Game Theoretic Channel Selection for Opportunistic Spectrum Access with Unknown Prior Information

We investigate the problem of distributed channel selection in an opportunistic spectrum access (OSA) system, from a game theoretic perspective. We consider a practical scenario where the channel availability statistics and the number of secondary users are unknown. We formulate the problem of distributed channel selection as a repeated game and prove it to be an exact potential game. Since there is no prior information about the channel availability statistics and there is no information exchange between secondary users, then existing approaches are not feasible in our game. We then propose a learning automata based distributed channel selection algorithm. The proposed learning algorithm (1) achieves the Nash equilibrium (NE) of the channel selection game, (2) does not explicitly learn the channel availability statistics and the number of secondary users, and (3) does not need information exchange between secondary users.