A game-theoretic approach for optimizing intrusion detection strategy in WSNs

摘要

Due to the limited capabilities of sensor nodes in Wireless Sensor Networks (WSNs) in terms of computation, communication, and energy, selecting the profitable detection strategy for lowering resources consumption determines whether the IDS can be used practically. The signaling game is used to set up an intrusion detection game modeling the interactions between a malicious sensor node and an IDS agent, and its equilibriums are found for optimal detection strategy. The stage intrusion-detection game at each individual time slot is showed in aspects of its player's utilities and the mixed-strategy Bayesian Nash equilibrium BNE. As the game evolves, the stage intrusion detection game is developed into a multi-stage dynamic intrusion detection game in which the beliefs on malicious sensor node can be updated based on Bayesian rules. Depending on the current belief, the best response strategy for the IDS agent can be gained based on the Perfect Bayesian equilibrium (PBE). The simulation results have shown the effectiveness of the proposed games, thus, the IDS agents are able to select optimal strategy to defend the malicious sensor node's actions.