Random task queuing unloading optimization method based on edge computing

Abstract

The invention provides a random task queuing unloading optimization method based on edge computing and belongs to the technical field of wireless communication. The method comprises the following steps: firstly, the probability that a task generated by a user MDi is locally executed as the probability that the MDi unloads the task is represented through a macro base station and the probability that the task is unloaded through a small base station respectively, wherein the queuing model of the tasks processed locally is an M / M / 1 queue, and the queuing model of the tasks unloaded to the serverfor processing is an M / M / c queue. Secondly, a user-centered time delay and energy consumption minimization optimization target is established, and the decision probability is utilized to reflect the intention of the user to select different paths to execute tasks; the objective of the invention is to solve the problems of time delay and energy consumption minimization. A task allocation algorithmbased on a quasi-Newton interior point method is provided, the optimal solution of a target variable is regarded as a combination, the inverse of a target function Hessian matrix is replaced with an approximate matrix Dk by using a quasi-Newton condition, the Dk matrix is continuously updated through an iterative formula, the optimal search direction and the search step length are updated, and finally the optimal solution is approximated.

Description

technical field [0001] The present invention relates to a random task queuing and unloading optimization scheme based on edge computing, and in particular to the problem of extra consumption of the system caused by network congestion as the demand for data services grows in an edge computing network system. Background technique [0002] With the rapid development of technology, mobile device traffic has increased dramatically. However, due to their limited resources and computing performance, smart mobile devices may face insufficient capabilities when processing computing-intensive and time-sensitive applications. For this reason, an edge computing model that uses network edge nodes to process and analyze data emerged as the times require, and complements the traditional cloud computing model. However, edge devices often have the characteristics of light weight, how to rationally utilize the limited computing resources of the edge has become an important problem that edge ...

AI Technical Summary

Problems solved by technology

[0005] In patents based on edge computing task offloading, the modeling of traditional methods is too idealized and does not consider more offloading obstacles in real communication scenarios, so the application range is narrow and the scalability is low

Images