On the Use of Fake Sources for Source Location Privacy: Trade-Offs Between Energy and Privacy

Over the last decade the advent of wireless sensor networks has enabled several novel classes of application, including monitoring and tracking. In the case of monitoring applications, the deployment of sensor networks will vary from mission critical applications, such as military, health and asset monitoring, to non-critical applications, such as temperature and humidity control. For some critical applications, such as military surveillance or asset monitoring, security is an important requirement. One important aspect of security is privacy. Privacy is often important in critical applications, as sensitive information may need to be kept private. However, owing to the fact that wireless sensor networks operate in a broadcast medium, attackers can easily intercept messages and attempt to subvert a system. The privacy threats that exist for sensor networks can be broadly classified along two dimensions, namely (i) content-based privacy threats and (ii) contextbased privacy threats. Content-based privacy threats relate to threats that are based on the contents of messages, i.e., the threats are against the values generated by the various network layers, either at the application level, e.g., sensed values, or lower-layer levels, e.g., time-stamps. Context-based privacy threats are based on the context associated with the measurement and transmission of sensed data. Context is a multi-attribute concept that captures several environmental aspects associated with sensed data, including aspects such as location and time. While content-based threats are well-understood [16], context-based threats are becoming increasingly important. For content-based threats, nodes launching attacks are often modelled as Byzantine nodes [9], with cryptographic techniques often being used to address these problems [9] [2]. However, cryptographic techniques do not address context-based threats. One aspect of context that is important in several applications is location. For example, in a military situation, location can be that of a soldier. In asset monitoring, the location can be that of a very expensive painting in a museum. Location information can be embedded in a message, but still remain inacces-