Data fusion algorithm based on fuzzy logic

Based on fuzzy logic theory, a new data fusion method for target identification using D-S evidence theory (Dempster-Shafer theory) is presented. By acquiring signals from multiple sources, the belief function assignment value of each transducer is obtained from extracted eigenvalue by using fuzzy logic. The fusion belief function assignment is received by using D-S evidence theory. To solve the problem of combing evidences with high degree of conflict, a modified D-S evidence theory combination rule is introduced. The problem of inconsistent evidence is handled using the method. More accurate result of target identification can be obtained. A comparison of the results of target identification based on separate original data with those based on fusion data shows that the accuracy of target identification is increased.