首页|多设备多任务场景下基于改进粒子群优化的计算卸载策略

多设备多任务场景下基于改进粒子群优化的计算卸载策略

扫码查看
原文链接
  • 万方数据
  • 维普
在移动边缘计算网络中,针对多用户场景下本地设备上多个计算密集型任务的计算卸载问题,为获得最优的任务卸载决策和资源分配方案,提出了一种基于改进粒子群优化的计算卸载策略.首先,综合考虑时延和能耗相关的计算卸载总代价以及服务器任务均衡,通过本地设备的剩余能量和充电状态信息自适应调整时延与能耗权重,以最小化系统总代价为目标,建立多用户、多任务、多服务器的计算卸载模型.然后,使用改进粒子群优化算法来求解该问题,最终获得最优的任务卸载决策和资源分配方案.仿真结果表明,该方案相对于基于遗传算法的卸载方案,能够减小 20%系统代价.
Computation Offloading Strategy Based on Improved Particle Swarm Optimization in Multi-user and Multi-task Scenarios
In mobile edge computing(MEC)networks,to solve the problem of computation offloading for multiple computing-intensive tasks on multiple local devices,a computation offloading strategy based on improved particle swarm optimization(PSO)was proposed to obtain the optimal offloading decision and resource allocation scheme.First,aiming at minimizing the computational offload cost related to the delay and the energy consumption,considering the task balance between MEC servers,a computation offloading system model in a multi-user,multi-task,and multi-server scenario was built.The weights of the delay and the energy consumption were adjusted adaptively according to the residual energy and charging state of the devices.Then,to minimize the corrected weight sum of the delay and energy consumption,an improved PSO algorithm was proposed to obtain the optimal task offload decision and computing resource allocation scheme.The simulation results showed that,the proposed scheme could reduce the system cost by 20%compared with the offloading scheme based on genetic algorithm.

mobile edge computingcomputation offloading strategyparticle swarm optimizationmixed integer nonlinear program-mingresource allocation

蒋鹏、富爽、丁晨阳

展开 >

黑龙江八一农垦大学信息与电气工程学院,大庆 163319

移动边缘计算 计算卸载 粒子群算法 混合整数非线性规划 资源分配

黑龙江省自然科学基金优秀青年基金国家留学基金黑龙江八一农垦大学青年创新人才培养计划

YQ2019F014201708230301ZRCQC201807

2024

10.3969/j.issn.1002-2090.2024.01.015

黑龙江八一农垦大学学报
黑龙江八一农垦大学

黑龙江八一农垦大学学报

影响因子：0.888
ISSN：1002-2090
年,卷(期)：2024.36(1)
  • 25