Research On Privacy Preservation Of Location Data

With the pervasively existence of the mobile Internet and the development of the positioning technologies,the location information of people as well as the devices became more significant in various of scenarios.Today,people use location to exchange the Location-Based Services(LBS)in many kinds of situations;the location is also be considered as a critical query condition or the result in the spatial searching.On the other hand,in order to get the location,the location services are widely adopted during the procedure of the positioning.A widespread location privacy concerns exist in the lifecycle of both the positioning and the application of location data.Both the provider of the third-party positioning service and the LBS could observe and collect the location data while they providing the the service.On one hand,the user has no choice but send their location to the LBS provider in order to get the LBS result,on the other hand,more and more positioning behavior of the end device are utilizing the “third party” information service,this leads to the same location privacy threat than LBS.Finally,as the rapid deployment of the mobile Internet,a diversity of LBS becomes prosperous,and the corresponding threat model of traditional location privacy is also kept up with it.On the basis of the above-mentioned issues,in this paper,for critical aspects of location privacy threats are considered,following with the corresponding location privacy preserving mechanisms.The main contribution of this paper is as follow:First,to overcome the location privacy threat in third-party positioning service,a noise addition-based k-anonymity method is proposed to preserve the location privacy.With the applicability of the mainstream positioning technologies,the paper presents the analysis and the modeling of the spatial distribution of wireless access points,and the principle of “positionable” relationship is presented.Based on this model,a noise addition-based method is proposed to generate the noise data.Buy adding noise data into the real access points,the method ensures the indistinguishability of the real data and the noise data,and hence preserve the location privacy.Second,a velocity-based threat is proposed in our paper.In trajectory LBS scenario,the abnormal detection threat against the generalized trajectory data is pointed out,such threat could erode those generalized data without considering the velocity information of the trajectory.Then a velocity-aware location privacy preserving method is proposed for resisting such threat,by generating the trajectory data with high similarity to the real trajectory in velocity,the proposed method outperforms the existing method while against the threat.Third,in order to solve the isolation of location privacy preservation between the positioning and LBS scenario,an access point-point of interest mapping nearest neighbor(NN)query method is proposed.The method performs NN query by directly using the access point indicators,without the location information.Then,an R-tree based multi-search structure is proposed for the privacy-aware NN query.The detach from the location data enhance the location privacy in NN query scenario.Finally,for an emerging type of LBS and the corresponding location privacy threat,the definition of the device-dependent LBS(DLBS)is proposed to formally define the scenario and the location privacy threat,which is hard to be solved by the traditional methods.Then,a virtual currency-based provider behavior constraint mechanism is proposed.By binding the location privacy and the DLBS system functionality,the method forces the provider to get the uncessary location information only by sacrificing his service profit.The proposed method effectively constrains the excessive collection of location data,and hence preserved location privacy in DLBS.