In large-scale wireless sensor networks, sensors are vulnerable to false report and false vote injection attacks since they are deployed in hostile environments. These attacks drain their limited energy resources of forwarding nodes and drops important data. Probabilistic voting-based filtering scheme simultaneously detects both the attacks through vote verification. To effectively detect them, it is important to define the vote length of the reports since the vote length is fixed at the initial phase. We find the effective vote length using a simulation model since it is nearly impossible to evaluate the security protocol performance on the real nodes. We demonstrate that the security protocol, in which the vote length is five, achieves better detection ratio against the two attacks.
A probabilistic voting-based filtering scheme in wireless sensor networks
