Scholarlink

<p>In today’s era of cutting-edge technology, Wireless Sensor Networks (WSNs) play a crucial role in powering the Internet of Things (IoT), the Industrial Internet of Things (IIoT), and the Internet of Healthcare Things (IoHT). Various state-of-the-art methods have been deployed to improve energy efficiency and data dissemination techniques in WSNs. In event-driven applications, the data generation rate may vary across different segments of the deployed network. Situations may arise where the Mobile Sink (MS) is unavailable at clusters with high data generation rates or remains stationed at clusters with minimal data generation, leading to inefficiencies. To address this issue, this article introduces an intelligent scheduling technique for an MS in a clustered WSN. By leveraging a Reinforcement Learning approach, the MS learns to optimize its path by skipping nodes with no new data in subsequent episodes. This significantly reduces data collection time, enhancing network efficiency and latency. The proposed work computes the variation in the data generation and accordingly manages the path of MS in dynamic network behaviour. The simulation results highlight the effectiveness of the proposed work in comparison to existing WSN models. The results are motivating for further use in healthcare applications. The simulation results clearly illustrate the superior efficiency compared to existing protocols, including LEACH, JBMDBM, and QWRP, with respect to parameters like node death, average packet loss percentage, and network lifetime.</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

Q-learning framework for data prioritization and route optimization for efficient data collection in IoHT applications

  • 期刊论文

  • 发表日期：

    2026年3月31日

查看全文
  • Rakhi Puri,
  • Teek Parval Sharma

摘要

In today’s era of cutting-edge technology, Wireless Sensor Networks (WSNs) play a crucial role in powering the Internet of Things (IoT), the Industrial Internet of Things (IIoT), and the Internet of Healthcare Things (IoHT). Various state-of-the-art methods have been deployed to improve energy efficiency and data dissemination techniques in WSNs. In event-driven applications, the data generation rate may vary across different segments of the deployed network. Situations may arise where the Mobile Sink (MS) is unavailable at clusters with high data generation rates or remains stationed at clusters with minimal data generation, leading to inefficiencies. To address this issue, this article introduces an intelligent scheduling technique for an MS in a clustered WSN. By leveraging a Reinforcement Learning approach, the MS learns to optimize its path by skipping nodes with no new data in subsequent episodes. This significantly reduces data collection time, enhancing network efficiency and latency. The proposed work computes the variation in the data generation and accordingly manages the path of MS in dynamic network behaviour. The simulation results highlight the effectiveness of the proposed work in comparison to existing WSN models. The results are motivating for further use in healthcare applications. The simulation results clearly illustrate the superior efficiency compared to existing protocols, including LEACH, JBMDBM, and QWRP, with respect to parameters like node death, average packet loss percentage, and network lifetime.