Scholarlink

<p>Mobility is crucial in the industrial internet of things (IIoT), while the widely used IPv6 routing protocol for low-power and lossy networks (RPL) faces performance issues in the presence of mobile nodes (MNs). This paper presents methods to enhance mobility support and reliability in IIoT-based mobile networks while minimizing network overhead, energy consumption, and latency. Using a new cross-layer metric, sensor nodes select their next hop based on an objective function (OF) that considers node mobility. Additionally, a reinforcement learning (RL) approach allows nodes to predict their operational state -whether they are undergoing hand-offs, retaining outdated routes, or in a stable state- facilitating timely updates and reducing failures. Our protocol demonstrates lower computational complexity and superior performance compared to RPL and existing approaches under various mobility speeds and node densities. These results highlight the effectiveness and lightweight design of our proposed protocol for IIoT applications.</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

RIMS-RPL: RL-based intelligent mobility-support of RPL for IIoT-based networks

  • 期刊论文

  • 发表日期：

    2025年9月2日

查看全文
  • Ismahene Alem,
  • Samira Yessad,
  • Louiza Bouallouche-Medjkoune

摘要

Mobility is crucial in the industrial internet of things (IIoT), while the widely used IPv6 routing protocol for low-power and lossy networks (RPL) faces performance issues in the presence of mobile nodes (MNs). This paper presents methods to enhance mobility support and reliability in IIoT-based mobile networks while minimizing network overhead, energy consumption, and latency. Using a new cross-layer metric, sensor nodes select their next hop based on an objective function (OF) that considers node mobility. Additionally, a reinforcement learning (RL) approach allows nodes to predict their operational state -whether they are undergoing hand-offs, retaining outdated routes, or in a stable state- facilitating timely updates and reducing failures. Our protocol demonstrates lower computational complexity and superior performance compared to RPL and existing approaches under various mobility speeds and node densities. These results highlight the effectiveness and lightweight design of our proposed protocol for IIoT applications.