Analysis and improvement for Proportional Conflict Redistribution rules

Abstract

The Proportional Conflict Redistribution (PCR) rules based on Dezert-Smarandache theory (DSmT) is a useful method for dealing with uncertainty problems. It is more efficient in combining conflicting evidence. Therefore, it has been successfully applied in identity identification. However, there exist shortcomings in PCR rule. So in this paper we propose a new improved rule which is based on new Proportional Conflict Redistribution. The six PCR rules (PCR1-PCR6) and improved PCR rule are analyzed and compared through numerical examples, and the results show that the improved rule is effective.

Fusing navigation and vision information with the Transferable Belief Model: Application to an intelligent speed limit assistant

“…In this case, both P CR5 and P CR6 behave similarly [35]. An interesting approach, based on the P CR6, is presented in [42]. Similarly to the approach presented in this paper, the first step in [42] consists in the conflict evaluation using a specific measure used as a weighting factor to combine the orthogonal sum results with those provided by a PCR6.…”

“…An interesting approach, based on the P CR6, is presented in [42]. Similarly to the approach presented in this paper, the first step in [42] consists in the conflict evaluation using a specific measure used as a weighting factor to combine the orthogonal sum results with those provided by a PCR6. The comparison of the P CR5 and the redistribution rule adopted in [42] represents an interesting topic for further researches.…”