Scholarlink

<p>Wireless sensor networks (WSN) are an economical solution for sensing region or coverage radius information updates. To extend WSN performance, an energy-efficient strategy is needed. Clustering-based routing is the best method for extending WSN performance parameters, supporting fault tolerance and reliable connectivity. However, achieving these metrics results in a shorter lifespan for the cluster head (CH). To overcome these issues, the DBSCAN based clustering approach with the DQN model has been designed for effective CH selection with an optimized routing path. The paper presents an optimization algorithm for efficient load balancing and energy conservation using DBDQN and wild geese migration optimization. Initially, nodes are placed in specific sensing regions. Method for clustering deployed nodes in massive quantities of data is density-based spatial clustering of applications with noise, or DBSCAN. The cluster head is identified using a deep Q-network-based reinforcement learning method. Based on residual energy and Euclidean distance, the Wild Geese Migration Optimized Routing Protocol for WSNs (WGMO) optimizes the selection of forwarder nodes. An Energy Level Matrix and a route recognizing algorithm is used to improve routing choices. Performance of DBDQN-WGMO is contrasted with that of additional resource-grabbing routing techniques on significant metrics such as 96.5% packet delivery ratio, 3.5% packet loss ratio, 5.33 packets per second (pps) throughput, 9.14&#xa0;s average delay, 3.13&#xa0;s jitter, 5.97&#xa0;s latency, 2.98&#xa0;J energy usage, 49.31&#xa0;h network lifespan, and 0.29&#xa0;s propagation delay. Therefore, the proposed routing procedure can significantly improve data transfer between users in wireless sensor networks.</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

Efficient load balancing and energy conservation protocol for WSN based on DBDQN and wild geeze migration optimization algorithm

  • 期刊论文

  • 发表日期：

    2026年2月3日

查看全文
  • G. Vinoth Rajkumar,
  • N. Muthukumaran

摘要

Wireless sensor networks (WSN) are an economical solution for sensing region or coverage radius information updates. To extend WSN performance, an energy-efficient strategy is needed. Clustering-based routing is the best method for extending WSN performance parameters, supporting fault tolerance and reliable connectivity. However, achieving these metrics results in a shorter lifespan for the cluster head (CH). To overcome these issues, the DBSCAN based clustering approach with the DQN model has been designed for effective CH selection with an optimized routing path. The paper presents an optimization algorithm for efficient load balancing and energy conservation using DBDQN and wild geese migration optimization. Initially, nodes are placed in specific sensing regions. Method for clustering deployed nodes in massive quantities of data is density-based spatial clustering of applications with noise, or DBSCAN. The cluster head is identified using a deep Q-network-based reinforcement learning method. Based on residual energy and Euclidean distance, the Wild Geese Migration Optimized Routing Protocol for WSNs (WGMO) optimizes the selection of forwarder nodes. An Energy Level Matrix and a route recognizing algorithm is used to improve routing choices. Performance of DBDQN-WGMO is contrasted with that of additional resource-grabbing routing techniques on significant metrics such as 96.5% packet delivery ratio, 3.5% packet loss ratio, 5.33 packets per second (pps) throughput, 9.14 s average delay, 3.13 s jitter, 5.97 s latency, 2.98 J energy usage, 49.31 h network lifespan, and 0.29 s propagation delay. Therefore, the proposed routing procedure can significantly improve data transfer between users in wireless sensor networks.