SDN – based Dynamic Bandwidth Allocation for Multiple Video-on-Demand Players over HTTP

Abstract

Nowadays, HTTP adaptive streaming (HAS) has become the standard for video streaming over multimedia networks. However, HAS alone cannot ensure a seamless viewing experience. The adaptation of HAS in network management faces significant challenges due to purely client-driven approaches. These approaches create unfair competition for available bandwidth resources when multiple bitrate-adaptive players share the same bottleneck network link. Each HAS client independently maximizes its own bandwidth share, leading to competition for network resources among clients and thereby reducing the Quality of Experience (QoE) for end users. This competition negatively impacts key metrics such as fairness, efficiency, and stability for each video player. In this paper, we propose a scheme for bitrate adaptation in HAS systems combined with Software Defined Networking (SDN)-based dynamic resource allocation. The objective is to enhance QoE among competing clients. Experimental results demonstrate that our proposed method significantly outperforms conventional approaches across all key QoE parameters under various scenarios.