Scholarlink

<p>Wireless Sensor Networks (WSNs) have emerged as a focal point of research and practical application in the field of computer networks and telecommunications. One of the foremost challenges in WSNs is the development of precise and efficient localization methods. In this research, we propose an enhanced DV-Hop localization algorithm. First, the neural network approach is applied to enhance localization accuracy by minimizing distance errors. The work is extended to cover energy efficiency by using optimization techniques for WSN-enabled DV-Hop. To solve the non-convex optimization for the WSN-enabled DV-Hop scheme, Alternating Direction Method of Multipliers (ADMM) is employed, which decomposes the global optimization task into three coordinated sub-problems—local variable update, global variable consensus, and dual variable update, allowing for efficient distributed computation. Results show that there is a 25.9% gain in accuracy when doubling the total number of nodes at certain anchor nodes. On the other hand, there is a 42.7% gain in energy efficiency when using double anchor nodes. Also, the results showed that the average localization error for the proposed method was 3.87% compared with 8.57% for DV-Hop. Finally, results show a drop in the energy consumption from 28.1% compared to DV-Hop to 11.59% for our proposed method.</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

Leveraging neural networks for energy-efficient DV-Hop localization using RSSI in wireless sensor networks

  • 期刊论文

  • 发表日期：

    2026年7月23日

查看全文
  • Abdelrahman Almomani,
  • Fadi Al-Turjman

摘要

Wireless Sensor Networks (WSNs) have emerged as a focal point of research and practical application in the field of computer networks and telecommunications. One of the foremost challenges in WSNs is the development of precise and efficient localization methods. In this research, we propose an enhanced DV-Hop localization algorithm. First, the neural network approach is applied to enhance localization accuracy by minimizing distance errors. The work is extended to cover energy efficiency by using optimization techniques for WSN-enabled DV-Hop. To solve the non-convex optimization for the WSN-enabled DV-Hop scheme, Alternating Direction Method of Multipliers (ADMM) is employed, which decomposes the global optimization task into three coordinated sub-problems—local variable update, global variable consensus, and dual variable update, allowing for efficient distributed computation. Results show that there is a 25.9% gain in accuracy when doubling the total number of nodes at certain anchor nodes. On the other hand, there is a 42.7% gain in energy efficiency when using double anchor nodes. Also, the results showed that the average localization error for the proposed method was 3.87% compared with 8.57% for DV-Hop. Finally, results show a drop in the energy consumption from 28.1% compared to DV-Hop to 11.59% for our proposed method.