摘要
研究了基于水声传感器网络的目标纯方位运动分析原理及方法,建立了基于水声传感器网络的目标运动分析模型。在此基础上,讨论了模型中多维非线性估计问题,提出了一种基于传感器网络新的水下目标运动分析方法。该方法采用改进的粒子滤波EKF-PF(扩展卡尔曼-粒子滤波)算法实现,并与传统的扩展卡尔曼滤波(EKF)和粒子滤波(PF)算法进行了比较。通过Monte Carlo仿真分析,表明基于水声传感器网络的目标运动分析方法充分利用了网络的优势和当前测量信息。这种方法对水下目标运动状态估计时,不仅降低了计算量而且表现出较高的估计精度。所得结论为水下传感器网络进行目标被动定位提供了参考。
Basic principle and method for target motion analysis,based on bearing measurement with underwater sensor networks,is studied in this paper.A new target motion analysis（TMA） algorithm which perfectly combines the extended kalman and particle filter（EKF-PF） is presented based on the discrete nonlinear model of passive localization in underwater sensor networks,with system design consideration and iterated method included.Furthermore,through Monte Carlo simulations the algorithm is compared with extended kalman filter（EKF） and particle filter（PF）.The results demonstrate that the proposed new TMA has less computation and better performance because it utilizes current data of measurement to acquire the posteriori estimation of the nonlinear system.
出处
《中国造船》
EI
CSCD
北大核心
2011年第1期90-96,共7页
Shipbuilding of China
关键词
水下传感器网络
纯方位
目标运动分析
粒子滤波
underwater acoustic sensor networks
bearing-only
target motion analysis
particle filter
