摘要
为了解决物联网定位中面临的诸多问题,对经典定位算法进行了仿真研究,并通过对仿真结果进行对比分析得到将最小二乘法、最大似然估计与拉格朗日日乘法结合起来使用的估计目标位置方法,即统一地利用加权最小二乘法/受限制的加权最小二乘法来计算各种算法.根据测量结果得到的非线性方程能够转化为线性方程,而这些线性方程应用加权最小二乘法与拉格朗日乘法来处理,这种方法可以很容易应用到其他算法上.在噪声相对较小时,上述算法性能已经可以达到克拉美罗下界和无偏性.
To solve the many problems faced in the location of Internet of Things, the classic positioning methods are studied. By the comparative analysis of the simulation results, an effectively method of estimated target location(least squares, maximum likelihood estimation and Lagrange combined positioning algorithm)were got, and an generalized weighted least square method / limited weighted least square method can be used to calculate various algorithm. The nonlinear equation obtained from the result of measurement can be transformed to linear equation. And then the linear equation can be solved with weighted Least squares and Lagrange algorithm. This method can be easily applied to other algorithm too. When noise is relatively low, the performance of the algorithm can reach the Cramer Rao lower bound and be unbiased.
出处
《武汉工程大学学报》
CAS
2015年第3期68-73,共6页
Journal of Wuhan Institute of Technology
基金
湖北省教育科学"十二五"规划2013年度立项课题(2013B060)
关键词
定位算法
最小二乘法
最大似然估计
拉格朗日日乘法
positioning algorithm
least squares
maximum likelihood estimation
lagrange algorithm