基于MCMC的无线传感器网络目标定位算法  

Target localization in wireless sensor networks based on Markov-chain Monte-Carlo

作　　者：张沛朋[1] 李冬锋 ZHANG Pei-peng;LI Dong-feng(Art Design Department,Jiyuan Vocational and Technical College,Jiyuan 459000,China;School of Information Engineering,North China University of Water Resources and Electric Power,Zhengzhou 450045,China)

机构地区：[1]济源职业技术学院艺术设计系,河南济源459000 [2]华北水利水电大学信息工程学院,河南郑州450045

出　　处：《计算机工程与设计》2022年第7期1864-1871,共8页Computer Engineering and Design

基　　金：河南省基础与前沿技术研究计划基金项目(142300410391)。

摘　　要：为解决基于时延和多普勒频率的无线传感器网络运动目标定位问题,提出一种基于马尔科夫链-蒙特卡罗(Markov-chain Monte-Carlo,MCMC)的直接定位算法。基于最大似然准则从各传感器节点接收信号模型中推导目标位置和速度估计的优化函数;针对该优化函数难以得到闭式解的问题,将优化函数转化为马尔科夫链的稳态分布,利用MCMC方法对目标位置和速度参数分布进行抽样,得到目标位置和速度参数的样本,通过统计样本均值得到目标位置和速度的估计值。实例仿真计算结果表明,该算法比现有算法具有更高的定位精度、稳健性和计算效率,在一般信噪比条件下,性能逼近克拉美罗界。To deal with the problem of moving target localization in wireless sensor networks(WSN)based on time delay and Doppler frequency,a direct position determination(DPD)algorithm based on Markov-chain Monte-Carlo(MCMC)was proposed.The optimization function of target position and velocity estimation was derived from the received signal model based on the maximum likelihood criterion.To address the problem that the optimization function can hardly be solved analytically,the optimization function was transformed into the steady-state distribution of a Markov chain,and the MCMC method was applied to draw samples from the distribution of target position and velocity parameters.The target position and velocity estimate was then obtained by computing the mean of the drawn samples.Results of the simulation indicate that the proposed algorithm has higher localization accuracy,robustness and computational efficiency than existing algorithms,and its performance is close to the Cramer-Rao lower bound under moderate signal-to-noise ratio conditions.

关 键 词：无线传感器网络 直接定位 时延 多普勒频率 马尔科夫链-蒙特卡罗 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]