Research On QoE Model For Video Telephone Service In Wireless Network

With the rapid development of multimedia computer technology and communication technology, people's requirements for communication are getting higher and higher, and are no longer satisfied with simple text and voice communications, but tend to choose multimedia communication which collects text, speech and image in a body, that is to say, people have put forward higher requirements on the form and content of communications. In the era of mobile Internet, more and more users start to use VoLTE, Wechat and QQ for video calls by utilizing the advantage of LTE (Long Term Evolution), and video telephone service gradually popularize. The perception quality for users to video telephone service has effect on the future development of video telephone service. At present, the research on users' perception quality evaluation model for video telephone service is still in progress. We focus on this topic in this paper.This paper first introduces the subjective and objective evaluation methods of users'QoE (Quality of Experience), the tools for establishment of mathematical modeling, then introduces current research state of video telephone service which was standardized as ITU-T G.1070 by ITU.G.1070 model covers a number of audio and video parameters which makes the formulas cumbersome, and practical application is not very convenient. Operators tend to have simpler and more intuitive business models. So we study the end-to-end QoE of video telephone service,propose parameters which affect the user experience and are easier to collect in practical application, such as resolution, bit rate, delay between audio and video; then design experiment program, make damage samples for video telephone service, conduct subjective test, collect and screen data,and then establish the QoE model. Linear regression analysis and BP neural network are used in the modeling, and the performance of the two models is compared and analyzed after that. On this basis, the impact of network factors on the QoE of video telephone service is analyzed from the perspective of network transmission, and this paper focuses on the impact of packet loss and delay on the video telephone user experience in wireless network environment.The main work of this paper is as follows:1. Research on end-to-end QoE of video telephone service. For parameters that may affect the perceived quality of the video telephone servi ce, the QoE model of video telephone service is established by using statistical multiple linear regression analysis and BP neural network of machine learning respectively. And then the performance of the two models are compared and analyzed. The verification results show that both models can reflect the users' perception of video telephone quality accurately.2. Research on the impact of packet loss ratio and delay on QoE of video telephone service from the perspective of network transmission. We use NS-3 simulation platform to build LTE wireless network environment,and simulate network throughput, packet loss and delay of the situation.Research on the impact of packet loss and delay on QoE of video telephone service in the WebRTC video broadcast platform using network simulator.In the study of delay factors, design an interactive experiment with non-free dialogue, and divide into two scenes according to the length of the average speaking time, and explore the effect of delay on perceived quality under different average speech time.