Bidding Privacy Preservation for Dynamic Matching Based Spectrum Trading

Abstract

Spectrum trading benefits secondary users (SUs) with accessing opportunities and primary users (PUs) with monetary gains in cognitive radio networks. In the spectrum trading market, bidding privacy is a serious concern for SUs, since untrustworthy PUs or spectrum traders may leverage bidding information to manipulate the trading or leak SU's bids to other SU competitors for illegal gains. Current centralized privacy preserving spectrum trading designs may incur extra infrastructure deployment, miss many instantaneous spectrum accessing opportunities, and have scalability issues. In this paper, we propose a novel privacy-preserving semi-distributed spectrum trading scheme, which has joint consideration of spectrum reuse, SUs' bidding privacy preservation, and PUs' revenue maximization. We use the conflict graph to characterize SUs' interference relationships, and jointly employ the Paillier Cryptosystem and dynamic matching with evolving preferences to conduct the privacy-preserving spectrum trading in a semi-distributed manner. Through security, complexity and performance analysis, we show that the proposed scheme can effectively preserve the privacy of SUs' bidding values, and notably increase the PUs' revenue and improve spectrum utilization with limited overhead.