北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2023, Vol. 46 ›› Issue (2): 29-36.

• 算力网络与分布式云 • 上一篇    下一篇

一种集成学习辅助DDPG的资源优化算法

公雨,魏翼飞   

  1. 北京邮电大学
  • 收稿日期:2022-03-15 修回日期:2022-08-31 出版日期:2023-04-28 发布日期:2023-05-14
  • 通讯作者: 公雨 E-mail:gongyu428@bupt.edu.cn
  • 基金资助:
    可编程绿色边缘网络架构及智能资源优化研究

An Ensemble Learning-Aided DDPG Resource Optimization Algorithm

1, 2   

  1. 1. Beijing University of Posts and Telecommunications
    2.
  • Received:2022-03-15 Revised:2022-08-31 Online:2023-04-28 Published:2023-05-14

摘要: 构建了一个通信、计算和缓存(3C)的集成体系架构,以解决任务调度和资源分配的联合优化问题。为了协调网络功能,动态分配有限的3C资源,采用深度强化学习方法,联合考虑用户请求业务的多样性和动态的无线信道条件,获得移动虚拟网络运营商的最大利润函数。仿真结果表明,基于DRL的资源分配方案明显优于其他两种比较策略。结合集成学习的DRL算法输出结果速度更快,性价比更高。

关键词: 网络切片, 多址边缘计算, 内容缓存, 资源分配, 深度强化学习, 集成学习

Abstract: An integrated architecture of communication, computing and caching (3C) is constructed to solve the joint optimization problem of task scheduling and resource allocation. In order to coordinate network functions and dynamically allocate limited 3C resources, the deep reinforcement learning algorithm is adopted to obtain the maximum profit function of mobile virtual network operators by jointly considering the diversity of user request services and dynamic wireless channel conditions. Simulation results show that the resource allocation scheme based on DRL is superior to the other two comparison strategies. The DRL algorithm combined with ensemble learning has a faster output speed and higher cost performance.

Key words: network slicing, multi-access edge computing, content caching, resource allocation, deep reinforcement learning, ensemble learning

中图分类号: