论文部分内容阅读
由于硬件与网络资源的极度受限,无线传感器网络(WSN)的故障诊断成为该领域内的一个研究难点.针对现有诊断方法网络开销大、方法复杂等缺陷,提出了一种基于感知数据的故障诊断(DSD)方法.以部署在实际森林环境中的GreenOrbs系统收集的大量感知数据为基础,通过分析感知数据特征分类的方法,建立与网络故障之间的对应关系,以自主学习的方式不断演化故障知识库,确定故障类型.实验结果表明,与其他诊断方法相比,DSD具有网络通信负担小、资源消耗低、诊断效率高等优点,并支持在大规模WSN的实际部署.
Due to the extremely limited hardware and network resources, the fault diagnosis of wireless sensor network (WSN) has become a difficult research in this field.Aiming at the shortcomings of existing methods such as large network overhead and complicated methods, this paper proposes a new method based on perceptual data Fault Diagnosis (DSD) Method Based on a large amount of perception data collected by GreenOrbs system deployed in real forest environment, by analyzing the method of feature classification of perceptual data, it establishes the correspondence with network failure and learns continuously Evolution fault knowledge base to determine the fault type.The experimental results show that compared with other methods, DSD has the advantages of small network communication load, low resource consumption and high diagnostic efficiency, and supports the actual deployment in large-scale WSN.