A Mechanism for Greediness Management when Streaming Multimedia to Portable Devices

The majority of streaming solutions use rate adaptation based on congestion avoidance mechanisms that try to obtain as much bandwidth as possible from the limited network resources. This is good if we assume that all clients have equal requirements. However, this is rarely the case as devices have unique characteristics (e.g. screen size, screen resolution, location, etc.) that influence streaming related parameters and consequently the experience obtained by the end user. Due to the inherent greediness of streaming protocols and lack of knowledge about the characteristics of the users device, this may result in certain devices receiving higher bandwidth share than they actually need and others not receiving enough, affecting the users perceived quality. Therefore there is a need to allow for client differentiation in order to provide an acceptable service for all clients, taking their device characteristics into account. This paper proposes a greediness control algorithm (GCA), a specially designed application layer solution that tunes the greediness of the multimedia streaming based on client priority, to make more efficient use of the wireless network and increase the overall user perceived quality.