Encrypted Traffic Identification Method Based on Flow Gradient Orientation

Abstract

The invention relates to an encrypted traffic identification method based on stream gradient orientation. The method comprises the steps of calculating data stream gradient orientation key identifiers in a known training set, extracting network data traffic, carrying out key identifier analysis, and calculating key identifiers of target encrypted traffic service and non-target encrypted traffic service in a network; for a to-be-measured unknown data stream, calculating the orientation key identifier, a related offset of the data stream gradient orientation key identifiers between the unknown traffic and the target encrypted traffic service, and the related offset of the data stream gradient orientation key identifiers between the unknown traffic and the non-target encrypted traffic service, judging the magnitude of the related offsets, further judging that the unknown stream is the target encrypted traffic service or the non-target encrypted traffic service. The method is high in identification rate and high in usability, and the method is applicable to any network encrypted traffic identification, supports evolution of the network and is compatible with network encrypted traffic identification possibly appearing in the future.

Description

technical field [0001] The invention belongs to the technical field of computer networks, and in particular relates to an encryption flow identification method based on flow gradient guidance. Background technique [0002] Peer-to-Peer (P2P) technology has been widely used in the current Internet, for example, streaming media services, VoIP, file sharing and many other fields adopt peer-to-peer network transmission technology. Because this technology is easy to implement, has strong carrying capacity, and is suitable for individual users, it has a very high usage rate in the network. However, the open nature of peer-to-peer network services makes its security unable to meet the needs of the current network. Various encrypted Trojan horse viruses, malware, and pirated information are widely spread in the peer-to-peer network. How to improve network security has become a severe problem. challenge. The existing encrypted traffic identification methods include: encrypted traff...

AI Technical Summary

Problems solved by technology

[0003] In order to overcome the deficiencies in the prior art, the present invention provides an encryption flow identification method based on flow gradient orientation, which solves the defects in the identification rate and complexity of encryption flow in the prior art, and realizes the gradient orientation of network data flow. Better recognition rate of encrypted flow, further guarantee the security and stability of network information, applied to all levels of nodes in data transmission network, applicable to any network encrypted flow recognition

Images