Study On Privacy Protection Scheme Based On Clustering In LBS

With the rapid development of mobile communications technology and positioning technology,location-based service(Location-based service,LBS)has been widely used in the worldwide.Especially in recent years,a variety of commercial LBS products constantly emerging,enrich the user's daily life and entertainment but also promote the development of the peripheral industry.And it already has an important role in People's Daily.However,with LBS giving users convenience,the privacy leakage problem also gradually appears.Especially once the problem of location privacy leakage cannot be well resolved;it will be a great threat to the user's life and property safety.Researchers have proposed a variety of related location privacy protection method so far,to protect the user's location privacy.One of the most studied approaches is location k-anonymity.While traditional location k-anonymity can provide some protection for the user's location privacy,it has some drawbacks.Such as when the users in cloaking region gather to a certain point,the specific location information of the user will still be locked.In order to solve the problem,in the protection of user's location privacy,we should not only consider the location k-anonymity but also need to consider location l-diversity,i.e.the cloaking region need to include l different physical location.The study of this aspect is less,so this paper proposed a clustering-based privacy protection scheme,which can satisfy the location k-anonymity and location l-diversity,and demonstrate the superiority of the scheme through experiments,and the specific contents are as follows:(1)Summarizes the existing location privacy protection technologies and system architecture,and highlights the location k-anonymity and location l-diversity.(2)Proposes a clustering-based privacy protection scheme,which can satisfy the location k-anonymity and location l-diversity.The algorithms of this scheme contains BAUC(Based-on analysis using clustering-algorithm)and(Based-on discrete grid cells).BAUC uses the density-based clustering algorithm to analyze the historical user's location data firstly,and uses B~+-tree to index the results of analysis,then generates cloaking region meeting the location k-anonymity and location l-diversity using fragmented cloaking region.Because it's necessary to obtain the true historical user's location data of the service area before executing BAUC algorithm,the scheme also includes BDGC anonymity algorithm.BDGC algorithm is used to provide location anonymity services for users and accumulate user location data in the initial phase of the scheme.This algorithm is based on grids algorithm to improve.In addition the scheme also includes the methods of anonymity-based query corresponding to the proposed anonymity algorithms.(3)The anonymity algorithms in the proposed scheme are implemented,and the experimental results are analyzed.The results show that proposed algorithms have the advantage on multiple evaluation indexes,and then prove the superiority of the proposed scheme in this paper.