Delay and life optimization method for cyber-physical fusion system

Abstract

The invention provides an information physical fusion system-oriented delay and life optimization method, which comprises the steps of determining system energy consumption and a system life cycle, training a long and short-term memory neural network to predict a base station task load, and designing an offline scheme of a multi-objective evolutionary algorithm based on dual local search. The optimal calculation unloading mapping and backup number is determined, and finally, an affinity-based online scheme is designed to update a calculation unloading mapping strategy; through an edge intelligent solution composed of an offline stage and an online stage, under the constraints of reliability, energy consumption and task schedulability, the purposes of simultaneously minimizing service delay and maximizing the system life are achieved, in the offline stage, a long-term short-term memory neural network is firstly used for predicting the task unloading rate of a single user group, and in the online stage, the task unloading rate of the single user group is predicted; and then determining optimal system static settings such as calculation unloading mapping and task backup number by using a double-local search multi-objective evolutionary algorithm.

Description

technical field [0001] The invention relates to the technical field of task scheduling in a cyber-physical fusion system, in particular to a delay and lifetime optimization method for a cyber-physical fusion system. Background technique [0002] In recent years, the great progress of information technology has made the physical world and the information world deeply integrated, and promoted the generation of cyber-physical fusion system. Typical applications of cyber-physical fusion system not only cover the fields of manufacturing and production, such as intelligent transportation, industrial process control, intelligent Power grids, humanoid robots, big data infrastructure, smart manufacturing, etc., also cover areas of daily life, such as healthcare, virtual / augmented reality, and smart homes, etc. How to design cyber-physical fusion systems in a low-latency manner Dealing with explosive growth of various data is a research hotspot and difficulty. To address this challen...

AI Technical Summary

Problems solved by technology

[0003] In view of the above problems, the purpose of the present invention is to propose a delay and lifetime optimization method for cyber-physical fusion systems. Under the conditions of energy consumption, reliability, and schedulability, by selecting the optimal calculation offloading mapping for tasks and the number of backups to solve the problems of high task response delay and low system life in the prior art

Images