10th EAI International Conference on Mobile Multimedia Communications

Research Article

An Objective Assessment Method of VoIP Video Quality Based on Network Parameters

Download22 downloads
  • @INPROCEEDINGS{10.4108/eai.13-7-2017.2270169,
        author={Xiaohan Zhao and Chengcai Li and Fei Wang and Jing Wang and Zesong Fei},
        title={An Objective Assessment Method of VoIP Video Quality Based on Network Parameters},
        proceedings={10th EAI International Conference on Mobile Multimedia Communications},
        publisher={EAI},
        proceedings_a={MOBIMEDIA},
        year={2017},
        month={12},
        keywords={voip video quality assessment pca network parameter neural network},
        doi={10.4108/eai.13-7-2017.2270169}
    }
    
  • Xiaohan Zhao
    Chengcai Li
    Fei Wang
    Jing Wang
    Zesong Fei
    Year: 2017
    An Objective Assessment Method of VoIP Video Quality Based on Network Parameters
    MOBIMEDIA
    EAI
    DOI: 10.4108/eai.13-7-2017.2270169
Xiaohan Zhao1, Chengcai Li1, Fei Wang1, Jing Wang1,*, Zesong Fei1
  • 1: Beijing Institute of Technology
*Contact email: wangjing@bit.edu.cn

Abstract

With the rapid development of Internet technology and popularization of the fourth generation of mobile communication system, VoIP audio & video communication is becoming more and more prevalent. On account of that all the data are transmitted through the IP network, some uncertain factors in the network such as packet loss, delay and jitter would have a strong impact on the quality of communication. Considering that circumstance, this paper proposes a kind of objective video quality assessment method based on network parameters and builds an evaluation model that could predict an objective value of the quality of video call on the basis of those network parameters. The proposed method mainly takes advantage of the Principal Component Analysis (PCA) algorithm and Neural Network (NN) technology which analyzes the parameters collected in network during VoIP video call and obtains the principle components to reduce the dimension of the data that is divided into training data and testing data further, so that it can predict the subjective score with the principle network parameters. According to the experimental results, the objective values forecasted by the proposed objective model have a strong correlation with the subjective scores of video quality gained by the video call.