End-to-end flow reconfiguration method in time-varying dynamic network

Abstract

The invention provides an end-to-end flow reconfiguration method in a time-varying dynamic network. The end-to-end flow reconfiguration method comprises the following steps of: step 1, carrying out end-to-end flow wavelet transform; step 2, extracting time-frequency domain characteristic parameters of an end-to-end flow; step 3, reconfiguring the end-to-end flow; and step 4, correcting a reconfiguration value of the end-to-end flow. The end-to-end flow is reconfigured by using a multi-fractal wavelet model; a link load is used as a restrain parameter; the end-to-end flow is accurately reconfigured through multi-resolution ratio analysis; and the ill-conditioned problem of the end-to-end flow reconfiguration is solved from a perspective of time-frequency domain combined analysis by both considering self-similarity property and multi-fractal property of the end-to-end flow, so that the end-to-end flow is more accurately reconfigured. The end-to-end flow reconfigured by using the method provided by the invention is used for carrying out activities in network management, such as network planning, load balance, fault diagnosis, route optimization and the like.

Description

technical field [0001] The invention belongs to the field of network technology, relates to the field of large-scale end-to-end flow measurement and analysis, in particular to a reconstruction method of end-to-end flow in a time-varying dynamic network. Background technique [0002] With the rapid development of the Internet in recent years and the increasing number of network applications, the traffic characteristics in the backbone network have undergone tremendous changes, and the end-to-end traffic shows randomness, time-varying and non-stationarity. And according to recent research, end-to-end traffic in large backbone networks has fractal properties, including long-term correlation, burstiness and self-similarity. The traditional end-to-end traffic model can no longer describe the traffic characteristics of the current backbone network well. End-to-end traffic is used for network management activities such as network planning, load balancing, fault diagnosis, and rout...

AI Technical Summary

Problems solved by technology

Although there are many ways to reconstruct the end-to-end traffic, but in the IP backbone network, the end-to-end traffic reconstruction itself is a highly pathological problem

The time-varying dynamics of large-scale backbone networks make end-to-end traffic random, time-varying, non-stationary, and multifractal, making end-to-end traffic reconstruction more difficult

Images