Research on the fuzzy set theory of evidence fusion algorithm with time-varying in multi-sensor detection network

As a relatively weaker certain deterministic inference algorithm than probability theory, Dempster-Shafter (D-S) evidence fusion theory is widely used in the multi-sensor detection network, By measuring uncertain information reasonably and comprehensive can improve the detection accuracy of the distributed detection network. Considered the evidence of time variability, An evidence fusion algorithm based on fuzzy set theory is proposed in this paper. The basic probability assignment is obtained by using fuzzy rough set theory to get a full measure of uncertainty information of each sensor, For dealing with the deficiency of the evidence conflict, Information content of each evidence was weighted for modifying evidence source. Then use Demspter evidence combination rule, the fusion result can be obtained by combining the new evidence, Transferable belief mode can translate reliability layer to probability layer. Then we can use mature decision-making mechanisms of probability level for decision-making. Finally, the simulation result shows that the fusion results have higher precision and reliability compared with other methods. Target recognition rate can be improved from 40% to 89%, which prove that the proposed fusion algorithm can effectively improve the accuracy of target identification.