基于图模型的传感器节点定位方法  

A method for sensor node localization based on graph model

作　　者：赵海兵 蒋俊正[1] ZHAO Haibing;JIANG Junzheng(School of Information and Communication,Guilin University of Electronic Technology,Guilin 541004,China)

机构地区：[1]桂林电子科技大学信息与通信学院

出　　处：《桂林电子科技大学学报》2019年第1期28-34,共7页Journal of Guilin University of Electronic Technology

基　　金：国家自然科学基金(61761011,61371186);广西自然科学基金(2017GXNSFAA198173,2017GXNSFBA198137);桂林电子科技大学研究生教育创新计划(2018YJCX34)

摘　　要：针对传感器网络中节点间测距存在误差致使定位不准确的问题,提出了一种基于图模型的传感器节点定位方法。该方法在图模型基础上充分考虑了节点间的连通性,利用节点间距离设置了目标函数中各求和项的权重值。考虑到目标函数是关于节点位置的高度非线性非凸函数,难以求解,采用两步法求解该优化问题。第一步,利用三点定位法对节点进行粗略的初步定位;第二步,基于三点定位得出的初始值,结合二阶泰勒近似给出的修正海森矩阵,采用修正牛顿法对定位问题进行求解。仿真实验表明,与现有方法相比,新方法在不同程度的测距误差下定位更快更准确。A method is presented to estimate the positions of sensor nodes based on graph model under inaccurate pairwise distance measurements.Because of the connectivity,the weight of each sum in the objective function is set by the distance between nodes.Considering that it is hard to the solve objective function which is a highly nonlinear nonconvex function with respect to the positions of nodes,a two-step scheme is proposed.The first step is to locate the nodes roughly by three-point positioning algorithm.Starting from the initial value by the first step,the modified Newton method is applied to further solve the optimization problem,specially the modified Hessian matrix is given by second-order approximation of the Taylor’s series.Compared with the existing methods,the proposed method can be positioned faster and more accurately.

关 键 词：无线传感器网络 节点定位 无约束优化 三点定位 修正牛顿法 

分 类 号：TN911.7[电子电信—通信与信息系统]