Multi-Sensors Information Fusion Based on Momentis Method and Euclid Distance

Abstract:

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.

• Periodical: Advanced Materials Research (Volumes 383-390)
• Pages: 5447-5452
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.383-390.5447
• Online since: November 2011
• Authors: Feng Yong Lang, Xiao Gang Li
• Keywords: Euclid Distance, Information Fusion, Momentis Method, Multi-Sensors Information, Prior Distribution

References

[1] Jing Feng. Research on Methods and Applications of Reliability Information Fusion for Complex System with Small Sample test, National University of Defense Technology. 2004, pp.24-28.

[2] Hanyan Huang, Xiaojun Duan, Zhengming Wang. A Novel Posterior-weighted Bayesian Estimation Method Considering the Credibility of the Prior Information, Acta Aeronautica et Astronautica Sinica. 2008, vol29(5), pp.1245-1251.

[3] Jian Chai. Information fusion method and application on System Reliability Assessment, Northwestern Polytechnical University, 2006, pp.24-28.

[4] Xie M, Hong G Y, Wbhiin C, Software reliability Prediction incorporating information from a similar Project, The Journal of system and Software, 1999, vol49, pp.43-48.

DOI: 10.1016/s0164-1212(99)00065-5

[5] Lesley W, John Q, Building Prior distributions to support Bayesian reliability growth modeling using expert judgment, Reliability Engineering and System Safety, 2001, vol74, pp.117-128.

DOI: 10.1016/s0951-8320(01)00069-2

[6] Xu-Qing TANG, Ping ZHU, Jia-Xing CHENG. Analysis of Structural Clustering Based on Normalized Metric, Pattern Recognition and Artificial Intelligence. 2009, pp.678-688.

[7] Yu rong LI, Guo bo XIANG. A Linear Discriminant Analysis Classification Algorithm Based on Mahalanobis Distance, Computer Simulation. 2006, pp.86-88.

[8] Wei Yang. Research of uncertain information fusion based on belief function theory, National University of Defense Technology. 2008, pp.16-17.

[9] Xuezhong Yin, Jiegui Wang. A New Combination Rule of Heterogeneous Sources Data Fusion on Conflict Evidences. Journal of Astronautics. 2009, 30(04), pp.1663-1666.

[10] Fang Genhai. Research on the Multi-source Information Fusion Techniques in the process of Reliability Assessment, Hefei University of Technology. 2006, pp.36-37.