Privacy Enhancement on Internet of Things Using Differential Privacy Strategies

Abstract

Abstract

With recent developments in communication and data storage technologies, the Internet is gathering and storing an explosive amount of information. While such a large volume of data provides great opportunities for the exploration of knowledge, organizations do not want to share their information for legal or competitive reasons. This raises the problem of information mining while retaining privacy. Current efficient data mining privacy preservation algorithms are focused on the assumption that all intermediate results during data mining operations are appropriate to be published. It has been shown, however, that such intermediate outcomes can still leak private information. Differential privacy is used in this situation to quantitatively restrict such leakage of information. Differential privacy is a newly emerged concept of privacy that can provide strong observable assurances of privacy. The rapid growth of the Internet of Things has contributed to the pervasive deployment of low-cost smart devices and the widespread use of high-speed wireless networks (IoT). IoT welcomes numerous physical artifacts that have not been active in the conventional Internet and allows a wide variety of IoT applications to be provided through their interaction and cooperation. Many IoT services which require a detailed understanding and analysis of data collected across a large number of physical devices that challenge both the privacy of personal information and IoT creation. IoT information privacy is a broad and complex term since its understanding and interpretation vary between individuals and involves efforts from both legislation and technology to implement it. In this paper we review IoT privacy enhancement using different privacy strategies on various IoT applications and challenges with Security & Privacy.

Keywords: Internet of Things, Differential Privacy, Security, Privacy, Data Mining, Data Set, Server.