Video transmission data fine measurement method

Abstract

The invention discloses a refined measurement method for video transmission data. The method comprises the following steps: acquiring network flow data; constructing a multi-stream space-time characteristic traffic matrix according to the network stream data; performing feature dimension reduction on the multi-stream space-time characteristic traffic matrix to obtain low-dimensional features of the network stream; and training and classifying the low-dimensional features in the low-dimensional feature space, and associating the network flow data. Through detailed analysis of time-space interactivity of network flows, relevance of the network flows in terms of time and space characteristics is mined, spatial-temporal characteristics of multiple encrypted flows and non-encrypted flows associated with applications are represented by using a flow matrix, dimension reduction is carried out on the complex and high-dimensional flow matrix by using a manifold theory, and the logic process between traffic behavior modes and behaviors in a mainstream application communication process is mined so as to realize refined measurement of encrypted traffic.

Description

technical field [0001] The invention relates to the technical field of communication, in particular to a method for finely measuring video transmission data. Background technique [0002] With the rapid development of Internet networks and online applications, in order to ensure communication security and privacy, more and more video network traffic is encrypted. Encryption is an important means of protecting privacy. It can protect our data from prying eyes and prevent crimes. Molecules steal users' credit card information, app usage habits, or passwords. [0003] Obtaining the description of attack behavior can provide us with certain intelligence information, but it is necessary to accurately classify attack behavior in order to achieve follow-up requirements such as attack defense, filtering, or portrait of the attacker. However, there are generally multiple network flows in the communication process of different normal applications and malware, and these network flows ...

AI Technical Summary

Problems solved by technology

[0005] In view of this, the embodiment of the present invention provides a refined measurement method for video transmission data to solve the problem of multiple normal applications and malicious software communication processes. Network flow, these network flows include encrypted flow or non-encrypted flow, and the spatio-temporal interaction behavior between these network flows is quite complex, and the interaction modes of different applications are different, which leads to the difficulty of fine-grained measurement of encrypted traffic

Images