Computer Engineering and Applications ›› 2022, Vol. 58 ›› Issue (7): 150-161.DOI: 10.3778/j.issn.1002-8331.2109-0155

• Pattern Recognition and Artificial Intelligence • Previous Articles     Next Articles

Improved Artificial Bee Colony Algorithm and Its Application in Edge Computing Offloading

ZHANG Chengrui, KE Peng, YIN Mei   

  1. 1.School of Computer Science and Technology, Wuhan University of Science and Technology, Wuhan 430065, China
    2.Hubei Province Key Laboratory of Intelligent Information Processing and Real-Time Industrial System, Wuhan 430065, China
  • Online:2022-04-01 Published:2022-04-01



  1. 1.武汉科技大学 计算机科学与技术学院,武汉 430065
    2.智能信息处理与实时工业系统湖北省重点实验室,武汉 430065

Abstract: Mobile edge computing(MEC) reduces computing latency and energy consumption by placing computing power at the edge of the network. An artificial bee colony algorithm based on one-dimensional and multi-dimensional dynamic population(OMABC) strategy is proposed to realize the offloading of computationally intensive and time-sensitive application scenarios. Firstly, establish an edge computing offloading model that includes cloud servers, and construct a cost function with energy consumption as a penalty term to minimize delay. Secondly, the offloading decision of the computing task is transformed into the process of optimizing the cost function of the artificial bee colony algorithm. Finally, the effectiveness of OMABC is verified on the CEC 2017 test function. In the edge computing simulation, it is compared with the local offloading strategy, random offloading strategy, the offloading strategy based on particle swarm optimization(PSO) and the offloading strategy based on artificial bee colony algorithm(ABC). The results show that the edge computing offloading strategy based on OMABC can effectively reduce the cost function of the MEC system and provide more efficient services.

Key words: artificial bee colony algorithm, mobile edge computing, computing offloading, multi-dimensional update, dynamic population

摘要: 移动边缘计算(MEC)通过将算力下沉到网络边缘来降低计算时延和设备能耗。针对计算密集型和时延敏感型应用场景,提出了一种单多维动态种群策略的人工蜂群算法(OMABC)来实现计算任务的卸载。建立一个包含云服务器的边缘计算卸载模型,并构建一个以能耗为惩罚项的代价函数;将计算任务的卸载决策转化为人工蜂群算法对代价函数的寻优过程。通过仿真实验,在CEC 2017测试函数上验证了OMABC的有效性,并在边缘计算模型仿真中与本地卸载策略、随机卸载策略、基于粒子群算法(PSO)的卸载策略、基于人工蜂群算法(ABC)的卸载策略进行对比。实验结果表明,基于OMABC的边缘计算卸载策略能够有效降低MEC系统的时延及代价函数,提供更高效的服务。

关键词: 人工蜂群算法, 移动边缘计算, 计算卸载, 多维更新, 动态种群