Scholarlink

<p>The Internet of Things (IoT) networks requires an efficient Medium Access Control (MAC) protocol with high throughput, low latency, and energy efficiency. Existing carrier sensing based medium access protocols and its derivatives are often fail to meet these requirements. This paper introduces Enhanced Q-Learning enabled Distributed Queuing MAC (EQL-DQMAC) protocol for IoT networks. EQL-DQMAC combines the benefits of Q-learning, distributed queuing, joint contention tree structure, and concurrent multichannel data transfer to improve network performance efficiently. EQL-DQMAC adaptively modifies the number of contending nodes within the contention period using Q-Learning with an enhanced reward function and performs distributed queuing mechanism based on joint contention tree structure for efficient contention resolution and data transfer. Extensive simulations demonstrate the superiority of EQL-DQMAC over traditional DQMAC and Q-Learning based QL-DQMAC approaches in terms of throughput, delay and energy consumption. The algorithm strives for balanced optimization of throughput, delay, and energy efficiency, addressing challenges in prevailing MAC protocols for 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

An enhanced Q-learning enabled distributed queuing MAC (EQL-DQMAC) protocol for IoT networks

  • 期刊论文

  • 发表日期：

    2026年7月2日

查看全文
  • Atul Kumar Pandey,
  • Abhishek Kumar

摘要

The Internet of Things (IoT) networks requires an efficient Medium Access Control (MAC) protocol with high throughput, low latency, and energy efficiency. Existing carrier sensing based medium access protocols and its derivatives are often fail to meet these requirements. This paper introduces Enhanced Q-Learning enabled Distributed Queuing MAC (EQL-DQMAC) protocol for IoT networks. EQL-DQMAC combines the benefits of Q-learning, distributed queuing, joint contention tree structure, and concurrent multichannel data transfer to improve network performance efficiently. EQL-DQMAC adaptively modifies the number of contending nodes within the contention period using Q-Learning with an enhanced reward function and performs distributed queuing mechanism based on joint contention tree structure for efficient contention resolution and data transfer. Extensive simulations demonstrate the superiority of EQL-DQMAC over traditional DQMAC and Q-Learning based QL-DQMAC approaches in terms of throughput, delay and energy consumption. The algorithm strives for balanced optimization of throughput, delay, and energy efficiency, addressing challenges in prevailing MAC protocols for IoT applications.