摘要
针对目前水下目标定位的数据融合算法定位误差较大,精度缺乏良好性能的情况,提出一种应用于水下分布式探测考虑节点可信度的基于线性最小方差估计(LMSE)和递归最小二乘(RLS)的自适应融合算法。该算法采用两级自适应调整得到最优加权因子,首先利用线性最小方差估计(LMSE)算法得到权系数的初始值,然后利用训练节点和递归最小二乘(RLS)算法自适应地调整达到最优。对水下静态和运动目标定位进行的仿真表明,相比单传感器定位,提出的融合算法的定位精度有约1～2个数量级的提高。
In view of the present underwater target location data fusion algorithm accuracy was of large error, This paper proposed a new data fusion algorithm of underwater target positioning for the distributed sensor net- work based on the linear minimum square estimation （LMSE） criterion. The algorithm used two-stage adaptive adjustment to acquire optimized weighting factor by the recursive least squares （RLS） algorithm. Simulation re- sults of underwater static and moving target positioning showed that, compared with the single node localiza- tion, the positioning precision of the fusion algorithm was about one or two orders of magnitude higher.
出处
《探测与控制学报》
CSCD
北大核心
2013年第2期33-36,共4页
Journal of Detection & Control
关键词
水下目标定位
分布式传感器网络
数据融合算法
节点可信度
两级自适应调整
underwater target location
distributed sensor network
data fusion algorithm
node credibility
two-stage adaptive adjustment
