Scholarlink

<p>A wireless sensor network (WSN) can cause disruption in the communication if there are any malfunctioning nodes. Identification of malfunction nodes is a difficult task because of the variety of operating conditions and the limited resources of sensor nodes. The defect detection mechanism described in the paper is proposed to be implemented by using an Adaptive Neuro-Fuzzy Inference System along with Reinforcement Learning (ANFIS-RL). The method starts with the deployment stage, where the sensor nodes (SNs) are given the deployment parameters, including position, communication range, packet transmission rate, energy, speed and temperature. The parameters are used to form fuzzy rules, which are then applied in a grid-based partitioning approach to the identification of faults. The ANFIS-RL system can identify the fault nodes effectively, which can improve the performance of the system. The proposed model is dynamically adaptable to different network conditions, and it is robust to different failure scenarios. The approach is contrasted to some of the existing methods, and its accuracy in fault detection and reliability of the network is analyzed.</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 Accurate Fault Node Identification Using Hybrid Adaptive Neuro‑Fuzzy Inference System-Based Reinforcement Learning in Wireless Sensor Network

  • 期刊论文

  • 发表日期：

    2026年7月4日

查看全文
  • Sujay Chakraborty,
  • Ajay Singh Raghuvanshi

摘要

A wireless sensor network (WSN) can cause disruption in the communication if there are any malfunctioning nodes. Identification of malfunction nodes is a difficult task because of the variety of operating conditions and the limited resources of sensor nodes. The defect detection mechanism described in the paper is proposed to be implemented by using an Adaptive Neuro-Fuzzy Inference System along with Reinforcement Learning (ANFIS-RL). The method starts with the deployment stage, where the sensor nodes (SNs) are given the deployment parameters, including position, communication range, packet transmission rate, energy, speed and temperature. The parameters are used to form fuzzy rules, which are then applied in a grid-based partitioning approach to the identification of faults. The ANFIS-RL system can identify the fault nodes effectively, which can improve the performance of the system. The proposed model is dynamically adaptable to different network conditions, and it is robust to different failure scenarios. The approach is contrasted to some of the existing methods, and its accuracy in fault detection and reliability of the network is analyzed.