Scholarlink

<p>The integration of machine learning approaches in the management of cellular communication systems yields substantial improvements in service quality and performance. This paper presents a comprehensive approach to address the challenges of traffic slicind and predictive analysis of delay and packet loss rate (PLR) in cellular networks. The proposed framework emphasizes the importance of fine-grained traffic slicing to enable accurate predictions of delay and PLR, thus enhancing resource allocation efficiency and quality of service (QoS) standards. The study formulates the problem by highlighting the complexities associated with traffic classification and the need for precise predictive analysis to ensure QoS adherence. Leveraging machine learning models, the framework categorizes traffic based on diverse parameters and key performance indicators (KPIs) such as delay and PLR. Additionally, the paper introduces an adaptive scheduling strategy that combines instantaneous and reservation-based scheduling to optimize resource allocation and maintain QoS standards. The adaptive approach dynamically transitions between scheduling methods based on network conditions and traffic demands, ensuring low latency for time-sensitive applications while guaranteeing QoS for others. Furthermore, the proposed architecture integrates SDN and NFV into a hybrid terrestrial-satellite network, enhancing resource management, network coverage, and connectivity. The architecture features double-level traffic slicing, effectively offloading traffic to satellite components to alleviate congestion and extend coverage to underserved areas. The paper also emphasizes the importance of SDN controllers in orchestrating the 5G network, highlighting their role in managing network resources efficiently. The implementation of this scheduling system improves network performance by lowering downtime and increasing user satisfaction. This approach, in particular, benefits high-performance networks by lowering maintenance costs while increasing revenue. The proposed framework effectively reduces operational disruptions, improves customer satisfaction, and generates financial benefits for both commercial and critical infrastructure networks. This study lays a solid foundation for future network optimization initiatives. The study extends its application to satellite communication systems, acknowledging their unique role in the modern communication landscape. The proposed methodology, while emphasizing terrestrial networks, showcases adaptability for satellite networks, contributing to improved reliability and efficiency.</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

Improved SDN-hybrid satellite-terrestrial architecture integrating intelligent fine-grained traffic slicing

  • 期刊论文

  • 发表日期：

    2025年4月28日

查看全文
  • Saloua Hendaoui,
  • Nawel Zangar

摘要

The integration of machine learning approaches in the management of cellular communication systems yields substantial improvements in service quality and performance. This paper presents a comprehensive approach to address the challenges of traffic slicind and predictive analysis of delay and packet loss rate (PLR) in cellular networks. The proposed framework emphasizes the importance of fine-grained traffic slicing to enable accurate predictions of delay and PLR, thus enhancing resource allocation efficiency and quality of service (QoS) standards. The study formulates the problem by highlighting the complexities associated with traffic classification and the need for precise predictive analysis to ensure QoS adherence. Leveraging machine learning models, the framework categorizes traffic based on diverse parameters and key performance indicators (KPIs) such as delay and PLR. Additionally, the paper introduces an adaptive scheduling strategy that combines instantaneous and reservation-based scheduling to optimize resource allocation and maintain QoS standards. The adaptive approach dynamically transitions between scheduling methods based on network conditions and traffic demands, ensuring low latency for time-sensitive applications while guaranteeing QoS for others. Furthermore, the proposed architecture integrates SDN and NFV into a hybrid terrestrial-satellite network, enhancing resource management, network coverage, and connectivity. The architecture features double-level traffic slicing, effectively offloading traffic to satellite components to alleviate congestion and extend coverage to underserved areas. The paper also emphasizes the importance of SDN controllers in orchestrating the 5G network, highlighting their role in managing network resources efficiently. The implementation of this scheduling system improves network performance by lowering downtime and increasing user satisfaction. This approach, in particular, benefits high-performance networks by lowering maintenance costs while increasing revenue. The proposed framework effectively reduces operational disruptions, improves customer satisfaction, and generates financial benefits for both commercial and critical infrastructure networks. This study lays a solid foundation for future network optimization initiatives. The study extends its application to satellite communication systems, acknowledging their unique role in the modern communication landscape. The proposed methodology, while emphasizing terrestrial networks, showcases adaptability for satellite networks, contributing to improved reliability and efficiency.