基于TSK模糊系统的非均匀分簇算法在无线传感网中的应用  

Application of Nonuniform Clustering Algorithm Based on TSK Fuzzy System in Wireless Sensor Networks

作　　者：卢伟 吴延军 苗丹 汪婷 LU Wei;WU Yanjun;MIAO Dan;WANG Ting(Communication and Information Engineering Institute,Guangdong Vocational College of Post and Telecom,Guangzhou 510630,China;Physics and Electronic Engineering Institute,Xingtai University,Xingtai 054000,China)

机构地区：[1]广东邮电职业技术学院信息与通信工程学院,广州510630 [2]邢台学院物理与电子工程学院,河北邢台054000

出　　处：《计算机测量与控制》2023年第12期258-264,共7页Computer Measurement &Control

基　　金：广东省特色创新项目(2021KTSCX259);广东省教育厅课题项目(2022GXJK538);广东省教指委项目(YJXGLW2022Z05)。

摘　　要：无线传感网是一个重要的数据调度工具,但其传感器数量有限,并且极易被无关因素影响;因此,研究提出一种基于区间二型TSK模糊系统和高效数据融合的非均匀分簇路由算法,分簇是为了最大程度地减小网络损耗,延长其生存时间,一般的分簇方法,可能会导致负载失衡,为解决这种热区现象,研究使用区间二型TSK模糊逻辑算法,进行非均匀分级分簇;同时引入高效数据融合技术,将调度过程划分为几个周期,利用时间间隙进行数据采集,并进行降维操作,进一步提升数据传输效率;研究在Matlab平台,对该算法以及自适应分簇层次等其余4种算法进行对照分析实验,并将其分为200 m×200 m和1 000 m×1 000 m的监测范围,实验结果表明,在不同大小的监测区域中,研究使用算法的分簇效果、网络吞吐量以及节点损耗率指标,都明显优于其他算法;其中,在小范围区域内,其剩余能力均值比自适应分簇层次算法提升了49.7%;在大范围区域内,该算法的HND指标比自适应分簇层次算法提升了98.7%;因此,研究采用算法具有极佳的性能。Wireless sensor network is an important data scheduling tool,but the number of sensors is limited and easily affected by irrelevant factors.Therefore,a non-uniform clustering routing algorithm based on interval II TSK fuzzy system and efficient data fusion is proposed.Clustering is to minimize network losses to the greatest extent and extend its life time,while general clustering methods may lead to load imbalance.To solve this hot zone phenomenon,the interval II TSK fuzzy logic algorithm is used to perform the heterogeneous classification and clustering.At the same time,the efficient data fusion technology is introduced to divide the scheduling process into several cycles,the time gap is used to collect the data,and the dimensionality is reduced to further improve the efficiency of data transmission.On the MATLAB platform,this algorithm and other four algorithms,such as adaptive clustering lev-el,are analyzed and compared,and divided into the monitoring ranges of 200 m ×200 m and 1000 m×1000 m.The experimental re-sults show that in monitoring areas of different sizes,the clustering effect,network throughput and node loss rate of the proposed al-gorithm are significantly better than that of other algorithms.Among them,compared with the adaptive clustering hierarchical algo-rithm,in a small range,the average residual capability of the proposed algorithm is increased by 49.7%.In a large range,the HND index of the proposed algorithm by 98.7%.Therefore,the proposed algorithm has an excellent performance.

关 键 词：区间二型 模糊逻辑 数据融合 路由算法 无线传感网 

分 类 号：TP392[自动化与计算机技术—计算机应用技术]