Packet Scheduling for Maximizing Revenue in a Network Node

In the future Internet, different applications such as Voice over IP (VoIP) and Video-on-Demand (VoD) arise with different demands on Quality of Service (QoS). Different kinds of service classes (e.g. gold, silver, bronze) should be supported in a network node. In the network node, packets are queued using a multiqueue system, where each queue corresponds to one service class. The customers of different classes will pay different prices to network providers based on multi-class pricing models. In this paper, we considered the optimization problem of maximizing the revenue attained in a network node under linear pricing scenario. A revenue-aware scheduling approach is introduced, which has the closed-form solution to the optimal weights for revenue maximization derived from revenue target function by Lagrangian optimization approach. The simulations demonstrate the revenue maximization ability of our approach.