一种基于FCM的DV-HOP定位算法  

DV-Hop positioning algorithm based on FCM

作　　者：孙爱晶 李益佳 SUN Aijing;LI Yijia(School of Communications and Information Engineering,Xi’an University of Posts and Telecommunications,Xi’an 710121,China)

机构地区：[1]西安邮电大学通信与信息工程学院,陕西西安710121

出　　处：《西安邮电大学学报》2023年第2期10-18,共9页Journal of Xi’an University of Posts and Telecommunications

基　　金：国家自然科学基金项目(U1965102)。

摘　　要：对锚节点稀少且未知节点分布不均匀的定位环境下距离矢量跳数(Distance Vector Hop,DV-HOP)定位算法定位误差大的问题进行研究,提出一种基于模糊C-均值(Fuzzy C-means,FCM)的DV-HOP定位算法。在DV-HOP定位算法中引入FCM算法将锚节点进行分簇并提出分簇策略,未知节点则按照设定的算法流程选择入簇,在通过簇内定位后进行簇间坐标的融合从而实现全局定位。为了验证基于FCM的DV-HOP定位算法的有效性,将所提算法与DV-HOP定位算法、基于粒子群的DV-HOP(DV-HOP for Particle Swarm Optimization,PSODV-HOP)定位算法以及基于几何改进的DV-HOP(Improved DV-HOP,IDV-HOP)定位算法等3种算法对比。仿真结果表明,基于FCM的DV-HOP定位算法与其他3种定位算法相比,其平均定位误差分别降低了52%、45%与31%,能更好地适应不均匀的节点分布环境。The problem of large positioning error of distance vector hop(DV-HOP)positioning algorithm in the positioning environment of few anchor nodes and uneven distribution of unknown nodes is studied,and a DV-HOP positioning algorithm based on fuzzy C-means(FCM)is proposed.In the DV-HOP positioning algorithm,the FCM algorithm is introduced to cluster the anchor nodes and a clustering strategy is proposed.The unknown nodes are selected into the cluster according to the given algorithm process,and then the coordinates between the clusters are fused after positioning in the cluster to achieve global positioning.In order to verify the effectiveness of the DV-HOP[JP+3]positioning algorithm based on FCM,the proposed algorithm is compared with the DV-HOP positioning algorithm,the DV-HOP particle swarm optimization(PSODV-HOP)positioning algorithm,and the DV-HOP(Improved DV-HOP,IDV-HOP)positioning algorithm based on geometric improvement.The simulation results show that the DV-HOP positioning algorithm based on FCM reduces the average positioning error by 52%,45%and 31%compared with the other three positioning algorithms respectively,which can better adapt to the uneven node distribution environment.[JP]

关 键 词：无线传感器网络 节点定位 距离矢量跳数定位算法 模糊C-均值算法 最小二乘法 

分 类 号：TN92[电子电信—通信与信息系统]