摘要
由于移动边缘计算网络在边缘位置部署,在多用户并发的情况下带宽资源优化策略容易出现高计算负荷;提出基于萤火虫算法的移动边缘计算网络带宽资源优化策略;在服务器之间数据连续传输的情况下,确定网络内用户分布情况,计算网络运行需要消耗的能量,利用萤火虫算法建立以带宽资源为中心的数学模型,以移动边缘计算网络的各项参数作为依据,对数学模型求解,得到最优解后,以用户最大收益为目标部署优化策略;实验结果表明:提出的基于萤火虫算法中当移动用户数量从10个增加到100个,网络平均能耗在8J到13J之间;当移动用户数量从10个增加到80个,边缘计算能力分别是4 GHz、8 GHz两种计算条件下,平均延迟均在1 s以内;提出的算法在收敛速度和计算性能方面都具有优势,具有很好的参考价值。
Because mobile edge computing networks are deployed at the edge,the bandwidth resource optimization strategy is prone to the high computing load in the case of multiple users.In order to solve this problem,a mobile edge computing network bandwidth resource optimization strategy based on Firefly algorithm is proposed.In the case of continuous data transmission between servers,the distribution of users in the network is determined,the energy consumed by the network operation is calculated,the Firefly algorithm is used to establish a mathematical model centered on bandwidth resources,and solve the mathematical model based on various parameters of the mobile edge computing network.After the optimal solution is obtained,the optimization strategy is deployed with the maximum benefit of users as the goal.The experimental results show that when the number of mobile users increases from 10 to 100,the average energy consumption of the Firefly-based algorithm is between 8 J and 13 J.When the number of mobile users increases from 10 to 80,the edge computing capabilities are 4 Ghz and 8 Ghz respectively,the average delay of the edge computing is within 1 s.The proposed algorithm has advantages in the rate of convergence and computational performance.It has a great reference value for mobile edge computation.
作者
王宏杰
徐胜超
陈刚
杨波
毛明扬
WANG Hongjie;XU Shengchao;CHEN Gang;YANG Bo;MAO Minyang(School of Data Science,Guangzhou Huashang College,Guangzhou 511300,China)
出处
《计算机测量与控制》
2023年第11期280-285,共6页
Computer Measurement &Control
基金
国家自然科学基金面上项目(61772221)
广州华商学院校级导师制科研项目(2023HSDS06)。
关键词
萤火虫算法
边缘计算
网络带宽
资源优化
策略分析
firefly algorithm
edge calculation
network bandwidth
resource optimization
strategy analysis
