PROMISE: A framework for truthful and profit maximizing spectrum double auctions

Auctions provide a platform for licensed spectrum users to trade their underutilized spectrum with unlicensed users. Existing spectrum auctions either do not apply to the scenarios where multiple sellers and buyers both make offers, or assume the knowledge of the users´ valuation distribution for maximizing the profit of the auction. To fill this void, we design PROMISE, a framework for spectrum double auctions, which jointly considers spectrum reusability, truthfulness, and profit maximization without the distribution knowledge. We propose a novel technique, called cross extraction, to compute the bid representing a group of secondary users, who can share a common channel. We prove that PROMISE is computationally efficient, individual-rational, and truthful. In addition, PROMISE is guaranteed to achieve an approximate profit of the optimal auction.