Abstract: The paper proposes an improved combination method based on weight and matrix, which can reduce the calculating complexity and solve the evidence combination problem of highly conflict evidences. The method uses weight value to distinguish the importance among all evidences, and thus it can deal with the conflicting evidences. Matrix is used when evidences are proceeded, which can reduce the computational complication of D-S combination rules. Simulation results show that the improved combination method improves the reliability and rationality of fusion results, and it decreases the time complexity.

Key words: evidence theory, weight, matrix computation, evidence combination, data fusion

摘要： 针对D-S证据理论计算量大和不能有效合成高度冲突证据的不足，提出一种基于加权和矩阵运算的证据组合方法。该方法通过引入权重系数区分不同证据在数据融合中的重要程度，因此能有效合成高度冲突的证据。同时，在证据合成过程中通过引入矩阵运算，有效地降低了算法的计算量。仿真结果表明，与传统的基于D-S证据理论的多传感器数据融合方法相比，改进方法不但能提高证据合成的可靠性和准确性，而且具有较小的计算量。

关键词: 证据理论, 权, 矩阵运算, 证据合成, 数据融合