On the Effectiveness of Intrusion Detection Strategies for Wireless Sensor Networks: An Evolutionary Game Approach

摘要

In Wireless Sensor Networks a large number of tiny nodes cooperate to control or monitor critical infrastructures, industrial processes, environments and other real-time sensor-based applications. However, environmental and technological factors present in that networks increase considerably the likelihood of successful attacks due to the presence of malicious nodes. In this paper, an intrusion detection mechanism (IDS) to identify malicious nodes in wireless sensor networks it is proposed. Here the interactions between malicious and IDS nodes are modelled as an evolutionary game based on such that assumption that both players have limited rationality and bounded strategic reasoning. Analytical results demonstrate that it is possible to achieve the most effective detection strategy during the malicious-IDS game if IDS's population follows the evolutionarily stable strategy. Finally, simulation results show that setting adequate configuration parameters the devised model can effectively detect malicious nodes, while improving the resources usage at the same time.