MECDLRL

AI Driven Heterogeneous MEC System with UAV Assistance for Dynamic Environment - Challenges and Solutions

Matlab code to reproduce our works.

• AIUE.m: Convert independent output AI solutions into time slice AI solutions

• allocUE.m: Convert the generated solution into a normative solution (satisfy constraints)

• CalR.m: Calculate the distance between UE and MEC

• CalT.m: Calculate the delay time

• cluUE.m: Classify current users according to the distance from MEC

• corresult.m: Correct the output solution of the neural network

• DNNopt2.m: Code for DNN in a dynamic environment to solve a and f

• drawfigs.m: Optimization and plot

• DRLopt.m: The code use DRL to solving a and f

• epsgreedy.m: Random Algorithm in RL

• fitfunc.m: Calculation time delay as the fitness function

• GreedyUE.m: Greedy algorithm to solve this problem

• LocalUE.m: Execute the tasks locally

• RandomUE.m: Random algorithm to solve this problem

• SA.m: The simulated annealing algorithm in RL

• WheelSel.m: Use roulette algorithm to generate random individuals

• main.m: run this file, including setting system parameters

About our works

1. Feibo Jiang ; Kezhi Wang ; Li Dong ; C. Pan; W. Xu and K. Yang. AI Driven Heterogeneous MEC System with UAV Assistance for Dynamic Environment - Challenges and Solutions [J]. IEEE network, 2020, DOI: 10.1109/MNET.011.2000440.

About authors

• Feibo Jiang (jiangfb@hunnu.edu.cn) is with Hunan Provincial Key Laboratory of Intelligent Computing and Language Information Processing, Hunan Normal University, Changsha, China,
• Kezhi Wang (kezhi.wang@northumbria.ac.uk) is with the department of Computer and Information Sciences, Northumbria University, UK,
• Li Dong (Dlj2017@hunnu.edu.cn) is with Key Laboratory of Hunan Province for New Retail Virtual Reality Technology, Hunan University of Commerce, Changsha, China,
• Cunhua Pan (Email: c.pan@qmul.ac.uk) is with School of Electronic Engineering and Computer Science, Queen Mary University of London, London, E1 4NS, UK,
• Wei Xu (wxu@seu.edu.cn) is with NCRL, Southeast University, Nanjing, China,
• Kun Yang (kunyang@essex.ac.uk) is with the School of Computer Sciences and Electrical Engineering, University of Essex, CO4 3SQ, Colchester, UK and also with University of Electronic Science and Technology of China, Chengdu, China

Required

• Matlab 2020a

• DeepLearnToolbox

How the code works

• run the file, main.m