系统工程与电子技术 ›› 2018, Vol. 40 ›› Issue (3): 509-517.doi: 10.3969/j.issn.1001-506X.2018.03.04

• 电子技术 • 上一篇    下一篇

基于犹豫模糊集统计相关系数的多源异类数据融合识别

关欣, 孙贵东, 衣晓, 赵静   

  1. 海军航空大学电子信息工程系, 山东 烟台 264001
  • 出版日期:2018-02-26 发布日期:2018-02-24

Multi-source heterogeneous data fusion recognition based on statistical correlation coefficients between hesitant fuzzy sets

GUAN Xin, SUN Guidong, YI Xiao, ZHAO Jing   

  1. Department of Electronics and Information Engineering, Naval Aviation University, Yantai 264001, China
  • Online:2018-02-26 Published:2018-02-24

摘要:

为解决由犹豫模糊数、直觉模糊数、区间数和实数4类基本数据组成的多源异类数据的融合识别问题,在犹豫模糊框架内,提出犹豫模糊集相关系数计算方法进行识别判定。首先,将多源异类数据用犹豫模糊数统一描述;其次,分析现有的犹豫模糊集相关系数的局限性,提出了既满足统计学直觉,又不需要各犹豫模糊数中隶属度个数相同,并具有更强数学概念的犹豫模糊集统计相关系数;最后,应用到多源异类数据的融合识别中,利用最大相关系数准则进行识别判定。仿真算例对比分析验证了犹豫模糊集统计相关系数的有效性,并具有精度高、区分度好的优点。

Abstract:

In order to solve the multi-source heterogeneous data fusion recognition problem in which the multi-source heterogeneous data is expressed as hesitant fuzzy element (HFE), intuitionistic fuzzy number (IFN), interval number and real number, a correlation coefficient between hesitant fuzzy sets (HFSs) is proposed to recognize these data in the HFS domain. Firstly, the multisource heterogeneous data is transformed into the HFS domain. After towards, pointing out the weakness of the existing correlation coefficients between HFS, we propose a correlation coefficient between two HFSs which has advantages in three aspects: firstly it conforms to the statistics intuition, secondly, it is not necessary to have the same length in membership, and thirdly, it is more theoretical in mathematics. Finally, apply the proposed correlation coefficient to recognize the multi-source heterogeneous data based on the principle of the maximum correlation coefficient. The simulation examples demonstrate the vadidity of the proposed correlation coefficient with a comparision analysis, and prove its advantages in high precision and discrinination.