Video lag detection method based on flow analysis

Abstract

The invention provides a video lag detection method based on flow analysis. The video lag detection method comprises the following steps: collecting flow at a network access point of an operation service provider; filtering out hot video traffic from user traffic acquired by the network access point according to the hot video server domain name, the ip address, the port and the hot video number IDmatching rule; detecting a video transmission pulse phenomenon and extracting pulse information in the video data stream; calculating the extracted pulse information to obtain an average playing coderate in a video playing process; collecting user traffic data at the network access point of the operation service provider and filtering hot video traffic; extracting video playing information; calculating the size of the video data in the video buffer area of the user terminal by utilizing the solved average playing code rate and the video playing information; and detecting the video lag condition of the terminal user. The method can be used for an operation service provider to evaluate the video service quality of the user terminal on the network access point, and provides a basis for improving the access service quality.

Description

technical field [0001] The invention belongs to the technical field of network space security, and relates to a video freeze detection method based on flow analysis. Background technique [0002] In recent years, due to the popularity of 4G networks and the continuous improvement of network infrastructure, video traffic has occupied a dominant position in global traffic. By the end of 2019, video downstream traffic has reached 60.6% of global application traffic. At the same time, with the development of 5G, the application of video services will become more and more extensive. Such rapid growth has forced network mobile operators to fundamentally reconsider and optimize their operating networks. For such optimization and capacity planning, operators must have in-depth knowledge and monitoring of the quality of experience (QoE) of video delivery. Aiming at limited mobile network bandwidth, to reasonably provide network access services and provide high-quality mobile video ...

AI Technical Summary

Problems solved by technology

However, this type of method has the following problems: on the one hand, it is difficult for the operation service provider to decode the video played on the client at the network access point and extract image features; on the other hand, it is very difficult for the operation service provider to It is difficult to reproduce the real user terminal playback scene, so the availability of this solution for network operators needs to be further studied

However, there are problems in the method of obtaining these parameters in this type of method. For example, when calculating the playback bit rate of video clips, some prior knowledge is needed, such as the video playback bit rate range that needs to be satisfied by different video platforms at different resolutions; A large amount of video freeze data is trained to obtain a model to predict freeze. Before training, a large number of video traffic features need to be extracted, which is difficult to implement

Images