PROJECT TITLE :

Detecting Malicious Data Injections in Event Detection Wireless Sensor Networks

ABSTRACT:

Wireless sensor networks (WSNs) are vulnerable and will be maliciously compromised, either physically or remotely, with doubtless devastating effects. When sensor networks are used to detect the prevalence of events such as fires, intruders, or heart attacks, malicious data will be injected to create pretend events, and therefore trigger an undesired response, or to mask the prevalence of actual events. We tend to propose a unique algorithm to spot malicious knowledge injections and build measurement estimates that are resistant to several compromised sensors even after they collude in the attack. We additionally propose a methodology to apply this algorithm in different application contexts and evaluate its results on three totally different datasets drawn from distinct WSN deployments. This leads us to spot completely different tradeoffs in the design of such algorithms and how they are influenced by the applying context.