Scholarlink

<p>The diverse nature of Wireless Sensor Network (WSNs) and the placement of sensor node (SN) deployment within these networks provide obstacles for fault identification and diagnosis. However, WSNs are susceptible to interruptions of regular processes resulting from the failure of individual sensor nodes, which may occur due to environmental degradation, hardware reliability, and battery life of each sensor node. Faulty sensor nodes can generate inaccurate data, which substantially impacts their network’s trustworthiness, quality, and longevity. The proposed methodology utilizes an intelligent strategy based on the evidence theory and fuzzy fault tree analysis to detect and diagnose faults in the sensor nodes. Ensuring data reliability involves prioritizing fault diagnosis in sensor nodes. In this context, an effective method of fault diagnosis is presented using fuzzy fault trees and evidence theory, and the Improved Artificial Jellyfish Optimization (IAJO) algorithm. The proposed framework is employed to diagnose and classify network faulty sensor nodes as normal, experiencing battery condition faults, sensor circuit faults, or communication module faults. The efficacy of the proposed framework was assessed using MATLAB via core analysis and simulations. The proposed framework demonstrates more robustness and superior efficiency than other approaches. The numerical findings indicated elevated diagnostic accuracy, accompanied by a reduced false positive rate and decreased energy use. </p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

An fuzzy fault tree analysis with evidence theory based efficient fault node diagnosis in wireless sensor networks

  • 期刊论文

  • 发表日期：

    2025年9月25日

查看全文
  • Nagarajan B,
  • Santhosh Kumar SVN

摘要

The diverse nature of Wireless Sensor Network (WSNs) and the placement of sensor node (SN) deployment within these networks provide obstacles for fault identification and diagnosis. However, WSNs are susceptible to interruptions of regular processes resulting from the failure of individual sensor nodes, which may occur due to environmental degradation, hardware reliability, and battery life of each sensor node. Faulty sensor nodes can generate inaccurate data, which substantially impacts their network’s trustworthiness, quality, and longevity. The proposed methodology utilizes an intelligent strategy based on the evidence theory and fuzzy fault tree analysis to detect and diagnose faults in the sensor nodes. Ensuring data reliability involves prioritizing fault diagnosis in sensor nodes. In this context, an effective method of fault diagnosis is presented using fuzzy fault trees and evidence theory, and the Improved Artificial Jellyfish Optimization (IAJO) algorithm. The proposed framework is employed to diagnose and classify network faulty sensor nodes as normal, experiencing battery condition faults, sensor circuit faults, or communication module faults. The efficacy of the proposed framework was assessed using MATLAB via core analysis and simulations. The proposed framework demonstrates more robustness and superior efficiency than other approaches. The numerical findings indicated elevated diagnostic accuracy, accompanied by a reduced false positive rate and decreased energy use.