Maintaining the Balance between Privacy and Data Integrity in Internet of Things

The recent proliferation of human-carried mobile and smartphone devices has opened up opportunities of using crowd-sensing to collect sensory data in Internet of Things (IoT). As tapping into the sensory data and resources of the smartphones becomes common place, it is necessary to ensure the privacy of the device user while maintaining the accuracy and the integrity of the data collected. IoT system devices often sacrifice either user privacy or data integrity. It has also become important to limit the computational cost and burden on the user devices, as increasingly more services desire to tap into the resource that these devices provide. In this paper we propose a balanced truth discovery (BTD) framework that attempts to meet all three of the aforementioned needs: user privacy, data integrity, and limited computational cost. The BTD framework also reduces user participation in the truth discovery process. The nature of the BTD framework provides the possibility for easy modification (e.g. cryptography and weight assignment). This reduces computation cost for the user device, but also limits the interactions between the devices and the server, which is essential to data integrity. BTD framework also takes steps to blur the user device's original sensory data, by processing results in groups called zones. An enhanced method takes privacy preservation a step further, by protecting the user from an untrusted data-collecting party. Analysis of simulations running the framework provides evidence for the preservation of data integrity.