TY - GEN
T1 - AnonySense
T2 - 6th International Conference on Pervasive Computing, Pervasive 2008
AU - Kapadia, Apu
AU - Triandopoulos, Nikos
AU - Cornelius, Cory
AU - Peebles, Daniel
AU - Kotz, David
PY - 2008
Y1 - 2008
N2 - Opportunistic sensing allows applications to "task" mobile devices to measure context in a target region. For example, one could leverage sensor-equipped vehicles to measure traffic or pollution levels on a particular street, or users' mobile phones to locate (Bluetooth-enabled) objects in their neighborhood. In most proposed applications, context reports include the time and location of the event, putting the privacy of users at increased risk-even if a report has been anonymized, the accompanying time and location can reveal sufficient information to deanonymize the user whose device sent the report. We propose AnonySense, a general-purpose architecture for leveraging users' mobile devices for measuring context, while maintaining the privacy of the users.AnonySense features multiple layers of privacy protection-a framework for nodes to receive tasks anonymously, a novel blurring mechanism based on tessellation and clustering to protect users' privacy against the system while reporting context, and k-anonymous report aggregation to improve the users' privacy against applications receiving the context. We outline the architecture and security properties of AnonySense, and focus on evaluating our tessellation and clustering algorithm against real mobility traces.
AB - Opportunistic sensing allows applications to "task" mobile devices to measure context in a target region. For example, one could leverage sensor-equipped vehicles to measure traffic or pollution levels on a particular street, or users' mobile phones to locate (Bluetooth-enabled) objects in their neighborhood. In most proposed applications, context reports include the time and location of the event, putting the privacy of users at increased risk-even if a report has been anonymized, the accompanying time and location can reveal sufficient information to deanonymize the user whose device sent the report. We propose AnonySense, a general-purpose architecture for leveraging users' mobile devices for measuring context, while maintaining the privacy of the users.AnonySense features multiple layers of privacy protection-a framework for nodes to receive tasks anonymously, a novel blurring mechanism based on tessellation and clustering to protect users' privacy against the system while reporting context, and k-anonymous report aggregation to improve the users' privacy against applications receiving the context. We outline the architecture and security properties of AnonySense, and focus on evaluating our tessellation and clustering algorithm against real mobility traces.
UR - http://www.scopus.com/inward/record.url?scp=44649150419&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=44649150419&partnerID=8YFLogxK
U2 - 10.1007/978-3-540-79576-6_17
DO - 10.1007/978-3-540-79576-6_17
M3 - Conference contribution
AN - SCOPUS:44649150419
SN - 3540795758
SN - 9783540795759
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 280
EP - 297
BT - Pervasive Computing - 6th International Conference, Pervasive 2008, Proceedings
Y2 - 19 May 2008 through 22 May 2008
ER -