Location Privacy Preserving Dynamic Spectrum Auction in Cognitive Radio Network

Dynamic spectrum auction offers the flexibility and capability for bidders to request and acquire unoccupied channels from spectrum license holders. Compared with the conventional auction, spectrum auction allows various buyers to utilize the same channel simultaneously based on their locations, which is denoted as spectrum reusability. In this paper, we consider a novel kind of attack, which could compromise location privacy of bidders by observing the bid items as well as bid price. To thwart this attack, we introduce a new Location Privacy Preserving Dynamic Spectrum Auction (LPPA) scheme which consists of two components: Privacy Preserving Bid Submission protocol (PPBS) and Private Spectrum Distribution protocol (PSD). Based on the prefix membership verification scheme, PPBS allows the auctioneer to construct the conflict relationship between different users and obtain the maximum value of bids on various channels without leaking users' location information. Furthermore, PSD is proposed to efficiently distribute the spectrum among bidders and securely charge the winners with the help of periodically available TTP (Trusted Third Party). To demonstrate the effectivenss of the proposed scheme, we implement our attack and scheme on data extracted from Google Earth Coverage Maps released by FCC. The experiment results show the efficacy and efficiency of our approach.