Multi-sensors information fusion based on momentis method and Euclid distance

A method based on data level fusion for solving the problem of multi-sources information fusion is discussed in this paper. Firstly, the credibility of multi-sources information is calculated based on the Euclid distance. Calculating the distance between the multi-sensors information and the experiment data, the shorter the distance is the better the degree of association between prior information and the population information is. The distance of multi-sources information is normalized to be the credible weight. Secondly, the unknown parameters of various probable distribution function is estimated with momentis method, in order to establish an optimal fused prior distribution. According to the momentis method, different random variable may have the same expectation and variance, in this paper we take some distribution function commonly seen for instance. Finally, demonstrations are carried out with MATLAB simulation to validate this method.