摘要
为了提高数据处理的精度 ,提出了一种适用于单传感器系统的数据融合方法。该方法先将来自单传感器的n个样本观测值细分成 k组 ,并把各细分组所对应的样本均值假想为来自 k个不同传感器的样本观测值 ,借助于极大似然函数估计法导出数据融合公式。基于该融合公式 ,证明了在单传感器系统中 ,当样本观测值一定时 ,分组数据融合的估计效果优于单组算术平均的估计效果 ;若再细分原有的分组 ,则细分后的融合估计效果优于细分前的融合估计效果。
To improve the precision of data processing, a method of the data fusion suitable for the single sensor system is presented. The n sample values comed from the single sensor is divided into k groups. The n sample mean of each group is assumed to be sample values of k different sensors. And the formula of data fusion by the maximum likelihood estimation method is established and the following conclusions are proved. In a single sensor system when the sample amount is constant, the estimation result of data fusion to subdivided into groups is superior to arithmetic average of single group. If original groups are subdivided again, the result before subdividing is not superior to one after subdividing. Actual data analysis shows that the conclusion is correct.
出处
《数据采集与处理》
CSCD
北大核心
2005年第1期88-90,共3页
Journal of Data Acquisition and Processing
基金
浙江省教育厅科研计划基金 ( 2 0 0 2 0 2 70 )资助项目