Task unloading and resource allocation joint optimization method based on edge cooperation

Abstract

The invention relates to a task unloading and resource allocation joint optimization method based on edge cooperation. The invention belongs to the technical field of mobile communication, the methodaims an edge network environment with sharp mobile service flow and non-uniform space distribution, real-time perception and centralized management and control are carried out on resources of networkequipment through a centralized system architecture based on a software defined network (SDN), and a theoretical model of user fairness time delay minimization is established under the constraint of user maximum tolerance delay and MEC server computing resources, so that the maximum time delay in all users is minimized, and joint optimization is performed on a task unloading strategy and a resource allocation scheme through a heuristic iterative algorithm framework based on a simulated annealing thought. According to the method, the computing load of the MEC server is balanced, the advantagesof the MEC technology are brought into full play, the user task execution delay is reduced while the fairness of the user is ensured, and the resource utilization rate and the system performance of the MEC system are improved.

Description

technical field [0001] The invention belongs to the technical field of mobile communication, and relates to a joint optimization method for task offloading and resource allocation based on edge cooperation. Background technique [0002] Mobile edge computing (MEC) effectively solves the high latency and backhaul problems faced by mobile cloud computing (MCC) by migrating the cloud computing platform from the core network to the edge of the mobile access network. The limitation problem is regarded as one of the key technologies of the fifth generation (5G) mobile communication network. However, with the vigorous development of mobile communication technology and the Internet, emerging applications represented by virtual reality (Virtual Reality, VR), augmented reality (Augmented Reality, AR), interactive games, etc. As a result, the MEC server with limited resources cannot provide continuous and endless computing offloading services for all associated user terminals at the s...

AI Technical Summary

Problems solved by technology

[0005]Most of the existing literature focuses on the scenario of multiple users on a single MEC server. There are few studies on multi-MEC servers and the cooperation between servers is not considered, which reduces the number of MEC servers. Secondly, some literatures offload all the tasks to the MEC server for execution when studying the edge collaboration problem, ignoring the computing power of the user terminal itself

In addition, since a centralized network architecture is not considered, frequent signaling interactions are required to obtain resource information of each edge node, which increases system energy consumption; finally, when designing offloading strategies and resource allocation schemes, all MEC servers in the network must For optimization, user fairness should also be considered to improve user QoE and system performance

Images