TSA: A framework of truthful spectrum auctions under the physical interference model

Auction is an effective method of allocating scarce spectrum resources in cognitive radio networks, where the primary users are sellers and the secondary users are buyers. In order for the buyers and sellers to act honestly during the auction, truthfulness has been identified as an important property. Current research focuses on the truthfulness and spatial reusability by either assuming that a conflict graph is given under the protocol model, or assuming that the grouping result is given under the physical interference model without power control. To fill this void, we design a framework of truthful double auctions, named TSA, for spectrum sharing in cognitive radio networks. TSA finds a feasible grouping profile such that users in the same group can be assigned to the same channel while each gets a satisfactory SINR value by an appropriate transmitting power allocation. We prove that TSA guarantees all the desired economic properties: individual rationality, budget-balance, computational efficiency, and truthfulness. Extensive performance evaluation also supports our theoretic analysis.