Fair pricing policy for multipoint-to-point service in Internet wide-area network

Price and quality differentiation are valuable tools that can provide higher revenues and increase utilization efficiency of a network, and thus in general social welfare. A well-designed pricing scheme should also have the fairness property, and how to define fairness within a multicast group, and among multicast groups and unicast connections is a crucial concern in the fairness property. In this paper, we propose a new resource allocation scheme that is based on power spectral analysis and general weight max-min (GWM) fairness to compensate for QoS degradation services. We prove that the QoS degradations of the input traffic, particularly delay and delay variations, are closely related to the power spectrum and derive an efficient discount factor based on perturbation analysis. The GWM fairness in the multipoint-to-point ABR service is also evaluated to avoid and alleviate congestion. Simulation results are presented to demonstrate the efficiency of proposed schemes.