Impact of the Quality-Affecting Parameters on Multimedia Quality

A second important problem is the evaluation of the impact of many basic parameters on multimedia quality. The relationship between quality and these parameters is complex and difficult to apprehend. Multimedia quality is affected by many factors. However, there is no complete study in the literature that shows the impact of these parameters on the quality, mainly because subjective quality tests are expensive to carry out. Moreover, to analyze the combined effect, for instance, of three or four parameters, we need to build a very large set of human evaluations in order to reach a minimal precision level. Previous studies are either concentrated on the effect of network parameters without paying attention to encoding parameters, or the contrary. Papers that consider network parameters and use subjective tests for the evaluation restrict the study to only one or two of the most important ones. For example, in the case of speech, the loss rate and packetization interval metrics are studied in [61], while [28] studies mainly the effect of the loss rate on several speech codecs. In the case of video, loss rate and loss burst size metrics are studied in [58], while [153] studies mainly the effect of bit rate and [53] works only on the effect of frame rate. Concerning objective approaches, there is no previously published objective quality test that can take into account the direct influence on the quality of, for instance, the whole set of previously mentioned parameters simultaneously. To attack this problem, we need a tool that can measure multimedia quality by taking into account the direct influence of a large range variation of these parameters. The measured quality should correlate well with human perception. The tool should predict the quality within the range of each parameter's values without having to do the subjective tests for all their possible combinations. By understanding the impact of the parameters on the quality, this may help in: $\bullet$

• developing control mechanisms in order to deliver the best possible video quality given the current network situation (cf. the next subsection),
• and in better understanding of QoS aspects in multimedia engineering.

Our objective in this area is to provide studies of the combined effects of several quality-affecting parameters on both speech and video quality. We aim also to provide general guidelines that can be followed to perform similar studies in other kinds of network technologies and, if useful, other parameters.