Attack Prevention Scheme for Privacy Preservation (APSP) Using K Anonymity in Location Based Services for IoT

Abstract

Internet of Things (IoT) is tagging our day-to-day usable objects across the internet. To make the life easy one should be more connective and communicative with the objects which are usable in daily life. In IoT, the privacy of users may be sacrificed while enjoying the conveniences and flexibilities provided by it. All the queries generated by the users, contain some relevant information about their location and identity. Location Based Server may uses this information maliciously and also reveal to third parties. In this paper, a new scheme, named APSP, has been proposed to preserve privacy of user’s location and to make it difficult for the attacker to identify the real user’s location. APSP implements an attack prevention algorithm on DLP for a given degree of anonymity. It selects number of dummy locations to ensure the maximum entropy with minimum variance. Simulation results show that, in APSP, there is least probability to reveal the identification of real user’s location under attack and achieves better privacy preservation for selected dummy locations. APSP maximize the complexity for attacker by providing the optimal set of locations. The attacker gets confused with the same historical query probabilities of all selected locations and fails to identify the real user’s location.