基于主元分析的无线传感器网络数据融合研究

由于传感器节点能量有限且难以补充,无线传感器网络设计的一个重要目标就是在严格的能量约束下最大化网络的生存时间。网内数据融合是减少传感器节点数据传输量、降低网络总能耗,最大化网络生命周期的一种有效途径。本文利用主元分析法PCA中的数据降维技术,提出一种在无线传感器网络的簇内进行数据融合的方法。该方法的主要思想是利用网络的簇头节点对具有一定相关性的多个簇内节点的实测变量进行线性变换,将其转化为少数几个不相关的综合指标后再进行数据转发,从而达到减少网络中的数据传输量和能量消耗的目的。理论分析和实例验证表明了该方法的有效性和实用性。

Research on the PCA-Based Data Aggregation in Wireless Sensor Networks

An important goal in the design of wireless sensor networks is to maximize the lifetime under limited energy because of the battery capacity is difficult to reinforce.In network data aggregation is an effective approach to reducing the amount of data transfer and total energy consumption aiming at lifetime maximization in wireless sensor networks.This paper proposes an in cluster data aggregation method by utilizing the technology of reducing dimensions in the principal component analysis theory.The main idea of this method is that the cluster head node executes lineal transformations to transform the practical correlative variables which transform the member nodes into a few of disrelated compositive indexes, and subsequently transmits these resultant data in order to minimize the data transfer and energy consumption.The theoretical analysis and the instance validation show that this method has considerable validity and practicability.