首页|基于PSO的联合任务卸载与缓存算法研究

基于PSO的联合任务卸载与缓存算法研究

扫码查看
原文链接
  • 万方数据
  • 维普
随着各类计算敏感(Computation-Sensitive,CS)型和高数据率(High-Rate,HR)服务的不断涌现,诸如移动游戏、认知辅助和虚拟/增强现实服务等,移动终端(Mobile Terminal,MT)对存储和计算资源的需求越来越大.为满足移动边缘计算(Mobile Edge Computing,MEC)网络中MT对CS和HR服务的通信需求,联合任务卸载、缓存与资源分配算法的研究被广为关注.现有研究鲜有涉及多蜂窝无线回程网络,不利于现实应用与低网络成本部署.针对支持CS和HR服务的多蜂窝无线回程网络,以MT所需资源占其服务基站的所有MT所需资源的比例分配频谱、计算与缓存资源.在该资源分配方式下,提出了最小化MT平均时延的问题,涉及MT关联指示的优化,其中MT关联兼顾基站的选择、计算和缓存模式的选择.为处理该问题,开发了基于粒子群优化(Particle Swarm Optimization,PSO)的MT关联与资源分配(PSO-based MT Association and Resource Allocation,PMARA)算法.仿真结果表明,在上述资源分配方式下,同其他现有算法相比,所设计的算法通常可以获得更低的MT时延.
Research on Joint Task Offloading and Caching Algorithm Based on PSO
With the emerging of various Computation-Sensitive(CS)and High-Rate(HR)services,such as mobile games,cognitive assistance and virtual/augmented reality,Mobile Terminal(MT)have an increasing demand for storage and computation resources.In order to meet the communication requirements of CS and HR services of MT in Mobile Edge Computing(MEC)networks,the research on the algorithms of joint task offloading,caching and resource allocation has been widely concerned.However,few existing studies focus on multi-cellular wireless backhaul networks,which are not conducive to real-world applications and low-cost network deployment.As for multi-cellular wireless backhaul networks with CS and HR services,the spectral,computing and caching resources are allocated according to the proportion of the resources required by an MT to the resources required by all MTs of its served base station.Under such resource allocation mode,an issue of minimizing the total time of MT is proposed,which involves the optimization of MT association indices.Significantly,MT indices refer to the selection of base station,and computing and caching modes.To address the above issue,a Particle Swarm Optimization(PSO)-based MT Association and Resource Allocation(PM ARA)algorithm is developed.The simulation results show that the designed algorithm generally achieves a lower delay than another existing algorithm under the above-mentioned resource allocation mode.

MECmobile edge cachingPSOresource allocation

周天清、许铭

展开 >

华东交通大学信息工程学院,江西南昌 330013

移动边缘计算 移动边缘缓存 粒子群优化 资源分配

国家自然科学基金国家自然科学基金国家自然科学基金国家自然科学基金国家自然科学基金国家重点研发计划江西省自然科学基金江西省自然科学基金江西省自然科学基金江西省自然科学基金江西省自然科学基金研究生创新资金项目

62261020620620346217111961961020620012012020YFB180720120232ACB21200520224BAB20200120232BAB20201920212BAB20200420212BAB212001YC 2022-s549

2024

10.3969/j.issn.1003-3106.2024.03.001

无线电工程
中国电子科技集团公司第五十四研究所

无线电工程

影响因子：0.667
ISSN：1003-3106
年,卷(期)：2024.54(3)
  • 19