Private Pooling: A Privacy-Preserving Approach for Mobile Collaborative Sensing

Due to the emergence of embedded sensors in many mobile devices, mobile and people-centric sensing has become an interesting research field. A major aspect in this field is that quality and reliability of measurements highly depend on the device’s position and sensing context. A sound level measurement, for instance, delivers highly differing values whether sensed from inside a pocket or while carried in a user’s hand. Mobile collaborative sensing approaches try to overcome this problem by integrating several mobile devices as information sources in order to increase sensing accuracy. However, sharing data with other devices for collaborative sensing in return raises privacy concerns. By exchanging sensed values and context events, users might give away sensitive data, which should not be linkable to them. In this paper, we present a new mobile collaborative sensing protocol, , which protects the users’ privacy by decoupling the data from its contributors in order to allow for anonymous aggregation of sensing information.