Scholarlink

<p>The increasing complexity and dynamic behavior of Internet of Things (IoT) networks, characterized by frequent topology changes, node mobility, and resource constraints, necessitates efficient and secure routing protocols. This paper presents Adaptive Neuro-Fuzzy Trust-Based Routing (ANF-TBR), an innovative routing framework specifically designed to address these challenges. ANF-TBR employs an Adaptive Neuro-Fuzzy Inference System (ANFIS) to dynamically evaluate node trustworthiness, significantly enhancing security, reliability, and adaptability within dynamic IoT environments. Its lightweight computational design ensures minimal CPU and memory usage, making it highly suitable for deployment on resource-constrained IoT devices. By integrating adaptive trust evaluation with optimized routing decisions, ANF-TBR effectively improves network performance, achieving a 10% higher packet delivery ratio (PDR), a 15% reduction in end-to-end delay, and 30% lower CPU and memory usage compared to existing trust-based routing protocols. Additionally, the proposed model introduces a dynamic hysteresis mechanism to minimize unnecessary parent node changes, significantly enhancing network stability. Extensive simulations conducted using the ROUT-4-2023 benchmark dataset and a customized dataset generated through the Cooja simulator demonstrate ANF-TBR’s superior please capability in identifying and mitigating routing attacks such as Blackhole, Rank, and Sinkhole attacks. These results highlight ANF-TBR as a practical and reliable solution for a wide range of IoT 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

A new adaptive neuro-fuzzy trust-based routing for dynamic IoT networks

  • 期刊论文

  • 发表日期：

    2025年9月11日

查看全文
  • Zohre Shoaei,
  • Rasool Esmaeilyfard,
  • Reza Javidan,
  • Ahmad Jalili

摘要

The increasing complexity and dynamic behavior of Internet of Things (IoT) networks, characterized by frequent topology changes, node mobility, and resource constraints, necessitates efficient and secure routing protocols. This paper presents Adaptive Neuro-Fuzzy Trust-Based Routing (ANF-TBR), an innovative routing framework specifically designed to address these challenges. ANF-TBR employs an Adaptive Neuro-Fuzzy Inference System (ANFIS) to dynamically evaluate node trustworthiness, significantly enhancing security, reliability, and adaptability within dynamic IoT environments. Its lightweight computational design ensures minimal CPU and memory usage, making it highly suitable for deployment on resource-constrained IoT devices. By integrating adaptive trust evaluation with optimized routing decisions, ANF-TBR effectively improves network performance, achieving a 10% higher packet delivery ratio (PDR), a 15% reduction in end-to-end delay, and 30% lower CPU and memory usage compared to existing trust-based routing protocols. Additionally, the proposed model introduces a dynamic hysteresis mechanism to minimize unnecessary parent node changes, significantly enhancing network stability. Extensive simulations conducted using the ROUT-4-2023 benchmark dataset and a customized dataset generated through the Cooja simulator demonstrate ANF-TBR’s superior please capability in identifying and mitigating routing attacks such as Blackhole, Rank, and Sinkhole attacks. These results highlight ANF-TBR as a practical and reliable solution for a wide range of IoT applications.