Intelligent Flow Control Under Game Theoretic Framework

In this chapter, we organize our work in two parts. The first part deals with the connection admission control (CAC) problem in ATM networks. By taking dynamic CAC from a different point of view, we propose a cooperative game model that considers certain combination of user preferences to optimize the call block probability while maintaining certain negotiated QoS. After deriving the product form of utility function of different traffic type as user preferences, we use genetic algorithm to optimize this objective function to maintain fair share during the transmission along each traffic connection. The second part tries to model the noncooperative user behavior in resource allocation process under a generalized auction framework. We propose an optimal auction rule that leads to the equilibrium in repeated dynamic game. We also use neural networks to model the user strategy in bidding and briefly discuss the formation of common knowledge in repeated game and the allocation efficiency. Contrary to our CAC scheme as a centralized network control mechanism, the auction to allocate network resources is suitable to a decentralized control environment. The common trait for these two parts is that we use some computational intelligence techniques to solve the complicated modeling and optimization problem in resource allocation