Anonymous location sharing in urban area mobility

GPS-equipped mobile devices are increasingly becoming an integral part of our daily lives. Consequently, the number of social networks that provide location based services (LBS) are increasing more and more. A typical service provided by these LBS's is the location check-in services. Location check-in via location-based service have brought privacy issues as each location check-in show the actual location in the geographic plane. This thesis studies the problem of privacy-preserving location update in the context of urban area people mobility. As a result, a data-centric framework is introduced. In the first section, an effective cloaking method has been developed to construct cloaking regions with respect to users' location privacy profile specification under road network. And a mobility model which utilizes these cloaking regions is proposed. Afterwards, related location privacy models have been proposed and algorithms providing location k-anonymity has been developed under road network and velocity-based attack constraints. In the second section, co-localization attacks has been detailed for co-location likely user groups and algorithms providing location k-anonymity has been developed under inference channels arising from user interactions. Within the scope of the thesis, location k-anonymity model is followed as a means to protect location privacy, i.e., the shared location (corresponding to a subgraph) is considered to be location privacy preserving if it contains at least k vertices. In addition, simulated datasets and trajectories are generated, comprehensive experimental studies that implements the proposed mobility model and algorithms has been carried out and their results are presented as well.