基于扇形重叠区域聚类分析的定位算法
首发时间:2015-12-24
摘要:enewcommand{ aggedright}{leftskip=0pt ightskip=0pt plus 0cm} aggedright 针对无线传感器网络中基于RSSI的加权质心定位算法无法满足边缘地区对目标准确定位的问题,提出一种基于扇形重叠区域聚类分析的定位算法 (Localization algorithm based on clustering analysis of overlapping sector areas, LA-CAOSA)。 该算法首先利用高斯筛选剔除误差相对较大的RSSI测量值;其次,根据RSSI对定位误差分布的影响,建立基于聚类分析的扇形重叠区域,在不同划分区域中利用参考节点间的几何位置加入修正系数进行距离修正。同时,利用动态加权质心定位算法进行定位,得到未知节点的位置。实验结果表明,该算法可以有效地提高定位准确度,能够满足目标的定位追踪需求。
关键词: 无线传感器网络; 定位; 聚类; 扇形重叠区域; 高斯筛选
For information in English, please click here
Localization algorithm based on clustering analysis of overlapping sector areas
Abstract:enewcommand{ aggedright}{leftskip=0pt ightskip=0pt plus 0cm} aggedright Aiming at the problem that RSSI-based weighted centroid localization algorithms can not meet the requirement of accurate positioning for targets in edge regions in wireless sensor network (WSN), we present a localization algorithm based on clustering analysis of overlapping sector areas (LA-CAOSA). The proposed algorithm first uses Gaussian filter to remove RSSI measurements with relatively large error. Then, according to the influence of RSSI on the distribution of positioning error, the algorithm establishes overlapping sector areas based on clustering analysis and introduces correction coefficients in different zones by using the geometric positions of the reference nodes. At the same time, LA-CAOSA adopts dynamic weighted centroid localization algorithms to locate unknown nodes. Experimental results show that the proposed algorithm can effectively improve the positioning accuracy and meet the needs of target tracking.
Keywords: WSN localization clustering overlapping sector areas gaussian filter
论文图表:
引用
No.4672022103438914****
同行评议
共计0人参与
勘误表
基于扇形重叠区域聚类分析的定位算法
评论
全部评论0/1000