摘要
为应对开放型无线接入网(Open Radio Access Network,O-RAN)中的数据传输成本过高及网络兼容不足等问题,研究了面向O-RAN的多级边缘服务资源分配与部署联合优化问题。首先,利用四层融合模型将多目标联合优化问题转化为异构边缘服务器数量选择及位置确定问题,并提出了一种负载约束和迭代优化的异构边缘服务器资源分配算法,解决了O-RAN网络中的异构资源分配与数据传输问题。然后,提出了一种能效驱动的异构节点部署优化算法,解决了多级异构资源最佳部署位置问题。最后,利用上海电信基站的真实数据集,验证了所提资源优化与部署算法的有效性,实验结果表明,所提算法较其它算法在部署成本上至少降低了22.5%,能效比值上至少提高了25.96%。
To address issues such as high data transmission costs and insufficient network compatibility in the Open Radio Access Network(O-RAN),research has been conducted on the joint optimization of multi-tier edge service resource allocation and deployment targeted at O-RAN.Initially,a four-layer fusion model is employed to transform the multi-objective joint optimization problem into an issue of selecting the number and determining the locations of heterogeneous edge servers.Subsequently,an algorithm for optimizing heterogeneous edge server resources,incorporating load constraints and iterative optimization,is proposed,effectively resolving the challenges of heterogeneous resource allocation and data transmission within the O-RAN network.Furthermore,an energy-efficient driven algorithm for optimizing the deployment of heterogeneous nodes is introduced to address the problem of determining the optimal deployment locations for multi-tiered heterogeneous resources.Finally,the effectiveness of the proposed resource optimization and deployment algorithms is validated using a real dataset from Shanghai Telecom base stations.The experimental results demonstrate that the proposed algorithm,compared to alternative methods,reduces deployment costs by at least 22.5%and increases energy efficiency by a minimum of 25.96%.
作者
赵晓焱
薛永辉
袁培燕
ZHAO Xiao-yan;XUE Yong-hui;YUAN Pei-yan(College of Computer and Information Engineering,Henan Normal University,Xinxiang Henan 453007,China;Henan Provincial Engineering Laboratory for Intelligent Business and Internet of Things Technology,Xinxiang Henan 453007,China)
出处
《计算机仿真》
2024年第9期392-396,401,共6页
Computer Simulation
基金
国家自然科学基金(62072159,61902112)
河南省科技攻关项目(222102210011,232102211061)。
