Scholarlink

<p>Many investigations have been conducted to demonstrate the impact of new technologies such as 5G on serving a variety of IoT applications, which have led to new architectural designs such as the Open Radio Access Network (O-RAN) and Multi-Access Edge Computing (MEC). An increase in the number of connected devices imposes more challenges in the infrastructure layer of the operators’ networks to cover new performance criteria such as extensive connectivity, security, ultra-low latency, and reliability. As O-RAN allows different RAN layers to be split and established as virtual functions, the RAN Intelligent Controller (RIC) will be responsible for controlling and optimizing these virtual functions. On the other hand, although some efforts have been made to demonstrate MEC server deployment in the O-RAN architecture as an effective solution to reduce heavy traffic loads and end-to-end delays, Quality of Service (QoS) is underestimated due to the complexity of deployment in the RIC. It is limited to non-prioritized solutions, which are not feasible in real networks. To ensure adequate QoS levels, this paper aims to examine the placement of MEC servers and user allocation, emphasizing latency and cost in an O-RAN architecture. For this purpose, the problem is formulated as a mixed-integer programming model based on a priority-based queuing approach. Due to the NP-hard nature of the proposed mathematical model, a Lagrangian relaxation algorithm and a matheuristic algorithm by hybridization of Genetic Algorithm and Lagrangian relaxation algorithm are developed to tackle large-scale instances. The effectiveness of the proposed model and solution methods has been validated via sensitivity analysis. The results show that the proposed model is efficient in reducing the number of lost latency-sensitive requests.</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

QoS-aware placement of MEC servers in an O-RAN architecture

  • 期刊论文

  • 发表日期：

    2026年2月12日

查看全文
  • Nasim Kazemifard,
  • Donya Rahmani

摘要

Many investigations have been conducted to demonstrate the impact of new technologies such as 5G on serving a variety of IoT applications, which have led to new architectural designs such as the Open Radio Access Network (O-RAN) and Multi-Access Edge Computing (MEC). An increase in the number of connected devices imposes more challenges in the infrastructure layer of the operators’ networks to cover new performance criteria such as extensive connectivity, security, ultra-low latency, and reliability. As O-RAN allows different RAN layers to be split and established as virtual functions, the RAN Intelligent Controller (RIC) will be responsible for controlling and optimizing these virtual functions. On the other hand, although some efforts have been made to demonstrate MEC server deployment in the O-RAN architecture as an effective solution to reduce heavy traffic loads and end-to-end delays, Quality of Service (QoS) is underestimated due to the complexity of deployment in the RIC. It is limited to non-prioritized solutions, which are not feasible in real networks. To ensure adequate QoS levels, this paper aims to examine the placement of MEC servers and user allocation, emphasizing latency and cost in an O-RAN architecture. For this purpose, the problem is formulated as a mixed-integer programming model based on a priority-based queuing approach. Due to the NP-hard nature of the proposed mathematical model, a Lagrangian relaxation algorithm and a matheuristic algorithm by hybridization of Genetic Algorithm and Lagrangian relaxation algorithm are developed to tackle large-scale instances. The effectiveness of the proposed model and solution methods has been validated via sensitivity analysis. The results show that the proposed model is efficient in reducing the number of lost latency-sensitive requests.