Novel machine learning-based game theoretic approach for wireless sensor networks

The capabilities and effectiveness of wireless sensor network (WSN) technologies are driving their rapid growth. Due to their unattended nature, WSNs are easily exploited by an attacker. This study addresses the issue of WSN security by introducing a machine learning (ML) based game theoretic approach (ML-GT). In this paper, the game theory is applied for routing to optimize energy consumption while using a security layer. And, the ML method called Kernel Adaptive Support Vector Machine (KASVM) is performed to detect the malicious nodes in WSN. We also conduct a simulation experiment to assess the ML-GT method's efficacy and provide a thorough evaluation of the simulation's outcomes. The proposed strategy can enable security with optimal end-to-end communication in WSN