This project addresses the challenge of strengthening control over location privacy for users of wireless devices such as smartphones. As these devices and their network services continuously monitor our environment, they enable many novel applications with tremendous societal benefits. However, they also raise significant privacy challenges by making it difficult for users to control when information about their whereabouts can be sensed or revealed. To address this challenge, this project studies the hitherto relatively unexplored concept of incorporating a comprehensive set of de-identification techniques into clients, which limit device-specific information that could allow extended tracking and eventually identification of the device?s user.
Project results are expected to lead to novel models that relate parameters such as spatio-temporal precision and accuracy, sampling frequency, and the presence of pseudo-identifiers to tracking and identification risks. These models complement existing models for transactional database records and can be used to inform users about their current level of privacy or guide system designers. In addition, the project is expected to provide fundamental insights on physical layer techniques that limit the accuracy with which infrastructure location sensors can locate a transmitting client, and techniques that can automatically detect candidate pseudoidentifiers in transmitted messages. The project also strengthens Rutgers University?s electrical and computer engineering curriculum by incorporating privacy and wireless mobile system topics. It further includes industry collaboration and an outreach plan in collaboration with New Jersey's Liberty Science Center to attract high school students from underrepresented groups to the computer engineering profession.
