Bayes分析中多源信息融合的最大熵-矩估计方法

Multi-source Information Fusion Based on MEMM in Bayes Analysis

本文提出了一种基于最大熵-矩估计理论的多源验前信息融合方法,简称为MEMM与现有其它融合方法不同的是,该方法完全从数据本身出发,不仅考虑了各验前信息源的可信度,还考虑了如何尽可能减少分析者的主观影响,得到的是最保守的融合估计结果。文章最后用仿真数据说明了该方法在实际工程中应用的有效性。

The approach proposed here of fusing multi -source iniormation by coupling the principle of maximum entropy with the moment-matching method, be named simply as MEMM.The inherent feature of MEMM is that the population is drawn from the genetic data without depending on any analysts' judemerit,an approach that is obviously different from many current techniques.At last, an example was given to illustrate the efficiency of this algorithm.