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

A HTTP-based dynamic adaptive encryption video traffic identification method

A dynamic self-adaptation and traffic recognition technology, applied in image communication, selective content distribution, electrical components, etc., can solve the problems of being easily affected by noise and high false positives

Active Publication Date: 2021-07-30
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, the common methods classify and identify by characteristics such as packet length distribution, packet arrival time interval, and transmission direction. The accuracy is relatively rational in some specific environments, but there are still some shortcomings: such as high false positives; most methods use closed model assumptions , that is, the range of streaming video is determined in advance; it is susceptible to noise; it is assumed that the attacker can obtain the encrypted traffic of the network layer and the physical layer, etc.

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
  • A HTTP-based dynamic adaptive encryption video traffic identification method
  • A HTTP-based dynamic adaptive encryption video traffic identification method
  • A HTTP-based dynamic adaptive encryption video traffic identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0039] As a most basic implementation of the present invention, this embodiment discloses a dynamic adaptive encryption video traffic identification method based on HTTP, including a data acquisition step, a data preprocessing step, a similarity calculation step and a classification and identification step:

[0040] The data acquisition step is to obtain the encrypted video data flow from the network, remove the irrelevant flow in the encrypted video data flow of the known video title, and save it as a training method for data packet burst flow feature extraction and video classification construction data set.

[0041] The encrypted video data flow includes a known title and an original video data flow to be identified, wherein the video data flow of a known title is used to construct a reference benchmark BPP time series, and BPP refers to the time sequence from the server to the client during a complete peak duration. The number of bidirectional transmission bits at the end....

Embodiment 2

[0057] As a preferred implementation of this embodiment, on the basis of Embodiment 1, further, the encrypted video data flow of the known video title and the encrypted video flow to be identified include several videos, and each video contains Both contain several video streams.

[0058] The BPP time series is a time series corresponding to video streams in each video.

[0059] In the data preprocessing step, each peak stage in the encrypted video traffic to be identified is obtained through network monitoring software.

[0060] In the similarity calculation step, the DTW algorithm recursively calculates the DTW value of the jth single video stream in the encrypted video traffic data video i to be identified by the formula of Cost=d(s[i], t[j]) , ie DTW[i, j]=Cost+Minimum(DTW[i-1, j], DTW[i, j-1], DTW[i-1, j-1]), DTW[i, j] The larger the value, the lower the similarity between the corresponding BPP time series and the time series of the data set. If the value of DTW[i, j]=0...

Embodiment 3

[0066] As another preferred implementation of this embodiment, this embodiment discloses an HTTP-based dynamic adaptive encrypted video traffic classification and identification method, including the following steps:

[0067] Step S1, network encrypted video traffic data collection and preprocessing;

[0068] Step S1 comprises the following steps:

[0069] Step S11. Obtain encrypted video data traffic of known video titles from the network, remove irrelevant traffic, and save files for feature extraction of data packet burst traffic to construct a training data set.

[0070] Step S12, preprocessing the data packets, the goal is to form a BPP time series with video title marks, that is, to form a marked BPP sequence for known titles, and to form a BPP sequence for the video traffic to be identified for testing, and the BPP is The number of bits of data packets sent and received by each peak is obtained through the preprocessing of the PCAP file.

[0071] Step S2, calculating ...

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 computer network security, discloses an HTTP-based dynamic self-adaptive encrypted video traffic classification and identification method, and belongs to the technical field of network security. Including encrypted video traffic data acquisition and preprocessing, calculation of time series similarity, similarity judgment through the proposed nearest neighbor and nearest class algorithms, etc. The advantage of the present invention is that only one feature of BPP of encrypted video flow is used, and the video in the flow can be identified without decrypting the flow. This method does not decrypt traffic, does not violate user privacy, and achieves better accuracy.

Description

technical field [0001] The invention belongs to the technical field of computer network security, and in particular relates to an HTTP-based dynamic self-adaptive encryption video flow identification method. Background technique [0002] A large number of Internet users will watch videos online. It is predicted that by 2020, online video viewing traffic will account for about 80% of the entire Internet traffic. Internet usage reports show that more than half of Internet traffic has been encrypted, and the field of analysis and identification of encrypted video traffic has attracted widespread attention from the academic and industrial circles of network security research. Mainstream video streaming sites such as Youtube, Netflix, etc. use HTTP (Hypertext Transfer Protocol)-based dynamic adaptive video streaming technology DASH (Dynamic Adaptive Streaming over HTTP: DASH or MPEG-DASH), which uses traditional HTTP web servers to spread video on the Internet Flow, which not on...

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): H04N21/234H04N21/44H04N21/2347H04N21/8547H04N21/24H04N21/442H04N21/845H04N21/8543
CPCH04N21/23418H04N21/2347H04N21/2407H04N21/44008H04N21/44245H04N21/8456H04N21/8543H04N21/8547
Inventor 周琨汪文勇唐勇黄鹂声张骏张文刘宝阳
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA