Chaotic dynamic congestion prediction-based time-varying rerouting method and system for space-air-ground integrated network

Abstract

The invention discloses a time-varying rerouting method and system based on chaotic dynamic congestion prediction for a space-air-ground integrated network. According to the method and the system, in order to meet the requirement of service quality, an LSTM neural network optimized by a PSO algorithm based on wavelet analysis is designed for the topology dynamics of space-air-ground integration, network congestion is predicted in advance to avoid congestion, meanwhile, attribute reduction is performed by using a neighborhood rough set model, and the service quality is improved. And selecting the parameter set with the maximum correlation with the network congestion to pre-process the network parameters. A time-varying thought is introduced, the state of a link is dynamically calculated, and a high-reliability and load-balanced path is found according to the state of the link. A simulation experiment verifies that the congestion prediction precision of the chaotic dynamic congestion prediction method is greatly improved, and the planned rerouting path has high reliability and stability.

Description

technical field [0001] The invention belongs to the field of air-space-ground integrated information networks, and relates to a time-varying rerouting method and system based on chaotic dynamic congestion prediction of an air-space-ground integrated network. Background technique [0002] Compared with the traditional network environment, the air-space-ground integrated network environment is quite different. The dynamics and heterogeneity of air-space-ground integrated networks bring challenges to network congestion prediction. [0003] Network congestion prediction can be regarded as traffic prediction in essence. There are mainly the following ways to predict network traffic, naive method, linear parameterized model and nonlinear parameterized model. The naive method is a simple and efficient time series forecasting method, which uses the average data of the past moment as the actual forecast data of the next moment, and the historical average (Historical Average, HA) is ...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art, and provide a time-varying rerouting method and system based on chaotic dynamic congestion prediction of the air-space-ground integrated network, so as to solve the problem that the traditional neural network model in the prior art is difficult to meet the actual The problem of network congestion prediction demand

Images