Privacy-Aware Community Sensing Using Randomized Response

Community sensing is an emerging system which allows the increasing number of mobile phone users to share effectively minute statistical information collected by themselves. This system relies on participants' active contribution including intentional input data through mobile phone's applications, e.g. Facebook, Twitter and Linkdin. However, a number of privacy concerns will hinder the spread of community sensing applications. It is difficult for resource-constrained mobile phones to rely on complicated encryption scheme. We should prepare a privacy-preserving community sensing scheme with less computational-complexity. Moreover, an environment that is reassuring for participants to conduct community sensing is strongly required because the quality of the statistical data is depending on general users' active contribution. In this article, we suggest a privacy-preserving community sensing scheme for human-centric data such as profile information by using the combination of negative surveys and randomized response techniques. By using our method described in this paper, the server can reconstruct the probability distributions of the original distributions of sensed values without violating the privacy of users. Especially, we can protect sensitive information from malicious tracking attacks. We evaluated how this scheme can preserve the privacy while keeping the integrity of aggregated information.