复杂环境下的小目标信号检测和跟踪技术研究
|
摘要
雷达要探测的目标,通常是运动着的物体,如飞机、船舶等,但在目标的周围经常存在着各种杂波干扰,如地杂波、云雨杂波、海浪杂波、敌方施放的箔条杂波等。
本文的研究目标是在强海杂波背景下,研究弱小目标信号的相关检测技术；研究在弱信噪比、低检测概率条件下,高机动目标的有效跟踪方法,实现在海面强杂波条件下,对弱小和高机动目标的可靠检测、稳定跟踪。
目标检测算法采用的是网格法。通过设定CFAR恒虚警系数,得到CFAR处理结果,并对满足门限的CFAR单元置标志送往后续做距离和方位检测,检测都采用M/N开始和结束准则。
目标跟踪算法主要研究了α-β算法Kalman滤波算法和IMM交互式多模型算法,并对三种算法做了比较：α-β滤波器的特点是不容易选择α和β的决策,Kalman滤波器可以提供α-β滤波器所不具有的滤波协方差阵和机动程度,而IMM算法能假定在一个给定的时间,动态系统的变化能用M个模型中的一个精确的表示。
通过Matlab对仿真目标应用三种滤波算法,对比三种跟踪算法的滤波结果和预测结果发现,α-β滤波器的滤波点与预测点的误差比较大,标准Kalman滤波算法,也存在系数优化难,机动准则难以判断等缺点,而IMM多模型算法具有机动自动判断与融合、算法稳健、需要调整的参数少等优点,而且能达到较好的跟踪效果。
Targets detected by radars are mostly moving objects, such as aircrafts and ships; and around them are often varieties of clutters cause by ground objects, clouds and rain, sea waves, and sometimes chaffs sent by enemies.
Studies are made on the technologies of detecting weak signals of small targets in the background of strong sea clutters, as well as the ways of tracking highly maneuverable targets with small SNR and low detection probability, so as to facilitate reliable detection and stable tracking of small, maneuverable targets in strong sea clutters.
The grid method is adopted as the algorithm for target detection. The result of CFAR processing is achieved through the setting of the factor of the CFAR. Flags are given to those CFAR cells that satisfy the threshold requirement and the cells are forwarded for the follow-up range and bearing detections in which the M/N start and end criteria are employed.
Three algorithms, which are respectively theα-βalgorithm, the Kalman filtering algorithm, and the Interactive Multiple Mode (IMM) algorithm, are studied; and comparisons are made among them. Results show that the decisions of theαand theβare not easily made for theα-βalgorithm; whereas the Kalman filtering algorithm is capable of providing the filtering covariance matrix and the maneuverability that are not possessed by the a-p algorithm. Besides, for the IMM algorithm, the varieties of the dynamic system can be precisely described by one of the M different models within a given period.
Simulations are carried out through Matlab for the three algorithms; and comparisons are made between the results of filtering processing and prediction. It is found that relatively big error exists between the filtered point and the predicted one for theα-βalgorithm; the typical Kalman filtering algorithm also has the weak points as difficulties in factor optimization and criteria judgment for maneuvers; and the IMM algorithm features to be robust, capable of making automatic judgment for maneuvers and fusions for judgment results, and having less parameters that need to be adjusted, making the tracking results better.
本文的研究目标是在强海杂波背景下,研究弱小目标信号的相关检测技术；研究在弱信噪比、低检测概率条件下,高机动目标的有效跟踪方法,实现在海面强杂波条件下,对弱小和高机动目标的可靠检测、稳定跟踪。
目标检测算法采用的是网格法。通过设定CFAR恒虚警系数,得到CFAR处理结果,并对满足门限的CFAR单元置标志送往后续做距离和方位检测,检测都采用M/N开始和结束准则。
目标跟踪算法主要研究了α-β算法Kalman滤波算法和IMM交互式多模型算法,并对三种算法做了比较：α-β滤波器的特点是不容易选择α和β的决策,Kalman滤波器可以提供α-β滤波器所不具有的滤波协方差阵和机动程度,而IMM算法能假定在一个给定的时间,动态系统的变化能用M个模型中的一个精确的表示。
通过Matlab对仿真目标应用三种滤波算法,对比三种跟踪算法的滤波结果和预测结果发现,α-β滤波器的滤波点与预测点的误差比较大,标准Kalman滤波算法,也存在系数优化难,机动准则难以判断等缺点,而IMM多模型算法具有机动自动判断与融合、算法稳健、需要调整的参数少等优点,而且能达到较好的跟踪效果。
Targets detected by radars are mostly moving objects, such as aircrafts and ships; and around them are often varieties of clutters cause by ground objects, clouds and rain, sea waves, and sometimes chaffs sent by enemies.
Studies are made on the technologies of detecting weak signals of small targets in the background of strong sea clutters, as well as the ways of tracking highly maneuverable targets with small SNR and low detection probability, so as to facilitate reliable detection and stable tracking of small, maneuverable targets in strong sea clutters.
The grid method is adopted as the algorithm for target detection. The result of CFAR processing is achieved through the setting of the factor of the CFAR. Flags are given to those CFAR cells that satisfy the threshold requirement and the cells are forwarded for the follow-up range and bearing detections in which the M/N start and end criteria are employed.
Three algorithms, which are respectively theα-βalgorithm, the Kalman filtering algorithm, and the Interactive Multiple Mode (IMM) algorithm, are studied; and comparisons are made among them. Results show that the decisions of theαand theβare not easily made for theα-βalgorithm; whereas the Kalman filtering algorithm is capable of providing the filtering covariance matrix and the maneuverability that are not possessed by the a-p algorithm. Besides, for the IMM algorithm, the varieties of the dynamic system can be precisely described by one of the M different models within a given period.
Simulations are carried out through Matlab for the three algorithms; and comparisons are made between the results of filtering processing and prediction. It is found that relatively big error exists between the filtered point and the predicted one for theα-βalgorithm; the typical Kalman filtering algorithm also has the weak points as difficulties in factor optimization and criteria judgment for maneuvers; and the IMM algorithm features to be robust, capable of making automatic judgment for maneuvers and fusions for judgment results, and having less parameters that need to be adjusted, making the tracking results better.
引文
[1]薛国义.强海杂波背景下实时目标检测与跟踪技术研究和系统实现.国防科技大学硕士学位论文.2003
[2]马玲.边扫描边跟踪雷达航迹跟踪算法的研究.西北工业大学硕士学位论文.2004
[3]刘剑.海杂波背景下目标检测方法研究,国防科技大学硕士学位论文.2002.11
[4]袁立权,靖涛.K—分布杂波的顺序统计恒虚警性能分析.现代雷达.1999.8(4)
[5]钟循进,张庆海.海浪环境中脉冲雷达的恒虚警检测器.舰船电子对抗.1998
[6]韩雁飞.IMM理论在超低空探测与跟踪上的应用.系统工程与电子技术.2003(1)
[7]许江湖.多模型滤波器.情报控制系统多与仿真技术.2001(4)
[8]许江湖.目标跟踪中的多模型估计算法综述,情报控制系统多与仿真技术.2002(4)
[9]于振华.雷达检测波门的设计方法与原理,舰船电子对抗.2002(5)
[10]Merrill I. Skolnik雷达手册.第2版.北京：电子工业出版社,2003.7
[11]戴耀.杂波环境下机动目标跟踪模拟.情报指挥控制系统与仿真技术.2001(12)
[12]韩兴斌,胡卫东,杨世海.应用交互多模IMM算法跟踪低空目标.火力与指挥控制.2003.10(5)
[13]张永胜.机动目标跟踪的模式集自适应IMM算法的设计和比较.情报指挥控制系统与仿真技术.2001.8
[14]张安明,杨世兴,李志舜.目标跟踪中的混合多模方法综述.系统工程与电子技术.2001(4)
[15]刘春恒,梁彦,周东华.基于最小二乘滤波起始的机动目标被动跟踪方法.系统工程与电子技术.2003(4)
[16]张永胜.自调整交互多模型.情报指挥控制系统与仿真技术.2001(1)
[17]王书宏.低分辨雷达目标识别系统中的检测、跟踪与数据压缩技术研究与实现.国防科技大学硕士学位论文.2002.11
[18]党莹.强杂波下机动目标跟踪技术研究.中科院长春光学精密机械及物理研究所博士学位论文.2001.4
[19]韩兴斌.雷达低空目标跟踪技术研究,国防科技大学硕士学位论文.2002.11
[20]杨世海.相控阵雷达低空目标探测与跟踪技术研究.国防科技大学博士学位论文.2002.12
[21]王威,姜义成,张宁.带有非相参积累的序列统计恒虚警检测器.系统工程与电 子技术.1997
[22]邹世迁,石岩,张辉.恒虚警概率新型实现途径及其检测方法.舰船电子对抗.2003(1)
[23]王新智.基于雷达PPI的目标检测与跟踪方法研究.国防科技大学硕士学位论文.2002.11
[24]CFAR at the end of Instrumented Range. Radar Conference,1999,250-255
[25]Mucahit K. Uner. Distributed CFAR detection in homogeneous and nonhomogeneous backgrounds. IEEE Trans on AES,1996,32(1)
[26]Michael E. Smith. Intelligent CFAR processor based on data variability. IEEE Trans on AES,2000,36(3)
[27]P. Tsakalides. Performance assessment of CFAR processors in Person-distributed Clutter. IEEE Trans on AES,2000,36(4)
[28]H.Rohling. Radar CFAR thresholding in clutter and Multiple target situations. IEEE Trans on AES,1983,19(4)
[29]Johnson G.. Construction of Particular Random Processes. IEEE Trans on AES, 1994,2(2):270-285
[30]L.James Marieretal. Correlated K-Distributed Clutter Generation for Radar Detection and Track. IEEE Trans on AES,1993,31(2):568-580
[31]K.J. Lewinski. Non-Stationary Probabilistic Target and Clutter Scatting Models. IEEE Trans on AP,1983,31(3):490-498
[32]D.R. Iskanderet. Estimating the Parameters of K_Distribution Using Higher-Order and Fractional Moments. IEEE Trans on AES,1999,35(4):1453-1457
[33]Guan J.. Distributed CFAR detector based on local test statistic. IEEE Trans on AES, 2000,80(2):373-379
[34]Viswanathan R.. A new Distributed constant false alarm rate detector. IEEE Trans on AES,1997,33(1):85-96
[35]Barkat M. Decentralized CFAR signal detection. IEEE Trans on AES,1989,25(2): 141-149
[2]马玲.边扫描边跟踪雷达航迹跟踪算法的研究.西北工业大学硕士学位论文.2004
[3]刘剑.海杂波背景下目标检测方法研究,国防科技大学硕士学位论文.2002.11
[4]袁立权,靖涛.K—分布杂波的顺序统计恒虚警性能分析.现代雷达.1999.8(4)
[5]钟循进,张庆海.海浪环境中脉冲雷达的恒虚警检测器.舰船电子对抗.1998
[6]韩雁飞.IMM理论在超低空探测与跟踪上的应用.系统工程与电子技术.2003(1)
[7]许江湖.多模型滤波器.情报控制系统多与仿真技术.2001(4)
[8]许江湖.目标跟踪中的多模型估计算法综述,情报控制系统多与仿真技术.2002(4)
[9]于振华.雷达检测波门的设计方法与原理,舰船电子对抗.2002(5)
[10]Merrill I. Skolnik雷达手册.第2版.北京：电子工业出版社,2003.7
[11]戴耀.杂波环境下机动目标跟踪模拟.情报指挥控制系统与仿真技术.2001(12)
[12]韩兴斌,胡卫东,杨世海.应用交互多模IMM算法跟踪低空目标.火力与指挥控制.2003.10(5)
[13]张永胜.机动目标跟踪的模式集自适应IMM算法的设计和比较.情报指挥控制系统与仿真技术.2001.8
[14]张安明,杨世兴,李志舜.目标跟踪中的混合多模方法综述.系统工程与电子技术.2001(4)
[15]刘春恒,梁彦,周东华.基于最小二乘滤波起始的机动目标被动跟踪方法.系统工程与电子技术.2003(4)
[16]张永胜.自调整交互多模型.情报指挥控制系统与仿真技术.2001(1)
[17]王书宏.低分辨雷达目标识别系统中的检测、跟踪与数据压缩技术研究与实现.国防科技大学硕士学位论文.2002.11
[18]党莹.强杂波下机动目标跟踪技术研究.中科院长春光学精密机械及物理研究所博士学位论文.2001.4
[19]韩兴斌.雷达低空目标跟踪技术研究,国防科技大学硕士学位论文.2002.11
[20]杨世海.相控阵雷达低空目标探测与跟踪技术研究.国防科技大学博士学位论文.2002.12
[21]王威,姜义成,张宁.带有非相参积累的序列统计恒虚警检测器.系统工程与电 子技术.1997
[22]邹世迁,石岩,张辉.恒虚警概率新型实现途径及其检测方法.舰船电子对抗.2003(1)
[23]王新智.基于雷达PPI的目标检测与跟踪方法研究.国防科技大学硕士学位论文.2002.11
[24]CFAR at the end of Instrumented Range. Radar Conference,1999,250-255
[25]Mucahit K. Uner. Distributed CFAR detection in homogeneous and nonhomogeneous backgrounds. IEEE Trans on AES,1996,32(1)
[26]Michael E. Smith. Intelligent CFAR processor based on data variability. IEEE Trans on AES,2000,36(3)
[27]P. Tsakalides. Performance assessment of CFAR processors in Person-distributed Clutter. IEEE Trans on AES,2000,36(4)
[28]H.Rohling. Radar CFAR thresholding in clutter and Multiple target situations. IEEE Trans on AES,1983,19(4)
[29]Johnson G.. Construction of Particular Random Processes. IEEE Trans on AES, 1994,2(2):270-285
[30]L.James Marieretal. Correlated K-Distributed Clutter Generation for Radar Detection and Track. IEEE Trans on AES,1993,31(2):568-580
[31]K.J. Lewinski. Non-Stationary Probabilistic Target and Clutter Scatting Models. IEEE Trans on AP,1983,31(3):490-498
[32]D.R. Iskanderet. Estimating the Parameters of K_Distribution Using Higher-Order and Fractional Moments. IEEE Trans on AES,1999,35(4):1453-1457
[33]Guan J.. Distributed CFAR detector based on local test statistic. IEEE Trans on AES, 2000,80(2):373-379
[34]Viswanathan R.. A new Distributed constant false alarm rate detector. IEEE Trans on AES,1997,33(1):85-96
[35]Barkat M. Decentralized CFAR signal detection. IEEE Trans on AES,1989,25(2): 141-149
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |