Computer Engineering and Applications ›› 2007, Vol. 43 ›› Issue (27): 124-126.
• 网络、通信与安全 • Previous Articles Next Articles
YU Hao,CHEN Xiong,FAN Jing-jing
Received:
Revised:
Online:
Published:
Contact:
于 浩,陈 雄,范晶晶
通讯作者:
Abstract: Resolve the problem of non-linear least square to realize maximum likelihood target localization based on energy.Get the minimum of target function using Gauss-Newtonmethod.Propose the selection principle of the initial value and pace length based on the analysis of the character of target function,and avoid local-minimum in the Gauss-Newton iterative process.When keeping high localization accuracy of search algorithm,this algorithm improves the speed of localization,and realizes the integration of speed and accuracy.
Key words: energy, energy ratio, maximum likelihood, Gauss-Newton
摘要: 将基于能量的最大似然法的目标定位问题转化为求解非线性最小二乘问题,并引入高斯-牛顿法来求解目标函数最小值,通过分析目标函数的特点,提出了能量法定位中高斯牛顿迭代的初始值和步长的选取原则,防止了迭代算法进入局部最小。这种算法在保证搜索法的定位精度的前提下,提高了定位速度,实现了精度与速度的结合。
关键词: 能量, 能量比例, 最大似然, 高斯-牛顿
YU Hao,CHEN Xiong,FAN Jing-jing. Application of Gauss-Newton to target localization based on energy[J]. Computer Engineering and Applications, 2007, 43(27): 124-126.
于 浩,陈 雄,范晶晶. 高斯-牛顿法在基于能量的目标定位中的运用[J]. 计算机工程与应用, 2007, 43(27): 124-126.
Add to citation manager EndNote|Ris|BibTeX
URL: http://cea.ceaj.org/EN/
http://cea.ceaj.org/EN/Y2007/V43/I27/124