In a cognitive radio network, collusion among selfish users may have seriously deleterious effects on the efficiency of dynamic spectrum sharing. The network users' behaviors and dynamics need to be taken into consideration for efficient and robust spectrum allocation. In this chapter, we model spectrum allocation in wireless networks with multiple selfish legacy spectrum holders and unlicensed users as multi-stage dynamic games. In order to combat user collusion, we present a pricing-based collusion-resistant approach for dynamic spectrum allocation to optimize overall spectrum efficiency, while not only keeping the incentives to participate of the selfish users but also combating possible user collusion. The simulation results show that the scheme achieves a high efficiency of spectrum usage even with severe user collusion.

Introduction

Traditional network-wide spectrum assignment is carried out by a central server, namely a spectrum broker. Distributed spectrum allocation approaches that enable efficient spectrum sharing solely on the basis of local observations have recently been studied. From the economic point of view, the deregulation of spectrum use further encourages market mechanisms for implementing efficient spectrum allocation.

Because of the spectrum dynamics and lack of centralized authority, the spectrum allocation needs to distributively adapt to the dynamics of wireless networks due to node mobility, channel variations or varying wireless traffic on the basis of local observed information. From the game-theoretic point of view, first of all, the spectrum allocation needs to be studied in a multi-stage dynamic game framework instead of the static game approach.