D-S Evidence Theory Fusion Method Based on Forecast Reliability
Jianpei Zhang, Li Wang, Lejun Zhang, Jing Yang · 2010
D-S evidence theory is one of the most widely used methods in data fusion, and the problems including evidence conflict and low fusion efficiency are also the focus subject that many scholars are studying. The idea of forecast reliability is introduced into this paper for the problems. First, an improved method based on forecast reliability coefficient is presented. Secondly, on the basis of maintaining all the characters of original combination rule, forecast reliability coefficient is introduced into combination rule to improve data fusion speed and resolve evidence conflict problem. Finally, the simulation experiment proved the effectiveness of the proposed algorithm.