Method and system for recognizing end-to-end flux

A technology of network traffic and P2P traffic, which is applied in the field of communication, can solve the problems of unable to identify P2P traffic, unable to effectively control P2P traffic, and unable to identify

Inactive Publication Date: 2011-09-14
HUAWEI DIGITAL TECH (CHENGDU) CO LTD +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The inventor found through research on the prior art that this method can only identify P2P traffic using the Transmission Control Protocol (TCP, Transmission Control Protocol), but cannot identify connections using the User Datagram Protocol (UDP, User Datagram Protocol). Therefore, it is impossible to accurately and comprehensively identify the P2P traffic in the network traffic, and thus cannot effectively control the P2P traffic.

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
  • Method and system for recognizing end-to-end flux
  • Method and system for recognizing end-to-end flux
  • Method and system for recognizing end-to-end flux

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0028] See figure 1 , The end-to-end traffic identification method provided in the first embodiment of the present invention includes the following steps:

[0029] Step 101: Construct a traffic feature vector according to the number of connections and the number of ports, and construct a training sample set according to the feature vector.

[0030] Step 102: Use the training samples in the training sample set to train a support vector machine SVM classification model.

[0031] Step 103: Use the SVM classification model to identify P2P traffic from the network traffic.

[0032] Through the method provided by the embodiment of the present invention, the traffic feature vector is constructed according to the number of connections and the number of ports, and the constructed feature vector can reflect the characteristics of TCP and UDP P2P traffic, and then use the feature vector to construct a training sample set and train support vector machine classification Model, and identify network...

Embodiment 2

[0034] See figure 2 , The P2P traffic identification method provided in the second embodiment of the present invention includes the following steps:

[0035] Step 201: Construct a feature vector of P2P traffic according to the number of connections and the number of ports.

[0036] By analyzing the difference between P2P traffic and non-P2P traffic, it can be known that for P2P applications, a certain node is connected to multiple nodes or super nodes in the network, but the number of connections between the two nodes is not many. This is because To ensure the load balance of data transmission, by limiting the number of connections between two nodes, the high load between the two nodes can be avoided, and the communication pressure can be reduced. For non-P2P applications, the nodes in the network appear as a certain node and A small number of nodes in the network have connections, but there are more connections between two nodes.

[0037] Constructing the feature vector of P2P tra...

Embodiment 3

[0062] In order to establish the SVM classification model, a large amount of data needs to be collected on the network as training samples and test samples. However, among these many samples, the support vector plays a role in establishing the classification model. Other training samples are not helpful for this, so you can Tailor the training samples to reduce the size of the training sample set.

[0063] See Figure 4 This embodiment provides a cropping processing method that can be used to perform cropping processing on a normalized training sample set to obtain a cropped training sample set. The cropping processing method includes:

[0064] Step 401: Select the first training sample set from the training sample set for training, and obtain the initial hyperplane of the initial classifier.

[0065] Assuming that the training sample set is M, a part of samples from M is randomly selected as the first training sample set N. Perform SVM training on the first training sample set N ac...

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 embodiment of the invention discloses an end-to-end flow recognition method and a system thereof, wherein the method comprises the following steps: an eigenvector is constructed according to the linking number and port number and a training sample set is constructed according to the eigenvector; the training sample set is used for training a classification mode of a support vector machine; and the classification mode of the support vector machine is used for recognizing P2P flow from network flow so as to realize accurate and complete recognition of the P2P flow in the network flow and toprovide conditions for the further effective control of the P2P flow.

Description

Technical field [0001] The present invention relates to the field of communication technology, in particular to an end-to-end (P2P, Peer to Peer) flow identification method and system. Background technique [0002] With the continuous expansion of P2P technology applications, especially the continuous growth of file sharing services based on P2P systems, the abuse of resources by P2P systems, especially network bandwidth resources, has been highly valued by various network operators and school network managers. Within the education network, Maze is widely used. Only from the traffic monitoring of the Tsinghua University campus network exit, Maze's traffic has already surpassed traditional traffic such as the Web, accounting for about 15% of the entire traffic. [0003] The continuous increase of P2P services will inevitably cause huge consumption of bandwidth, which will cause network congestion and reduce network performance. Therefore, identifying P2P traffic and effectively co...

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
Patent Type & Authority Patents(China)
IPC IPC(8): H04L12/56H04L29/06H04L29/08G06K9/62
Inventor 孙知信宫婧严晓倩王飞
Owner HUAWEI DIGITAL TECH (CHENGDU) CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap