基于深度神经网络的空间目标结构识别
|
摘要
利用空间目标雷达散射截面(Radar Cross Section, RCS)序列开展空间目标结构识别是空间态势感知的重要组成部分。文章针对RCS序列受目标物理特性、姿态特性影响大,序列信号非平稳特征明显的问题,利用深度神经网络(Deep Neural Network,DNN)算法解决空间目标结构特征识别的问题;针对特征提取不具区分度的问题,提出利用分形分析提取RCS序列的分数维特征,并利用Fisher判决率对传统特征进行选取;介绍了DNN算法以及数据处理过程;最后,利用一组仿真测试数据对算法进行了仿真验证。分析结果表明,DNN算法在解决利用RCS序列进行目标结构识别这一问题中具有鲁棒性强、识别准确的特点。
The use of Radar Cross Section(RCS) sequences for spatial target structure recognition is an important part of space situational awareness. RCS sequence is easily affected by the target physical characteristics and attitude characteristics,and the non-stationary characteristics of the sequence signal are obvious. In this paper, deep neural network(DNN) algorithm was used to solve the problem of spatial target structural feature recognition. For the problem of feature extraction without distinguishing degree, fractal features were used to extract the fractal features of RCS sequences, and the Fisher′s decision rate was used for selecting traditional features. What′s more, the DNN algorithm and data processing process were introduced. Finally, a set of simulation test data were used to verify the algorithm. The analysis results show that the DNN algorithm is robust and accurate in solving the problem of using RCS sequence to identify the target structure.
The use of Radar Cross Section(RCS) sequences for spatial target structure recognition is an important part of space situational awareness. RCS sequence is easily affected by the target physical characteristics and attitude characteristics,and the non-stationary characteristics of the sequence signal are obvious. In this paper, deep neural network(DNN) algorithm was used to solve the problem of spatial target structural feature recognition. For the problem of feature extraction without distinguishing degree, fractal features were used to extract the fractal features of RCS sequences, and the Fisher′s decision rate was used for selecting traditional features. What′s more, the DNN algorithm and data processing process were introduced. Finally, a set of simulation test data were used to verify the algorithm. The analysis results show that the DNN algorithm is robust and accurate in solving the problem of using RCS sequence to identify the target structure.
引文
[1] 曹斌. 国外空间目标监视系统的发展[J]. 军事文摘, 2015(21):18-22. CAO B. Development of foreign space target surveillance system[J]. Military Digest, 2015(21):18-22(in Chinese).
[2] 金胜,高梅国, 王洋. 基于RCS的空间目标识别技术[J]. 现代雷达, 2010, 32(6):59-62. JIN S, GAO M G, WANG Y. Technology of space target recognition based on RCS[J]. Modern Radar, 2010, 32(6):59-62(in Chinese).
[3] SATO T, WAKAYAMA T, TANAKA T, et al. Shape of space debris as estimated from radar cross section variations[J]. Journal of Spacecraft & Rockets, 1994, 31(31):665-670.
[4] LAMBOUR R, RAJAN N, MORGAN T, et al. Assessment of orbital debris size estimation from radar cross-section measurements[J]. Advances in Space Research, 2004, 34(5):1013-1020.
[5] 黄小红. 基于RCS序列的空间目标形状估计[J]. 航天电子对抗, 2005, 21(4):44-46. HUANG X H. Estimation of space object shapes from RCS times[J]. Aerospace Electronic Warfare, 2005, 21(4):44-46 (in Chinese).
[6] 张森. 卫星目标散射特性及识别方法研究[D]. 哈尔滨:哈尔滨工业大学, 2008. ZHANG S. RESEARCH OF SCATTERING CHARACTE-RISTICS AND RECOGNITION METHOD FOR SATELLITE TARGETS[D]. Harbin:Harbin Institute of Technology, 2008(in Chinese).
[7] 张建强, 高世家, 赵霁红. 舰船RCS特征提取与GA-BP神经网络分类研究[J]. 舰船科学技术, 2016, 38(3):125-130. ZHANG J Q, GAO S J, ZHAO J H. Research on RCS feature extraction and classification of GA-BP neural network[J]. Ship Science and Technology, 2016, 38(3):125-130(in Chinese).
[8] 詹武平,郑永煌,王金霞. 基于深度神经网络模型的雷达目标识别[J]. 现代雷达,2018,40 (1):16-19. ZHAN W P, ZHENG Y H, WANG J X. Radar target recogintion based on deep neural network[J]. Modern Radar,2018,40 (1):16-19(in Chinese).
[9] 黄小红, 邱兆坤, 陈曾平. 空间目标RCS序列的分形分析[J]. 中国空间科学技术, 2005, 25(1):33-36. HUANG X H, QIU Z K, CHEN Z P. Fractal analysis of spatial target RCS sequences[J]. Chinese Space Science and Technology, 2005, 25(1):33-36(in Chinese).
[10] GAO J, CAO Y, TUNG W W, et al. Multiscale analysis of complex time series: integration of chaos and random fractal theory, and beyond[M]. Wiley-Interscience, 2007.
[11] 行鸿彦,龚平,徐玮. 海杂波背景下小目标检测的分形方法[J]. 物理学报, 2012, 61(16):160504-1-160504-10. XING H Y, GONG P, XU W. Fractal method for small target detection in sea clutter background[J]. Acta Physica Sinica, 2012, 61(16):160504-1-160504-10(in Chinese).
[12] 耿庆田, 赵宏伟. 基于分形维数和隐马尔科夫特征的车牌识别[J]. 光学精密工程, 2013, 21(12):3198-3204. GENG Q T, ZHAO H W.License plate recognition based on fractal and hidden Markov feature[J]. Optics and Precision Engineering, 2013, 21(12):3198-3204.
[13] 吴岸城. 神经网络与深度学习[M]. 北京:电子工业出版社, 2016. WU A C. Neural network and deep learning[M].Beijing:Publishing House of Electronics Industry, 2016(in Chinese).
[14] 宋谊强. 某相控阵雷达目标识别技术研究[D].兰州:兰州大学,2008. SONG Y Q.The phased array radar target recognition technogy research[D]. Lanzhou:Lanzhou University,2008(in Chinese).
[15] 杜兰, 刘宏伟, 保铮. 一种用于雷达HRRP功率谱的加权特征压缩方法[J]. 西安电子科技大学学报(自然科学版), 2006, 33(2):173-177. DU L, LIU H W, BAO Z. A weighted feature reduction method for the power spectrum of radar HRRP[J].Journal of Xidian University(Science and Technology), 2006, 33(2):173-177(in Chinese).
[16] 周绍磊, 廖剑, 史贤俊. 基于Fisher准则和最大熵原理的SVM核参数选择方法[J]. 控制与决策, 2014, 29(11):1991-1996. ZHOU S L, LIAO J, SHI X J.SVM parameters selection method based on Fisher criterion and maximum entropy principle[J]. Control and Decision, 2014, 29(11):1991-1996(in Chinese).
[17] 徐灿, 李智. 基于改进Gordon方程的RCS快速算法[J]. 装备学院学报, 2016, 27(5):85-89. XU C, LI Z. Fast Algorithm for Calculating RCS Based on Improved Gordon Equation[J]. Journal of Equipment Academy, 2016, 27(5):85-89(in Chinese).
[2] 金胜,高梅国, 王洋. 基于RCS的空间目标识别技术[J]. 现代雷达, 2010, 32(6):59-62. JIN S, GAO M G, WANG Y. Technology of space target recognition based on RCS[J]. Modern Radar, 2010, 32(6):59-62(in Chinese).
[3] SATO T, WAKAYAMA T, TANAKA T, et al. Shape of space debris as estimated from radar cross section variations[J]. Journal of Spacecraft & Rockets, 1994, 31(31):665-670.
[4] LAMBOUR R, RAJAN N, MORGAN T, et al. Assessment of orbital debris size estimation from radar cross-section measurements[J]. Advances in Space Research, 2004, 34(5):1013-1020.
[5] 黄小红. 基于RCS序列的空间目标形状估计[J]. 航天电子对抗, 2005, 21(4):44-46. HUANG X H. Estimation of space object shapes from RCS times[J]. Aerospace Electronic Warfare, 2005, 21(4):44-46 (in Chinese).
[6] 张森. 卫星目标散射特性及识别方法研究[D]. 哈尔滨:哈尔滨工业大学, 2008. ZHANG S. RESEARCH OF SCATTERING CHARACTE-RISTICS AND RECOGNITION METHOD FOR SATELLITE TARGETS[D]. Harbin:Harbin Institute of Technology, 2008(in Chinese).
[7] 张建强, 高世家, 赵霁红. 舰船RCS特征提取与GA-BP神经网络分类研究[J]. 舰船科学技术, 2016, 38(3):125-130. ZHANG J Q, GAO S J, ZHAO J H. Research on RCS feature extraction and classification of GA-BP neural network[J]. Ship Science and Technology, 2016, 38(3):125-130(in Chinese).
[8] 詹武平,郑永煌,王金霞. 基于深度神经网络模型的雷达目标识别[J]. 现代雷达,2018,40 (1):16-19. ZHAN W P, ZHENG Y H, WANG J X. Radar target recogintion based on deep neural network[J]. Modern Radar,2018,40 (1):16-19(in Chinese).
[9] 黄小红, 邱兆坤, 陈曾平. 空间目标RCS序列的分形分析[J]. 中国空间科学技术, 2005, 25(1):33-36. HUANG X H, QIU Z K, CHEN Z P. Fractal analysis of spatial target RCS sequences[J]. Chinese Space Science and Technology, 2005, 25(1):33-36(in Chinese).
[10] GAO J, CAO Y, TUNG W W, et al. Multiscale analysis of complex time series: integration of chaos and random fractal theory, and beyond[M]. Wiley-Interscience, 2007.
[11] 行鸿彦,龚平,徐玮. 海杂波背景下小目标检测的分形方法[J]. 物理学报, 2012, 61(16):160504-1-160504-10. XING H Y, GONG P, XU W. Fractal method for small target detection in sea clutter background[J]. Acta Physica Sinica, 2012, 61(16):160504-1-160504-10(in Chinese).
[12] 耿庆田, 赵宏伟. 基于分形维数和隐马尔科夫特征的车牌识别[J]. 光学精密工程, 2013, 21(12):3198-3204. GENG Q T, ZHAO H W.License plate recognition based on fractal and hidden Markov feature[J]. Optics and Precision Engineering, 2013, 21(12):3198-3204.
[13] 吴岸城. 神经网络与深度学习[M]. 北京:电子工业出版社, 2016. WU A C. Neural network and deep learning[M].Beijing:Publishing House of Electronics Industry, 2016(in Chinese).
[14] 宋谊强. 某相控阵雷达目标识别技术研究[D].兰州:兰州大学,2008. SONG Y Q.The phased array radar target recognition technogy research[D]. Lanzhou:Lanzhou University,2008(in Chinese).
[15] 杜兰, 刘宏伟, 保铮. 一种用于雷达HRRP功率谱的加权特征压缩方法[J]. 西安电子科技大学学报(自然科学版), 2006, 33(2):173-177. DU L, LIU H W, BAO Z. A weighted feature reduction method for the power spectrum of radar HRRP[J].Journal of Xidian University(Science and Technology), 2006, 33(2):173-177(in Chinese).
[16] 周绍磊, 廖剑, 史贤俊. 基于Fisher准则和最大熵原理的SVM核参数选择方法[J]. 控制与决策, 2014, 29(11):1991-1996. ZHOU S L, LIAO J, SHI X J.SVM parameters selection method based on Fisher criterion and maximum entropy principle[J]. Control and Decision, 2014, 29(11):1991-1996(in Chinese).
[17] 徐灿, 李智. 基于改进Gordon方程的RCS快速算法[J]. 装备学院学报, 2016, 27(5):85-89. XU C, LI Z. Fast Algorithm for Calculating RCS Based on Improved Gordon Equation[J]. Journal of Equipment Academy, 2016, 27(5):85-89(in Chinese).