Author Login Chief Editor Login Reviewer Login Editor Login Remote Office

Computer Engineering ›› 2010, Vol. 36 ›› Issue (7): 76-78,8.

• Software Technology and Database • Previous Articles     Next Articles

High Complexity Optimization for DSmT Mixing Rule

CHEN Kai1,2, YANG Jun-an1,2, CHEN Hao1,2   

  1. (1. 309 Research Lab, Electronic Engineering Institute, Hefei 230037; 2. Key Laboratory of Electronic Restriction Technology, Anhui Province, Electronic Engineering Institute, Hefei 230037)
  • Received:1900-01-01 Revised:1900-01-01 Online:2010-04-05 Published:2010-04-05

DSmT混合规则的高复杂度优化

陈 凯1,2,杨俊安1,2,陈 昊1,2   

  1. (1. 电子工程学院309研究室,合肥 230037;2. 电子工程学院安徽省电子制约技术重点实验室,合肥 230037)

Abstract: Aiming at the problem of high complexity in calculation and memory for DSmT mixing rule, this paper proposes a matrix calculation algorithm. This algorithm codes the basic element of discernment frame into a set, simplifies procedure for hybrid rule of combination. Experimental results show that this algorithm makes execution time exponential rising with variation of discernment frame size, and linear variation with variation of evidence source and focal element, it avoids high complexity effectively.

Key words: combinatorial matrix, mixing rule, complexity

摘要: 针对DSmT混合规则计算和存储复杂度高的问题,提出一种矩阵计算法。该算法对辨识框架的基本元素进行集合编码,从而简化混合规则的组合过程。实验结果表明,该算法使执行时间随辨识框架大小呈指数增加,随证据源数目和焦元数呈线性变化,能有效避免高复杂度。

关键词: 组合矩阵, 混合规则, 复杂度

CLC Number: