基于半定规划的恶劣环境下定位修正算法  

Research on Semidefinite Programming Localization Enhancement in Complex NLOS Wireless Sensor Network Environment

作　　者：刘栓[1] 刘直良[1] 

机构地区：[1]黄淮学院信息工程学院

出　　处：《传感技术学报》2017年第5期766-771,共6页Chinese Journal of Sensors and Actuators

基　　金：河南省科技厅重点攻关计划项目(9412015Y1305);国家自然科学基金项目(U1404602)

摘　　要：无线传感网络WSNs(Wireless Sensor Networks)的定位精度严重受到环境的影响,尤其是高噪声电平和非视距连接的恶劣环境,定位精度急剧下降。为此,提出基于半定规划的恶劣环境下定位修正算法,记为ESDP_O算法。该修正算法以半定规划ESDP(Edge-Semi-Definite Programming)算法为基础,旨在提高定位精度,并降低算法复杂性,进而减少定位时间。ESDP_O算法通过引用抖动矩阵,对ESDP算法进行修改,提高了算法在恶劣环境的健壮性。同时,ESDP_O算法通过寻找低秩解,减少高噪声和非视距偏差。仿真结果表明,在高噪声和非视距NLOS(Non Line of Sight)的恶劣环境下,ESDP_O算法的定位精度优于基于同类算法,并且降低了定位的复杂度。The accuracy of localization in wireless sensor networks （WSNs） depends on noise level and the presence of nonline of sight （NLOS） connections. In this letter, a novel Edge semidefinite programming （ESDP） Robust locali- zation algorithm is proposed and its goal is to improve the accuracy and reduce the algorithm complexity, and it reduce the required time for wireless sensor network localization in harsh environments,which is marked as ESDP_O. We modified ESDP relaxation by a perturbation matrix in order to make it robust against large errors in distance measurements. ESDP_O algorithm modified ESDP in order to find a low rank solution, and introduced a new class of ESDP-based relaxation that is robust against high level noise and NLOS biases. Simulation results confirm that our proposed method outperforms others when the majority of connections are NLOS,noise level is high,and the proposed ESDP_O algorithm can effectively reduce the computational complexity.

关 键 词：无线传感网络 定位 半定规划 非视距 抖动矩阵 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]