Scholarlink

<p>In industries like agriculture, healthcare, industry, and large-area surveillance, wireless sensor networks, or WSNs, are essential for collecting and analyzing environmental data. However, the lifespan and overall efficacy of these networks are restricted by the sensor nodes’ limited battery capacity. This study presents an energy-efficient routing approach that combines Deep Reinforcement Learning + Fuzzy Logic (DRLF) to improve performance, particularly in dense Internet of Things (IoT) scenarios. While a fuzzy-logic layer chooses clusters and distributes energy across nodes, DRL predicts the path of a mobile sink and constantly improves routing methods. The resulting synergy increases throughput, reduces packet-delivery latency, and significantly reduces power consumption. The method swiftly recognizes changing network states and modifies its settings to maintain stability and scalability because of its self-adjusting nature. In comparison to conventional methods, simulations show a 62% increase in network lifetime and a 48% decrease in end-to-end latency, which are crucial for real-time applications like ecological monitoring and smart health systems. Overall, the results confirm that combining fuzzy inference and DRL significantly improves WSN performance in dynamic, resource-constrained environments.</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

Energy-Aware Adaptive Routing for Mobile Sink Nodes in Wireless Sensor Networks Leveraging Deep Reinforcement Learning and Fuzzy-Based Clustering

  • 期刊论文

  • 发表日期：

    2026年7月30日

查看全文
  • Jiangbo Tang

摘要

In industries like agriculture, healthcare, industry, and large-area surveillance, wireless sensor networks, or WSNs, are essential for collecting and analyzing environmental data. However, the lifespan and overall efficacy of these networks are restricted by the sensor nodes’ limited battery capacity. This study presents an energy-efficient routing approach that combines Deep Reinforcement Learning + Fuzzy Logic (DRLF) to improve performance, particularly in dense Internet of Things (IoT) scenarios. While a fuzzy-logic layer chooses clusters and distributes energy across nodes, DRL predicts the path of a mobile sink and constantly improves routing methods. The resulting synergy increases throughput, reduces packet-delivery latency, and significantly reduces power consumption. The method swiftly recognizes changing network states and modifies its settings to maintain stability and scalability because of its self-adjusting nature. In comparison to conventional methods, simulations show a 62% increase in network lifetime and a 48% decrease in end-to-end latency, which are crucial for real-time applications like ecological monitoring and smart health systems. Overall, the results confirm that combining fuzzy inference and DRL significantly improves WSN performance in dynamic, resource-constrained environments.