Audiovisual network service optimization by quality of experience estimation

With the growing popularity of audio and video communication services on the Internet, network operators, service providers and application developers are becoming increasingly interested in assuring that their services give the best possible experience to the users. Since real-time audio and video services are very sensitive to packet loss, latency and bandwidth variations, the performance of the network must be monitored in real time so that the service can be adapted to varying network conditions by mechanisms such as rate control, forward error correction and jitter buffer adaptation. However, in order to optimize a service in terms of the user’s experience, the subjective effect that various network perturbations have on the user should be taken into consideration in the service adaptation mechanism. In this paper we present a novel approach to performance optimization based on rate adaptation driven by real-time estimation of the subjective Quality of Experience of a videoconferencing service. A proof-of-concept service optimization framework consisting of network monitoring, quality estimation, rate adaptation and service optimization mechanisms is presented and a testbed configuration based on network emulation is described and used for evaluation. Our initial experiments show that the approach is viable in practice and can substantially improve the Quality of Experience of real-time audio-