networks,2015,17(5):491-498.
[4]任东方,张涛,韩洁.结合ITD与非线性分析的通信辐射源个体识别方法[J].信号处理,2018,34(3):331-339.Ren Dongfang,Zhang Tao,Han Jie.Individual identification method of communication radiation source combined with ITD and nonlinear analysis[J].Journal of Signal Processing,2018,34(3):331-339.(in Chinese)
[5]符颖,王星,周东青,等.基于模糊函数SVD和改进S3VM的雷达信号识别[J].计算机工程与应用,2017,53(6):264-270.Fu Ying,Wang Xing,Zhou Dongqing,et al.Recognition method of radar signal based on SVD of ambiguity function and improved S3VM algorithm[J].Computer Engineering and Applications,2017,53(6):264-270.(in Chinese)
[6]吴剑旗,田西兰.一种基于半监督学习的窄带雷达目标识别系统[J].中国电子科学研究院学报,2015,10(1):49-53.Wu Jianqi,Tian Xilan.A Narrow-Band Radar Target Recognition System Based on Semi-Supervised Learning[J].Journal of China Academy of Electronics and Information Technology,2015,10(1):49-53.(in Chinese)
[7]Akay O,Boudreaux-Bartels G F.Unitary and Hermitian fractional operators and their relation to the fractional Fourier transform[J].IEEE Signal Processing Letters,2002,5(12):312-314.
[8]陈诗国,张道强.半监督降维方法的实验比较[J].软件学报,2011,22(1):28-43.Chen Shiguo,Zhang Daoqiang.Experimental Comparisons of Semi-Supervised Dimensional Reduction Methods[J].Journal of Software,2011,22(1):28-43.(in Chinese)
[9]Kothari R,Jain V.Learning from labeled and unlabeled data using a minimal number of queries[J].IEEE Transactions on Neural Networks,2003,14(6):1496-1505.
[10]Wang Zengmao,Du Bo,Zhang Lefei,et al.A Novel Semisupervised Active-Learning Algorithm for Hyperspectral Image Classification[J].IEEE Transactions on Geoscience&Remote Sensing,2017,55(6):3071-3083.
[11]Wan Lunjun,Tang Ke,Li Mingzhi,et al.Collaborative Active and Semisupervised Learning for Hyperspectral Remote Sensing Image Classification[J].IEEE Transactions on Geoscience&Remote Sensing,2015,53(5):2384-2396.
[12]Demir B,Persello C,Bruzzone L.Batch-Mode ActiveLearning Methods for the Interactive Classification of Remote Sensing Images[J].IEEE Transactions on Geoscience&Remote Sensing,2011,49(3):1014-1031.
[13]何正日.雷达辐射源信号识别技术研究[D].西安:西安电子科技大学,2015.He Zhengri.Research on Radar Emitter Signal Recognition Technology[D].Xi’an:Xidian University,2015.(in Chinese)
[2]Rong Li,Wang Huaning,Cui Yanmei,et al.Solar flare forecasting using learning vector quantity and unsupervised clustering techniques[J].Science China,2011,54(8):1546-1552.
[3]Guo Qiang,Nan Pulong,Zhang Xiaoyu,et al.Recognition of radar emitter signals based on SVD and AF main ridge slice[J].Journal of Communications&Networks,2015,17(5):491-498.
[4]任东方,张涛,韩洁.结合ITD与非线性分析的通信辐射源个体识别方法[J].信号处理,2018,34(3):331-339.Ren Dongfang,Zhang Tao,Han Jie.Individual identification method of communication radiation source combined with ITD and nonlinear analysis[J].Journal of Signal Processing,2018,34(3):331-339.(in Chinese)
[5]符颖,王星,周东青,等.基于模糊函数SVD和改进S3VM的雷达信号识别[J].计算机工程与应用,2017,53(6):264-270.Fu Ying,Wang Xing,Zhou Dongqing,et al.Recognition method of radar signal based on SVD of ambiguity function and improved S3VM algorithm[J].Computer Engineering and Applications,2017,53(6):264-270.(in Chinese)
[6]吴剑旗,田西兰.一种基于半监督学习的窄带雷达目标识别系统[J].中国电子科学研究院学报,2015,10(1):49-53.Wu Jianqi,Tian Xilan.A Narrow-Band Radar Target Recognition System Based on Semi-Supervised Learning[J].Journal of China Academy of Electronics and Information Technology,2015,10(1):49-53.(in Chinese)
[7]Akay O,Boudreaux-Bartels G F.Unitary and Hermitian fractional operators and their relation to the fractional Fourier transform[J].IEEE Signal Processing Letters,2002,5(12):312-314.
[8]陈诗国,张道强.半监督降维方法的实验比较[J].软件学报,2011,22(1):28-43.Chen Shiguo,Zhang Daoqiang.Experimental Comparisons of Semi-Supervised Dimensional Reduction Methods[J].Journal of Software,2011,22(1):28-43.(in Chinese)
[9]Kothari R,Jain V.Learning from labeled and unlabeled data using a minimal number of queries[J].IEEE Transactions on Neural Networks,2003,14(6):1496-1505.
[10]Wang Zengmao,Du Bo,Zhang Lefei,et al.A Novel Semisupervised Active-Learning Algorithm for Hyperspectral Image Classification[J].IEEE Transactions on Geoscience&Remote Sensing,2017,55(6):3071-3083.
[11]Wan Lunjun,Tang Ke,Li Mingzhi,et al.Collaborative Active and Semisupervised Learning for Hyperspectral Remote Sensing Image Classification[J].IEEE Transactions on Geoscience&Remote Sensing,2015,53(5):2384-2396.
[12]Demir B,Persello C,Bruzzone L.Batch-Mode ActiveLearning Methods for the Interactive Classification of Remote Sensing Images[J].IEEE Transactions on Geoscience&Remote Sensing,2011,49(3):1014-1031.
[13]何正日.雷达辐射源信号识别技术研究[D].西安:西安电子科技大学,2015.He Zhengri.Research on Radar Emitter Signal Recognition Technology[D].Xi’an:Xidian University,2015.(in Chinese)