Mobile edge computing has emerged as a new paradigm to enhance computing capabilities by offloading complicated tasks to nearby cloud server. To conserve energy as well as maintain quality of ser-vice, low time complexity algorithm is proposed to complete task offloading and server allocation. In this paper, a multi-user with multiple tasks and single server scenario is considered for small network, taking full account of factors including data size, bandwidth, channel state information. Furthermore, we consider a multi-server scenario for bigger network, where the in-fluence of task priority is taken into consideration. To jointly minimize delay and energy cost, we propose a distributed unsupervised learning-based offloading framework for task offloading and server allocation. We exploit a memory pool to store input data and cor-responding decisions as key-value pairs for model to learn to solve optimization problems. To further re-duce time cost and achieve near-optimal performance, we use convolutional neural networks to process mass data based on fully connected networks. Numerical results show that the proposed algorithm performs bet-ter than other offloading schemes, which can generate near-optimal offloading decision timely.

Qiuming Liu;Jing Li;Jianming Wei;Ruoxuan Zhou;Zheng Chai;Shumin Liu

School of Software Engineering,Jiangxi University of Science and Technology,Nanchang 330013,China;Nanchang Key laboratory of Virtual Digital Factory and Cultural Communications,Nanchang 330013,China

中国通信（英文版）

[1]Qiuming Liu;Jing Li;Jianming Wei;Ruoxuan Zhou;Zheng Chai;Shumin Liu-.Efficient Multi-User for Task Offloading and Server Allocation in Mobile Edge Computing Systems)[J].中国通信（英文版）,2022(07):226-238

Efficient,Multi,User,Task,Offloading,Server,Allocation,Mobile,Edge,Computing,Systems,edge,computing,has,emerged,new,paradigm,enhance,capabilities,offloading,complicated,tasks,nearby,cloud,server,To,conserve,energy,well,maintain,quality,vice,low,complexity,algorithm,proposed,complete,allocation,In,this,paper,user,multiple,single,scenario,considered,small,taking,account,factors,including,data,size,bandwidth,channel,state,information,Furthermore,bigger,where,fluence,priority,taken,into,consideration,jointly,minimize,delay,cost,distributed,unsupervised,learning,framework,We,exploit,memory,pool,store,input,cor,responding,decisions,key,value,pairs,model,solve,optimization,problems,further,duce,achieve,optimal,performance,convolutional,neural,networks,process,mass,fully,connected,Numerical,results,show,that,performs,bet,ter,than,other,schemes,which,can,generate,timely

0.634446