Multi-user rate-based flow control: distributed game-theoretic algorithms

Flow and congestion control allow the users of a telecommunication network to regulate the traffic they send into the network according to the quality of service that they require. The flow control may be performed by the network, as is the case in ATM networks, or by the users themselves, as is the case in the Internet (TCP/IP). We consider in this paper both cases using optimal control techniques. For the first case, we obtain a formulation of a dynamic team problem. The second case is handled by dynamic non-cooperative game techniques. We establish the existence and uniqueness of a Nash equilibrium, and compute the corresponding performance measures and (dynamic) equilibrium policies. We further show that when the users update their policies in a greedy manner, not knowing a priori the utilities of the other players, their policies converge to the Nash equilibrium.