End-to-End Monitoring of Multidimensional User-Level QoS in Audio-Video IP Transmission

This paper proposes a method of multidimensional user-level QoS monitoring for audio-video transmission over IP networks in the context of ITU-T Recommendation J.148. In order to assess user-level QoS in a multidimensional way, we utilize the SD (semantic differential) method, which is one of the psychometric methods. By comparing results of the multidimensional assessment with overall subjective quality, which is represented by a scalar value, we find major factors which contribute to the user-level QoS. Moreover, in order to clarify implications of the major factors, we perform QoS mapping between user-level and application-level. The mapping functions can estimate the user-level QoS and therefore enables its real-time monitoring. From experimental results, we find that three factors contributes to the overall subjective quality; we then see that the three factors imply audio quality affected by video, video quality affected by audio, and the content type.