5th International ICST Conference on Testbeds and Research Infrastructures for the Development of Networks and Communities

Research Article

LA1 TestBed: Evaluation Testbed to Assess the Impact of Network Impairments on Video Quality

  • @INPROCEEDINGS{10.1109/TRIDENTCOM.2009.4976208,
        author={Mu Mu and Andreas Mauthe and John Casson and Gareth Tyson and Francisco Garcia},
        title={LA1 TestBed: Evaluation Testbed to Assess the Impact of Network Impairments on Video Quality},
        proceedings={5th International ICST Conference on Testbeds and  Research Infrastructures for the Development of Networks and Communities},
        publisher={IEEE},
        proceedings_a={TRIDENTCOM},
        year={2009},
        month={5},
        keywords={Testbed Network Impariments Video Quality},
        doi={10.1109/TRIDENTCOM.2009.4976208}
    }
    
  • Mu Mu
    Andreas Mauthe
    John Casson
    Gareth Tyson
    Francisco Garcia
    Year: 2009
    LA1 TestBed: Evaluation Testbed to Assess the Impact of Network Impairments on Video Quality
    TRIDENTCOM
    IEEE
    DOI: 10.1109/TRIDENTCOM.2009.4976208
Mu Mu1,*, Andreas Mauthe1,*, John Casson1,*, Gareth Tyson1,*, Francisco Garcia2,*
  • 1: Lancaster University Lancaster, United Kingdom
  • 2: Agilent Laboratories Edinburgh, United Kingdom
*Contact email: m.mu@comp.lancs.ac.uk, andreas@comp.lancs.ac.uk, jcasson@gmail.com, g.tyson@comp.lancs.ac.uk, frankie_garcia@agilent.com

Abstract

Currently, a complete system for analyzing the effect of packet loss on a viewer’s perception is not available. Given the popularity of digital video and the growing interest in live video streams where channel coding errors cannot be corrected, such a system would give great insight into the problem of video corruption through transmission errors and how they are perceived by the user. In this paper we introduce such a system, where digital video can be corrupted according to established loss patterns and the effect is measured automatically. The corrupted video is then used as input for user tests. Their results are analyzed and compared with the automatically generated. Within this paper we present the complete testing system that makes use of existing software as well as introducing new modules and extensions. With the current configuration the system can test packet loss in H.264 coded video streams and produce a statistical analysis detailing the results. The system is fully modular allowing for future developments such as other types of statistical analysis, different video measurements and new video codecs.