Unlock instant, AI-driven research and patent intelligence for your innovation.

SDN network flow prediction method based on graph convolutional network

A network traffic, convolutional network technology, applied in neural learning methods, biological neural network models, electrical components, etc., can solve problems such as inability to process network topology data, and achieve the effect of dynamic changes and accurate network traffic prediction.

Pending Publication Date: 2022-07-29
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The traditional Convolutional Neural Network (CNN) is only suitable for Euclidean space, and the network topology is a non-Euclidean space, so CNN cannot process the data of network topology

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • SDN network flow prediction method based on graph convolutional network
  • SDN network flow prediction method based on graph convolutional network
  • SDN network flow prediction method based on graph convolutional network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0038] S1. Obtain a sample data set

[0039] In this embodiment, ONOS is used as the network controller, and Mininet is used as the network simulation platform to simulate the SDN network environment. The adopted topology diagram includes a total of 12 switches, 15 hosts, 15 unidirectional links, and a total of 30 bidirectional links, such as figure 1 shown. Among them, s1 and s2 are the main switches, which are used as the source address and the destination address respectively. a3s1 to a3s10 are intra-area switches. s1 is connected to 4 hosts, and the rest of the switches are connected to one host. The host h1 sends a total of four services to h11, and the data packets are 1.25Mbps, 0.2Mbps, 0.5Mbps and 1.15Mbps respectively. The link bandwidth is uniformly set to 2Mbps. After the current three packets are sent, the link load reaches 1.95Mbps, which is close to the bandwidth threshold. When the fourth packet is sent, the threshold is clearly exceeded. The service flow...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention belongs to the technical field of network traffic prediction, and particularly relates to an SDN network traffic prediction method based on a graph convolutional network. According to the method, traffic prediction is carried out by using the graph convolutional network, the whole network is abstracted as a network topology, network nodes are abstracted as points in the graph, links connected between the nodes are abstracted as edges in the graph, dynamic changes of the network are better fitted, and more accurate network traffic prediction can be realized.

Description

technical field [0001] The invention belongs to the technical field of network traffic prediction, in particular to an SDN network traffic prediction method based on a graph convolution network. Background technique [0002] Under the background of the normalization of new crown pneumonia epidemic prevention and control, people's reliance on the Internet has deepened, and all aspects of work and life, such as big data itinerary codes, takeaway orders, smart homes, and online courses, are inseparable from the stable operation of the network. , the failure of one network node or link can affect the lives of millions of people. How to prevent the influx of big data traffic or sudden failures in advance and ensure the stability of the network environment has become an urgent problem to be solved. [0003] Predicting network traffic can sense the network in an all-round way, and can solve the above problems very well. Through the analysis and feature extraction of the network t...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): H04L41/147H04L41/142H04L41/14H04L41/12H04L41/16G06N3/04G06N3/08
CPCH04L41/147H04L41/142H04L41/145H04L41/12H04L41/16G06N3/08G06N3/047G06N3/045
Inventor 纪雅欣苏俭
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA