Computer Engineering and Applications ›› 2014, Vol. 50 ›› Issue (20): 86-90.

Previous Articles     Next Articles

Multi-sensor weighted data fusion method using LMS algorithm

CHEN Ziyu, ZHANG Xinwei, YE Lingyun   

  1. College of Biomedical Engineering and Instrument Science, Zhejiang University, Hangzhou 310027, China
  • Online:2014-10-15 Published:2014-10-28

基于LMS算法的多传感器数据加权融合方法

陈咨余,张新伟,叶凌云   

  1. 浙江大学 生物医学工程与仪器科学学院,杭州 310027

Abstract: For the optimal weighed fusion of multi-sensor data with the variance of the observation error is unknown and variant, this paper presents a new multi-sensor weighed data fusion method based on LMS (Leanest Mean Square) algorithm. This method converts the optimal weighed fusion of multi-sensor data to the optimal solution search on the performance surface of mean square error of estimated value. Adaptive LMS algorithm is applied to adjust the weighting coefficient of sensor. Simulation results show the effectiveness of this method.

Key words: multi-sensor, data fusion, adaptive

摘要: 针对目前多传感器数据融合过程中,传感器观测噪声不易确定,提出了一种基于LMS算法的多传感器自适应加权数据融合方法。该方法将传感器最优加权系数的求解,转化为估计值的均方误差性能表面的最优解搜索,通过加入自适应阶段,采用自适应最小均方误差(LMS)算法调整传感器加权系数。仿真结果表明该方法的有效性。

关键词: 多传感器, 数据融合, 自适应