Scholarlink

In low-power networks, sensors are typically deployed randomly or in a predetermined pattern to monitor an area for various applications such as environmental monitoring, surveillance, and disaster management. Efficient coverage strategies can significantly affect the energy consumption and operational lifetime of WSNs. By optimizing sensor deployment and coverage patterns, WSNs can minimize redundant sensing, communication overhead, and energy waste. This results in longer network lifetimes, reduced maintenance costs, and improved stability, especially in applications deployed in remote or harsh environments. In this paper, we aim to create a trade-off between balancing mobile sensors in IoT networks and achieving environmental coverage by presenting an approach based on the Q-learning reinforcement algorithm (VF-RL). This approach allows us to manage the mobility of the nodes in order to increase energy consumption and improve network coverage. For this purpose, the virtual force approach is used to calculate the distance of neighbors and their orientation. Simulation results with different scenarios show that the proposed approach performs well compared to VFA  [18], VFPSO  [16], ALO  [1] and VF-IALO  [17] algorithms in three different scenarios.

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

VF-RL: A Reinforcement Learning-Based Coverage Improvement in Mobile IoT Networks Using Virtual Force

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Raheleh Samadi,
  • Amin Nazari,
  • Jochen Seitz

摘要

In low-power networks, sensors are typically deployed randomly or in a predetermined pattern to monitor an area for various applications such as environmental monitoring, surveillance, and disaster management. Efficient coverage strategies can significantly affect the energy consumption and operational lifetime of WSNs. By optimizing sensor deployment and coverage patterns, WSNs can minimize redundant sensing, communication overhead, and energy waste. This results in longer network lifetimes, reduced maintenance costs, and improved stability, especially in applications deployed in remote or harsh environments. In this paper, we aim to create a trade-off between balancing mobile sensors in IoT networks and achieving environmental coverage by presenting an approach based on the Q-learning reinforcement algorithm (VF-RL). This approach allows us to manage the mobility of the nodes in order to increase energy consumption and improve network coverage. For this purpose, the virtual force approach is used to calculate the distance of neighbors and their orientation. Simulation results with different scenarios show that the proposed approach performs well compared to VFA  [18], VFPSO  [16], ALO  [1] and VF-IALO  [17] algorithms in three different scenarios.