无线传感器网络中基于减法聚类的定位算法

Localization Algorithm Based on Subtractive Clustering in Wireless Sensor Networks

定位算法是无线传感器网络中的关键技术。文中在传统的Dv-Hop算法的基础上,找出其产生误差的主要原因,即对未知节点与锚节点之间的估计距离做出了修正,提出一种无线传感器网络中基于减法聚类的定位算法。该算法用减法聚类的方法,根据节点自身的密度,选出锚簇头节点,使锚簇头节点在锚节点密集处产生;同时用所有锚簇头节点平均每跳距离的均值作为未知节点的网络平均每跳距离,提高了定位精度,减少了定位过程中的能量消耗。仿真实验表明,该算法比Dv-Hop算法有更好的定位精度和鲁棒性。

Node localization algorithm is one of the most critical technologies in wireless sensor networks. Based on Dv-Hop algorithm, point out the main reason for the error. Then a new algorithm named localization algorithm based on subtractive clustering in wireless sen- sor networks is proposed. In new algorithm , the subtractive clustering is used to choose the anchor cluster-head where has relatively high density;Besides, the average distance of all anchor cluster-heads is used as unknown nodes hop-distance in network. Compared with the traditional Dv-Hop localization algorithm , the improved algorithm is more reliable and more robust.