A Subset Coding Based k-Anonymization Technique to Trade-Off Location Privacy and Data Integrity in Participatory Sensing Systems

Success of participatory sensing system depends on the extent of voluntary participation by users. To increase participation, incentive such as rewards can be used only if reported data has associated user identification. This creates serious threat to participating users' location privacy. Existing techniques tried to solve it with spatial clocking, which suffers from inferior data integrity. In this paper, we present a subset coding based anonymization scheme that can safeguard users' location privacy with k-anonymity while preserving almost lossless data integrity at the destination server. Adversary threats to our scheme are comprehensively analyzed to develop robust strategies and analytical bounds on system parameters for location privacy risk mitigation. Applicability of the proposed scheme is established with extensive simulation results.