A decision-making method for task unloading and migration based on user mobility

Abstract

The invention belongs to the technical field of mobile communication, and particularly relates to a decision-making method for task unloading and migration based on user mobility. The method comprisesthe steps that enabling a mobile user to determine the cost and income of a mobile edge cloud system according to a random connection time and task processing time relation; establishing a revenue system model with the goal of maximizing the revenue of the mobile edge cloud system; adopting a multi-stage random planning method for the revenue model; obtaining the earnings of each mobile edge cloud computing MEC and the corresponding unloading task amount of the mobile edge cloud computing MEC; and adopting an upper confidence boundary algorithm to maximize the system average income within thelearning time T, so that the MEC with the maximum system income is obtained. According to the invention, the randomness problem of the connection time between the mobile user and the MEC and the problem of how to select the MEC specifically can be solved, so that the method can be widely applied to a mobile edge cloud system.

Description

technical field [0001] The invention belongs to the technical field of mobile communication, and relates to a decision-making method for task offloading and migration based on user mobility in Mobile Edge Computing (MEC). Background technique [0002] MEC is one of the important technical means for future mobile communication systems to enhance service application capabilities. By deploying a service application platform with computing, storage, and communication capabilities at the edge nodes of the wireless access network, MEC can effectively handle the high-timeliness business needs of end users, greatly shorten the end-to-end delay, and solve the problem of the core network. A large amount of data traffic calculation and other related issues. Compared with Mobile Cloud Computing (MCC), MEC has the following advantages: [0003] (1) Ultra-low latency. For example, in a dense cell network, the information propagation distance of MEC is usually tens of meters, generally ...

AI Technical Summary

Problems solved by technology

If a MEC randomly accessed by a mobile device is serving other devices or its computing power is insufficient, the resource allocation of the MEC will be unreasonable and eventually lead to a decline in user experience

[0008] (2) Actual mobility of users: Although the current research mainly focuses on predicting the user's movement trajectory by establishing the user's random walk model, it does not consider the random connection time of the mobile user with the MEC in the MEC network

In addition, dynamic computing task offloading and migration resource allocation caused by random connection time are also problems that need to be solved urgently.

Images