Malicious node identification strategy based on artificial bee colony considering reputation in WSN

Abstract:

In the complex application environment of wireless sensor networks (WSN), in order to resist the selective forwarding attack and dishonest recommendation attack launched by malicious nodes and improve the safety performance of the network, this paper proposes a malicious node identification strategy based on artificial bee colony (ABC) considering reputation (CR-ABC) in WSN. By utilizing a fuzzy trust model (FTM) and integrating a fuzzy comprehensive evaluation mechanism, the paper calculates the comprehensive trust value of nodes based on three influencing factors: communication features, data attributes, and physical attributes to improve the detection accuracy of the reputation model. The paper introduces the suggested deviation function and the interaction index deviation function and uses the ABC algorithm to optimize the FTM, aiming to ensure that the system still maintains a higher identification rate and a lower misjudgment rate when there are too many malicious nodes. The simulation results show that the identification rate of CR-ABC for selective forwarding attacks can reach over 90%, and the misjudgment rate for normal nodes can be reduced to less than 6%. For dishonest recommendation attacks, even if the number of dishonest nodes reaches 50%, CR-ABC still maintains a high identification performance, which can effectively improve the security and reliability of WSN in complex environments.