作　　者： ; ; ; ;

机构地区： 华南理工大学机械与汽车工程学院

出　　处： 《计算机测量与控制》 2012年第3期645-647,共3页

摘　　要： 针对固定预测时间间隔下目标机动对无线传感器网络目标定位预测效果影响较大的问题,分析目标预测误差产生机理及主要影响因素,提出WSN目标定位动态预测方法。该方法根据目标预测模型构造预测时间增量函数,通过固定预测时间增量函数取距离自变量值动态调节目标预测时间间隔,实现目标运动状态自适应动态目标预测。仿真平台分别应用运动学预测、粒子滤波预测方法建立预测模型,并进行目标预测实验;结果表明,目标机动情况下,目标动态预测方法误差相比固定预测时间时间间隔方法分别减小18.5%、12.8%,动态目标预测方法能较好改善机动性目标预测效果,增强预测方法对目标运动变化的自适应能力。 Aiming at the problem of target tracking in wireless sensor networks （WSN） based on fixed prediction time interval, a method of target tracking dynamic prediction in WSN is proposed. Error mechanism of target tracking and the major factors which influence the prediction error are researched. Increment function of prediction time is built based on target prediction model. Using different value for distance independent variable of increment function of prediction time, adjusting prediction time interval dynamically and achieving target dynamic pre- diction based on target motion state adaptively. Prediction models are built using quadratic polynomial motion modeling method and particle filter method individually. Experimental results show that prediction effect and adaptive ability of prediction method for maneuvering target are improved using dynamic prediction method. When target maneuvers, prediction error of dynamic method decreased by 18. 5% and 12.8% respectively in compare with fixing prediction time interval method.

关 键 词： 无线传感器网络 目标定位 动态预测 预测误差

领　　域： [自动化与计算机技术] [自动化与计算机技术]