Scholarlink

<p>The unpredictable environmental conditions create significant challenges for data transmission in underwater acoustic communication systems. Stochastic Network Calculus (SNC) provides a strong framework for analyzing and optimizing these networks, but traditional models struggle with predictive accuracy in rapidly changing environments. This paper presents an innovative approach by integrating deep learning models with SNC to enhance channel state prediction and improve network performance. The Deep Learning model used to predict future channel states and allowing proactive adjustments to the SNC model. The results show a 10% reduction in expected delay and an 18% reduction in backlog compared to traditional models. The method also enhances throughput and transmission efficiency by adjusting network parameters based on predictions. This integration fosters resilience to dynamic underwater environments affected by temperature, salinity changes and emphasizing the importance of optimizing underwater acoustic communication for marine exploration for environmental monitoring 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

Integrating Deep Learning Models into Stochastic Network Calculus for Enhanced Channel State Prediction in Underwater Acoustic Networks

  • 期刊论文

  • 发表日期：

    2025年2月1日

查看全文
  • M. Saravanan,
  • Rajeev Sukumaran

摘要

The unpredictable environmental conditions create significant challenges for data transmission in underwater acoustic communication systems. Stochastic Network Calculus (SNC) provides a strong framework for analyzing and optimizing these networks, but traditional models struggle with predictive accuracy in rapidly changing environments. This paper presents an innovative approach by integrating deep learning models with SNC to enhance channel state prediction and improve network performance. The Deep Learning model used to predict future channel states and allowing proactive adjustments to the SNC model. The results show a 10% reduction in expected delay and an 18% reduction in backlog compared to traditional models. The method also enhances throughput and transmission efficiency by adjusting network parameters based on predictions. This integration fosters resilience to dynamic underwater environments affected by temperature, salinity changes and emphasizing the importance of optimizing underwater acoustic communication for marine exploration for environmental monitoring applications.