一种基于泰勒级数多元变量展开模型定位算法

A localization algorithm based on Taylor series multivariable expansion model

为了提高无线传感器网络节点定位精度,构建了增加未知节点与未知节点间的距离信息的泰勒级数多元变量展开定位模型.在对该算法的求解过程中,首先利用最大似然估计法得到未知节点的初始位置,再运用加权最小二乘法计算其最优值作为未知节点的估计位置.仿真测试了不同距离测量误差和已知节点数目对定位误差的影响,以及算法的累计分布函数.结果表明,该算法能够有效提高节点定位精度.

In order to improve positioning accuracy in wireless sensor networks ,a new Taylor se‐ries multivariable expansion localization model is established by the method of adding the dis‐tances information betw een unknow n nodes . In process of the algorithm solution , firstly the maximum likelihood estimation is utilized to obtain initial values of unknow n nodes . T hen ,its optimal values are calculated as the estimated location of unknown nodes by the weighted least squares method .To evaluate the performance of this algorithm ,simulations test the impact of different distance measurement error and the number of know n nodes on positioning error ,and the cumulative distribution function of algorithm .Simulation results show that the proposed al‐gorithm has achieved better performance on positioning accuracy and efficiency .