Main Article Content

Review of multimedia communication quality assessment techniques


U. S. Ukommi

Abstract

Communication pattern is drifting from conventional audio to multimedia applications, streaming live and on-demand digital video contents over telecommunications and broadcasting networks. Currently, some businesses deploy video in promoting products in a far more enriching, entertainment and informative approach than typical traditional audio would allow. However, users of multimedia applications are interested in paying for a good acceptable video quality. Thus, this paper reviews methods used in Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Metric (SSIM) and Video Quality Metric (VQM) with the objective of providing useful information to wireless video services providers in assessing and monitoring of quality of multimedia services delivered to users. Experimental comparison of PSNR, SSIM and VQM assessment methods to examine their performances in evaluation of multimedia applications has been performed. Comparing the quality performance of PSNR, SSIM and VQM metrics for Akiyo and Crew standard test sequences, it has been observed that the quality performance of the metrics improves when the bitrates allocation increase. For the test of consistency and quality performance of the metrics, observation shows that for a given bitrates of 3.84Mb/s, under CABA test configuration, Akiyo test sequence, experienced better quality performance of 46.63dB (PSNR), 99.10% (SSIM) and 11.00% (VQM) compared to lower quality performance of 36.18dB (PSNR), 90.40% (SSIM) and 22.00% (VQM) experienced by Crew test sequence. Experimental results of the media quality metrics for Akiyo and Crew standard test sequences with different temporal activity levels show that the quality performance of media content also depends on the temporal activity of the media content. The experimental results further show consistency in quality performance of PSNR, SSIM and VQM metrics for CABA and UBA test configurations. Thus, the objective metrics can be harnessed for evaluation of quality performance of multimedia applications at different channel conditions.


Journal Identifiers


eISSN: 2467-8821
print ISSN: 0331-8443