Scholarlink

<p>Underwater wireless sensor networks (UWSN) require a suitable routing method due to slow signal travel time, numerous data errors, low energy in sensor nodes, and limited data transmission capabilities. Therefore, designing an efficient UWSN routing mechanism is essential for the performance improvement. Correct localization of the next forwarding relay is vital to avoid data loss, void node communication, and longer propagation delays. Traditional range-free localization approaches suffered from poor quality of service (QoS) and localization errors. The recent methods of UWSN localization techniques further need enhancement in terms of routing efficiency and complexity. We propose QoS-aware range-free localization using the hybrid Q-learning (QRL-QL) protocol to address the problems of existing methods. The novelty of the QRL-QL approach is that hybrid QoS parameters are incorporated to perform the range-free localization of underwater sensor nodes using the QL technique. To achieve optimal sensor localization for route construction, the Q-value is computed by concurrently evaluating five characteristics of sensor nodes throughout the UWSN routing cycle. To improve the overall performance of UWSNs, we specifically develop a hybrid reward function in QL localization parameters. This hybrid reward function in the QL approach incorporated the localization and QoS parameters, including propagation delay, transmission loss, localization error, energy level, and data transmission direction. The simulation results indicate that the proposed QRL-QL protocol demonstrates strong performance regarding throughput, energy efficiency, localization errors, and communication delay within the dynamic underwater network environment.</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

Hybrid Q-Learning Technique for Range Free Localization Protocol in Underwater Wireless Sensor Networks

  • 期刊论文

  • 发表日期：

    2026年7月29日

查看全文
  • Rupinder kaur,
  • Sonia Goyal

摘要

Underwater wireless sensor networks (UWSN) require a suitable routing method due to slow signal travel time, numerous data errors, low energy in sensor nodes, and limited data transmission capabilities. Therefore, designing an efficient UWSN routing mechanism is essential for the performance improvement. Correct localization of the next forwarding relay is vital to avoid data loss, void node communication, and longer propagation delays. Traditional range-free localization approaches suffered from poor quality of service (QoS) and localization errors. The recent methods of UWSN localization techniques further need enhancement in terms of routing efficiency and complexity. We propose QoS-aware range-free localization using the hybrid Q-learning (QRL-QL) protocol to address the problems of existing methods. The novelty of the QRL-QL approach is that hybrid QoS parameters are incorporated to perform the range-free localization of underwater sensor nodes using the QL technique. To achieve optimal sensor localization for route construction, the Q-value is computed by concurrently evaluating five characteristics of sensor nodes throughout the UWSN routing cycle. To improve the overall performance of UWSNs, we specifically develop a hybrid reward function in QL localization parameters. This hybrid reward function in the QL approach incorporated the localization and QoS parameters, including propagation delay, transmission loss, localization error, energy level, and data transmission direction. The simulation results indicate that the proposed QRL-QL protocol demonstrates strong performance regarding throughput, energy efficiency, localization errors, and communication delay within the dynamic underwater network environment.