Scholarlink

In various applications, Wireless Sensor Networks (WSN) plays a crucial role. Energy being one of the main factors in the sensor nodes, many challenges are faced during the transmission of data. A hybrid method is used, introducing Adaptive Neuro Fuzzy Inference System (ANFIS) and Deep Reinforcement Learning (DRL) to optimize energy efficiency in WSNs. The existing work shows traditional methods used for static routing and power management schemes that may not adapt well to changes in environmental conditions. This paper aims to use ANFIS model to predict the Cluster Head (CH) and Secondary Cluster Head (SCH) based on the state of environment. Along with it, Deep Q-Network (DQN) component that is responsible to adapt to dynamic network conditions by making informed decisions to maximize data transmission efficiency. Results indicate that the hybrid model significantly reduces energy consumption while enhancing data relay capabilities. This work underscores the potential of integrating ANFIS with DRL for advancing energy management strategies in WSNs.

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

ANF-DRL: Adaptive Neuro Fuzzy and Deep Reinforcement Learning for Optimized Energy Efficiency in Wireless Sensor Networks

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • A. P. Jyothi,
  • Shaista Tarannum,
  • Shreyanka Subbarayappa,
  • N. Bhavatarini,
  • S. R. Bhagyashree

摘要

In various applications, Wireless Sensor Networks (WSN) plays a crucial role. Energy being one of the main factors in the sensor nodes, many challenges are faced during the transmission of data. A hybrid method is used, introducing Adaptive Neuro Fuzzy Inference System (ANFIS) and Deep Reinforcement Learning (DRL) to optimize energy efficiency in WSNs. The existing work shows traditional methods used for static routing and power management schemes that may not adapt well to changes in environmental conditions. This paper aims to use ANFIS model to predict the Cluster Head (CH) and Secondary Cluster Head (SCH) based on the state of environment. Along with it, Deep Q-Network (DQN) component that is responsible to adapt to dynamic network conditions by making informed decisions to maximize data transmission efficiency. Results indicate that the hybrid model significantly reduces energy consumption while enhancing data relay capabilities. This work underscores the potential of integrating ANFIS with DRL for advancing energy management strategies in WSNs.