Rate adaptation performance and quality analysis of adaptive HTTP streaming methods

Abstract

Media streaming is the key innovation in technology era brings high quality on-demand and live media content to the consumer on-the-go anywhere, any time. Achieving best quality-of-service and quality-of-experience (QoE) are major concerns in the current circumstances due to increasing network usage and user demand to high quality media content. Conventional streaming approaches face numerous challenges in delivering media content to the uses without degradation of quality. An adaptive HTTP streaming is emerging content delivery approach which provides real time content delivery without compromising quality and ensuring superlative QoE. The dynamic and robust bit rate selection based on the network and client key performance indices helps to achieve best class of service to the consumer. As many standards and processes emerge in the field of Adaptive HTTP Streaming, it is necessary to study and compare the performance of different approaches based on the defined metrics. In this paper, we emphasis on the comprehensive study of different emerging adaptive streaming models such as HTTP Live Streaming, Microsoft Smooth Streaming and MPEG Dynamic Adaptive Streaming over HTTP (DASH). The experiments were conducted using Gstreamer adaptive HTTP streaming client to measure delivery performance of the approaches under varying network conditions for on-demand streaming and live streaming contents. The outcome for adaptive HTTP methods were measured and analyzed using defined performance indices. The analyzed data indicates that each delivery method performs well with respect to outlined metrics. However, overall DASH gives considerably stable performance across different network settings than other streaming approaches.