Adaptive resource negotiation based control for real time applications

In this paper, we focus on providing quality-of-service (QoS) control in terms of packet loss and delay for interactive real-time applications. We propose a new scheme to find the optimal values of token bucket parameters, token generation rate r and token bucket size b, from the observed real-time traffic. This can be adjusted based on the user's QoS requirement and the different classes of real-time applications. Based on these optimal token bucket parameters, we introduce an adaptive resource negotiation control scheme. Our proposed admission control scheme offers the re-negotiation feature in the resource reservation process allowing higher admission ratios and higher resource utilization to be achieved

Duke Scholars

Author Teh Yu Tan Thomas Lord Department of Mechanical Engineering and Materia ...