On the Impact of Burst Loss for QoE-Based Performance Evaluations for Video Streaming

In the broad area of performance evaluation of multimedia networking mechanisms, quality of experience (QoE) assessments are often used as a performance metric. Even though intensive efforts have been made to understand the impact of packet loss on QoE for different application scenarios, there is no consensus about how specific packet loss patterns impact the QoE for RTP video streaming. In this paper, we conduct an extensive study on the impact of uniformly distributed random and burst loss on the QoE-metrics Peak-Signal-to-Noise-Ratio (PSNR), Structural Similarity (SSIM), and Video Multi-Method Assessment Fusion (VMAF). Overall, our study indicates that for QoE-based performance evaluations, it is in most cases sufficient to only consider easier to understand and easier to model uniformly distributed random loss.