模糊集合下多传感器信息融合算法仿真  

Simulation of Multi-Sensor Information Fusion Algorithm under Fuzzy Set

作　　者：靳双燕 李浩亮[2] JIN Shuang-yan;LI Hao-liang(School of Information Engineering,Zhengzhou Technology and Business University,Zhengzhou Henan 450000,China;School of Electrical and Information Engineering,Zhengzhou University,Zhengzhou Henan 450000,China)

机构地区：[1]郑州工商学院信息工程学院,河南郑州450000 [2]郑州大学电气与信息工程学院,河南郑州450000

出　　处：《计算机仿真》2024年第7期276-280,共5页Computer Simulation

基　　金：河南省高等学校重点科研项目计划(22B510014)。

摘　　要：多传感器数据融合通常涉及大量数据,受到噪声、不完整性、不准确性等因素的影响,导致融合结果的不确定性增加。为此,提出一种基于模糊集合的多传感器信息融合算法。利用局部保持投影(Locality Preserving Projection, LPP)处理多传感器信息,采用主成分分析(Principal Component Analysis, PCA)分离特征值较大的反射信号,删除特征值较小的随机噪声。利用隶属函数得到不同传感器提供的可信度,将支持度以及可信度转换为基本概率分配函数,引入证据理论(Dempster-Shafer, D-S),实现多传感器信息融合优化。仿真分析表明,所提方法可以得到高精度和高效率的多传感器信息融合结果,峰值信噪比达60dB以上,信噪比一直处于12dB以上,融合最长耗时仅为2.01ms,使其融合性能得到有效优化。Typically,the data fusion from multiple sensors involves a large amount of data.Affected by factors such as noise,incompleteness and inaccuracy,the uncertainty of the fusion result also in creases.To address this,a multi-sensor information fusion algorithm based on fuzzy sets was proposed.Firstly,Loc ality Preserving Projection(LPP)was utilized to process multi-sensor in formation.Then,Principal Component Analysis(PCA)was employed to separate reflection signals with larger eige nvalues and remove random noise with smaller eigenvalues.Next,the membership function was used to calculate the credibility provided by different sensors.Mor eover,both the support and credibility were converted into basic probab ility assignment functions.Finally,the Dem pster-Shafer(D-S)theory was introduced to achieve the optimization of mu lti-sensor information fusion.Simulation a nalysis results show that the proposed method can obtain high-precision an d efficient fusion results.The peak signal-to-noise ratio reaches over 60dB,and the signal-to-noise ratio remains a bove 12dB.The longest fusion time is only 2.01ms.Therefore,the fusion performance is effectively optimized.

关 键 词：模糊集合 多传感器 信息融合 优化 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]