A general framework for adaptive filtering in wireless sensor networks based on fuzzy logic

Wireless sensor networks are vulnerable to false data injection attacks. In these, adversaries inject false reports into the network using compromised nodes, with the goal of deceiving the base station and depleting the energy resources of forwarding nodes. Several filtering schemes, and the adaptive versions of them, have been proposed to detect and drop such false reports during the forwarding process. In this paper, we propose a fuzzy-based framework to achieve adaptive filtering of false reports. A fuzzy rule-based system controls the security level of the filtering scheme through the determination of the required number of endorsements in every report. Compared to existing adaptive solutions, the proposed method uses more practical factors for the determination and can be applied to the network, regardless of scale. The proposed method can provide energy saving and sufficient resilience against false data injection attacks.