An Entropy Method For Multisource Data Fusion

Abstract- The present paper proposes a generic model of the multisource data fusion in the frame-work of the theory of information, with closer at-tention being given the dierent nature of data processed in common cases. This model that we have called entropy model is then used to elabo-rate processing methods able to face specic problems that may arise when multisource systems are imple-mented to achieve functions like classication and pattern recognition, matching of ambiguous obser-vations, estimation, detection or tracking. Crucial practical problems to data fusion are more specif-ically dealt with, such as information representa-tion, appropriate combination processing and deci-sion making. Some clues are given on the practical use and implementation of such an approach, for ex-ample, in the distributed estimation problem