计算机工程与应用 ›› 2017, Vol. 53 ›› Issue (12): 133-139.DOI: 10.3778/j.issn.1002-8331.1601-0177

• 模式识别与人工智能 • 上一篇    下一篇

移动视频体验质量的量化评价模型

张家辰,刘  浩   

  1. 东华大学 信息科学与技术学院,上海 201620
  • 出版日期:2017-06-15 发布日期:2017-07-04

Quantitative evaluation model for mobile video quality of experience

ZHANG Jiachen, LIU Hao   

  1. College of Information Science and Technology, Donghua University, Shanghai 201620, China
  • Online:2017-06-15 Published:2017-07-04

摘要: 移动视频体验质量(QoE)的研究正变得日益重要,现有的QoE评价模型缺少相应的量化方法。结合移动视频业务的特点,提出了两层权值的QoE量化评价模型,即QoE由六种“影响层面”按各自的一级权值叠加而成;进一步地,每个“影响层面”又由多个“影响因素”构成。除了丢包和抖动两个“影响因素”外,其他“影响因素”均具有独立的二级权值。提出通过层次分析法(AHP)计算出所有的一级权值,并且进一步计算出除了丢包、抖动外的其他二级权值。由于丢包和抖动这两个影响因素相互依赖,无法计算各自的二级权值,因此通过网络模拟实验,找出两者的双变量函数f(丢包,抖动),由此得到完整的两层权值QoE量化评价模型。所提QoE量化评价模型通过测评人员的QoE评分进行验证,实验结果表明该模型获得了较高的准确度,证明了上述模型量化方法的可靠性。

关键词: 移动视频, 体验质量, 量化评价模型, 层次分析法

Abstract: The research on mobile video Quality of Experience (QoE) is becoming increasingly important, but the existing QoE evaluation models lack the appropriate quantitative methodology. Considering the characteristics of mobile video services, a two-level weighting QoE quantitative evaluation model has been proposed in this paper. The proposed QoE model is made up of six kinds of influence levels according to their first-level weights;further, each influence level consists of multiple influence factors. In addition to packet loss and jitter factors, other influence factors have their second-level weights. This paper utilizes the Analytic Hierarchy Process (AHP) method to calculate all first-level weights, and then calculate most of second-level weights in addition to packet loss and jitter factors. Since packet loss and jitter are inter-dependent, it is unable to directly calculate their weights. Therefore, this paper finds their function f(packet loss, jitter) by network simulation experiments, and thus obtains a complete QoE quantitative evaluation model. By the QoE score testing, the proposed QoE quantitative evaluation model achieves a high precision, which proves the reliability of this methodology.

Key words: mobile video, Quality of Experience(QoE), quantitative evaluation model, Analytic Hierarchy Process(AHP)