Scholarlink

<p>The Internet of Things (IoT) has transformed data acquisition and decision-making across sectors, yet the limited energy of sensor nodes poses challenges for network longevity and efficiency. This paper proposes a multi-stage, multi-objective routing protocol using a hybrid of Ladybug Optimization (LBO), Butterfly Optimization Algorithm (BOA), and Q-learning. Virtual cluster heads are initially selected based on centrality and load balancing, followed by predictive energy-aware clustering to extend network life. Q-learning then enables dynamic, energy-efficient multi-hop routing based on energy levels and proximity. Simulation results show the method reduces energy consumption by up to 43% compared to FIAVOA in specific scenarios and extends network lifetime by up to 47% over GA-SDN. It also increases the number of alive nodes by up to 76% and delays the first node death time by up to 60%, enhancing network stability and coverage. These results underscore the approach’s effectiveness in sustaining IoT networks while ensuring efficient data transmission.</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

An AI-driven multi-stage routing protocol for energy-efficient IoT networks

  • 期刊论文

  • 发表日期：

    2025年8月31日

查看全文
  • Amin Nazari,
  • Seyedeh Shabnam Jazaeri,
  • Abolfazl Omidi,
  • Muharram Mansoorizadeh

摘要

The Internet of Things (IoT) has transformed data acquisition and decision-making across sectors, yet the limited energy of sensor nodes poses challenges for network longevity and efficiency. This paper proposes a multi-stage, multi-objective routing protocol using a hybrid of Ladybug Optimization (LBO), Butterfly Optimization Algorithm (BOA), and Q-learning. Virtual cluster heads are initially selected based on centrality and load balancing, followed by predictive energy-aware clustering to extend network life. Q-learning then enables dynamic, energy-efficient multi-hop routing based on energy levels and proximity. Simulation results show the method reduces energy consumption by up to 43% compared to FIAVOA in specific scenarios and extends network lifetime by up to 47% over GA-SDN. It also increases the number of alive nodes by up to 76% and delays the first node death time by up to 60%, enhancing network stability and coverage. These results underscore the approach’s effectiveness in sustaining IoT networks while ensuring efficient data transmission.