Extraction Method of Encrypted Network Flow Voiceprint Feature Vector

A voiceprint feature and extraction method technology, applied in data exchange networks, digital transmission systems, electrical components, etc., can solve problems such as the inability to effectively identify protocols, the inability of traffic classification systems to correctly classify private encrypted protocol traffic, and the inability to identify hidden protocols. , to achieve the effect of rich and reliable knowledge

CN103546333BActive Publication Date: 2016-08-10TSINGHUA UNIV
3 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Publication Date
2016-08-10

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
Patent Text Reader

Abstract

The present invention provides a method for extracting voiceprint feature vectors from encrypted network streams, comprising the steps of: S2. establishing a voiceprint feature vector extraction sliding window for encrypted network streams; S3. updating the voiceprint feature vector extraction sliding window using current data packets; S5. Filtering the voiceprint feature vector to extract the sliding window; S7. Screening out unit groups with more than k units in a row from the sliding window; S9. Extracting the voiceprint features of each unit group in sequence. The voiceprint features extracted by the invention can effectively describe the dynamic features of the network flow. The method of the present invention not only inherits the advantages of the traditional network flow statistical feature extraction method, but also can provide more abundant and reliable knowledge, and can provide technical support for the design and realization of a high-performance traffic classification system in a high-speed network.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention belongs to the technical field of flow classification in network technology, and in particular relates to a method for extracting a feature vector of encrypted network flow voiceprints. Background technique

[0002] The rapid development of network technology and network bandwidth has posed new challenges to network security: First, the data traffic in the network has doubled, and network data arrives at high speed and without interruption, presenting the characteristics of massive data, which cannot be stored locally. Therefore, the efficiency of the traditional traffic classification technology based on data packet load cannot meet the needs; secondly, the services carried by the network are becoming more and more abundant, and more and more network services are encapsulated and transmitted using encryption protocols. The wide use of encryption protocols makes The traditional traffic classification technology based on packet load cannot...

Examples

Embodiment Construction

[0039] The specific implementation manners of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0040] The statistical feature extraction method adopted by the current classification technology can only extract static statistical features from the network flow, which leads to the problem that the traffic classification system cannot correctly classify the private encryption protocol traffic, let alone effectively identify the protocol hidden in the encrypted proxy channel. The invention proposes a method for extracting feature vectors of encrypted network stream voiceprints. This method first uses a sliding window to record the time-series statistical characteristics of the encrypted network flow in a continuous period of time, then filters the processing units in th...