Providing Contextual Privacy in Sensor Networks (PARIS)

Abstract

The objective of this research is to protect contextual privacy in sensor networks.

Privacy threats that exist for sensor networks may be categorized into two broad classes: content-oriented threats and contextual threats. Content-oriented security and privacy threats are raised by an adversary observing/manipulating the exact content of packets, and due to its importance, many of the core problems related with content privacy are being addressed by the ongoing research in the security community. In this project, we focus on protecting contextual privacy associated with sensor communication, which is concerned with protecting the context associated with the measurement and transmission of sensed data. Among the wide spectrum of contextual information, we strive to hide the location of the message originator (source-location privacy) and the time when a packet is created (temporal privacy) from the adversary. To protect source- location privacy, we have proposed a novel routing scheme, called phantom routing, which first routes a packet to a random node in the network (i.e. a phantom source), and then delivers the packet to the destination along the original routing path. To protect temporal privacy, we introduce additional delivery delay by buffering packets at each intermediate forwarding node, and to address the buffer limit, we also design four preemption heuristics to make buffer vacancy for incoming packets when the buffer is full. Both of these two strategies carefully balance between privacy protection and energy conservation.

This project is one of the pioneer works in the field, and it has inspired a series of studies that look at contextual privacy issues for sensor networks.

Publications