NESL Technical Report #: 2011-6-1
Abstract: With the wide-spread use of mobile smartphones and body- worn sensors, continuous collection of sensor data about individuals becomes feasible, and many useful applications such as medical behavioral studies, personal health-care, and participatory sensing have emerged. Such applications have important privacy implications due to their nature of shar- ing personal sensor data. In addition, what is shared is not only the raw sensor data but also the information that can be inferred from the data, which raises more privacy concerns of users. This paper proposes SensorSafe, an ar- chitecture for managing such personal sensory information in a privacy-preserving way. Our architecture consists of multiple remote data stores and a broker so users can re- tain the ownership of their data and management of mul- tiple users can be well supported. SensorSafe also provides a fine-grained access control mechanism by which users can define their own sharing rules based on various conditions including context and behavioral status. Users define their privacy preferences and review their data by using our web- based user interface. We discuss our implementation of the SensorSafe architecture and provide application examples to show how our system can support user privacy. Our perfor- mance evaluation results demonstrate that building appli- cations using the SensorSafe architecture is feasible so user privacy can be better protected.
Publication Forum: The 8th VLDB Workshop on Secure Data Management
Public Document?: Yes
NESL Document?: Yes
Document category: Conference Paper