Research On Secure And Efficient Double Spectrum Auction Mechanisms

In recent years,with the rapid development and widespread popularity of wireless technology,the demand for spectrum has increased dramatically.However,most of the licensed spectrum which has been allocated under the traditional,static and rigid spectrum allocation polices remains unused or low utilization in many places much of the time.In order to improve spectrum utilization,secondary spectrum markets have arisen,where auctions are used to dynamically and efficiently redistribute spectrum.The auction here can greatly improve the spectrum utilization by exploiting spatial reusability of spectrum.In such an auction,the licensed spectrum users are encouraged to lease their selected idle channels for profit to secondary spectrum users who are starving for wireless spectrum.The first necessary property for a practical auction is truthfulness which simulating each bidder to report its true valuation while guarding against dishonest behaviors.Unfortunately,the true valuation which may reflect its economic status is privacy for secondary spectrum user(usually the wireless service provider),on the other hand,the true valuations leaked by open and market-based truthful auction may be used by a malicious auctioneer or malicious bidders to undermine the truthfulness of auction.The above security issues seriously hamper the practicability of auction.Moreover,existing privacy-preserving works in traditional auction cannot be transplanted to the spectrum auction directly for the spectrum can be reused among bidders simultaneously.Therefore,it is necessary and significant to design secure and efficient auction according to the characteristics of the spectrum auction mechanism for promoting the implementation of spectrum auction and improving spectrum utilization effectively.In this thesis,after investigating the existing spectrum auction mechanisms and the privacy protection spectrum auction schemes,we focus on the double spectrum auction that is more efficient and complex than single-sided auction while may put forward more challenges to the realization of the privacy protection.The major achievements of this thesis are concluded as follows.(1)Aiming at the privacy-preserving in double spectrum auction,a secure,efficient and practical double spectrum auction is designed.Specifically,by combining three security techniques:homomorphic encryption,secret sharing and garbled circuits,we design a secure tow-party protocol computing a socially efficient double spectrum auction,TDSA,without leaking any information about sellers' requests or buyers' bids beyond the auction outcome.Theoretical analysis and security proof are given to prove the security that the proposed protocol achieves.Experimental results show that our design is efficient and practical even for large-scale double spectrum auctions(thousands of bidders).(2)Based on existing buyer grouping algorithms,some optimization measures are proposed after the contrast and analysis of the existing algorithms.The buyer grouping algorithm,which is mainly used to form buyer groups in double spectrum auction so that one channel can be reused within one group,is equivalent to finding the independent sets of the conflict graph.As we know,the problem of maximum independent set has been proved to be an NPC-hard problem.Therefore,three algorithms that find the independent set,which is not necessarily the maximum independent according to nodes'degree,are explored and the analysis of the characteristic of independent set found by different algorithms are made.Although the buyer grouping algorithm is bid-independent,different buyer grouping methods have different influence on the result of the same auction.Considering the fairness demand of auction,two buyer grouping algorithms are proposed and compared with the above algorithms.Experimental results show that one of the proposed algorithms can improve the fairness of auction under the premise of guaranteeing the auction efficiency and the spectrum utilization.