Research On Price Privacy Protection Technology In Spectrum Allocation Of Cognitive Wireless Network

With the rapid development of the Internet,the coverage of wireless network is getting larger and larger.At the same time,the fierce international competition for 5G technology makes the research in wireless field all over the world.So far,a variety of wireless application devices have appeared in the market,and the communication of these wireless devices will use a non-renewable resource,namely spectrum resources.In recent years,with the fast development of radio communication technology,the demand for radio spectrum resources in the world is growing rapidly.However,the international allocation of spectrum is still static,which makes the spectrum utilization extremely low.As an effective means of buying and selling,auction mechanism is often used in the transaction of scarce commodities to make both buyers and sellers profitable by bidding.Therefore,this thesis focuses on the use of auction mechanism in spectrum allocation of cognitive wireless networks,and studies and implements a spectrum auction mechanism based on differential privacy.In view of the shortage of wireless spectrum resources and the single spectrum allocation strategy,a dynamic spectrum allocation scheme based on differential privacy is proposed.The idle spectrum owned by primary users is managed centrally,and then the auctioneer allocates the idle spectrum to the secondary users with communication needs by means of spectrum auction.Through this dynamic spectrum allocation scheme,the static spectrum allocation can be effectively utilized.Considering different aims of spectrum administrator,In this thesis,we analyze the spectrum auction process in single primary user multiple user scenarios and multi-primary user multiple user scenarios respectively.Based on integer programming technology,we study the winner selection strategy of auctioneers to improve the spectrum reusability,meanwhile maximize the benefits of spectrum auctions or minimize the spectrum efficiency under unit bidding price..In order to solve the problem of malicious users seeking the bidding price of other users to win the auction,the design of the auction scheme needs to satisfy the honesty and fairness.This thesis designs an honest auction mechanism to ensure that users can participate in spectrum auction based on real valuation and obtain maximum revenue while ensuring the fairness of auction.According to the problem of bidding price leakage in auction process,this thesis considers a method of bidding price privacy protection based on differential privacy theory,and protects users' bidding price privacy by adding random noise to winner selection strategy.Through the simulation test on MATLAB R2017 a platform,the results show that the auction mechanism designed in this thesis can not only protect the privacy of users' bidding price,but also ensure the effectiveness of the scheme.