修正时延扩展卡尔曼滤波MIMO声纳目标跟踪方法研究  

Research on target tracking based on modified extended Kalman filter in distributed multiple-input multiple-output sonar system

作　　者：罗胜男[1,2] 付广义[1,2] 贺旭[1,2] 李宇[1] 尹力[1] 

机构地区：[1]中国科学院声学研究所,北京100190 [2]中国科学院研究生院,北京100190

出　　处：《声学技术》2013年第6期458-463,共6页Technical Acoustics

基　　金：863计划项目(2009AA093601);国家自然科学基金项目(10904160)

摘　　要：分布式多输入多输出(Multiple-Input Multiple-Output,MIMO)声纳是一种通过MIMO技术规划时空信道来提高声纳探测性能的新型主动探测声纳体制。由于分布式MIMO声纳节点分布间隔大,水中声速较小,由各发射节点同步发射的测距信号将经过不同的时延到达目标,因此各接收节点测得的距离值分别对应于目标不同时刻的状态。常规的定位方法并没有考虑传播时延对测量值的影响,因而定位精度受到限制。提出了一种修正时延的扩展卡尔曼滤波方法(Modified Extended Kalman Filter,MEKF)对分布式MIMO声纳系统中的移动目标进行跟踪。仿真结果表明,与常规的目标定位跟踪方法相比,该方法有定位精度高、收敛速度快、跟踪性能稳定的特点。The distributed Multiple-Input Multiple-Output （MIMO） sonar system is new active detection sonar that offers detection improvement by planning space-time channels. The stations in such system are widely separated and the un- derwater sound velocity is small. There are different time delays for the signals launched from different transmitters to reach the target. Thereore, the distance measured at each receiver corresponds to the status of target at different time. The accuracy of common localization methods is limited because the time delay for transmission has not been taken into account. In this paper, a Modified Extended Kalman Filter method （MEKF） based on delay correction was proposed to deal with this target tracking problem. Simulation results showed that the MEKF method could provide higher accuracy, faster convergence and steadier performance.

关 键 词：分布式MIMO声纳系统 定位精度 修正时延的扩展卡尔曼滤波 目标跟踪 

分 类 号：TB56[交通运输工程—水声工程]