Dynamic QoE optimisation for streaming content in large-scale future networks

Multimedia content sharing services, such as YouTube, usually offer only low quality video streams. Additionally, it is expected the popularity of these services will increase even further in the future. Therefore, current streaming- content delivery architectures, with centralised searching and management components, might not be able to keep up with the ever-growing user bases. In this paper, we propose a scalable and fully distributed end- to-end architecture for delivering streaming multimedia content. Additionally, we designed several algorithms that allow content servers to intelligently adapt content quality to the available access-link bandwidth and server resources. In this way, the service can improve quality of experience, by offering higher bit-rates at times of low load, while still being able to satisfy as many content requests as possible when the load increases. In depth simulations were performed to validate and evaluate these algorithms. The results show that the proposed platform is capable of increasing delivered stream quality and accepted content requests in various large-scale scenarios.