Research On Protection Of Query Privacy In Location Based Services

With the rapid development of mobile positioning technology and wireless networktechnology, the applications of location-based service have becoming more and morepopular, however, despite of the facilitate service of location-based services, it also bringsin worrying privacy issues. Privacy in location-based services includes not only thelocation privacy, but also query privacy, most of the current privacy protection model justfor location privacy, while the query request content may contains more privacy related tousers, therefore, researching new privacy protection method facing query context is ofgreat significance.Firstly, analyzing the features of the existing privacy protection model inlocation-based services and its shortage in query privacy protection, on base of theexisting LBS k-anonymous model, we analyze of the semantics of the sensitivity of querycontent and propose the LBS (p, k) anonymous model, in this model queries are dividedinto two types of sensitive queries and non-sensitive queries, ensure attackers can’t link asensitive query and a user at a proportion of more than p by constraining the proportion ofsensitive queries no more than p to achieve the query privacy protection.Secondly, analyzing the large space redundancy issue of the search algorithm for theanonymous space of LBS k-anonymity model, this paper proposes new search algorithmfor the anonymous space based on grid and density, first the space is mapped to m×n grid,then iteration search neighborhood space of the grid of the user lies in until finds theMinimum contain space, then stripping the edges with smallest user distribution densityone by one according to the density matrix, on purpose of finding the anonymous user setmeeting the anonymity condition in a minimum range, and achieving a better balancebetween privacy and quality of service. On base of it, add p-sensitive constraint in thealgorithm, propose search algorithm for the anonymous space of LBS (p, k)-anonymitymodel the algorithm within the query of model of anonymous space search algorithm.Finally, the proposed method is verified with Java language on the Windows platformfrom anonymous cost, anonymous quality, anonymous time, anonymous success rate.