Adaptive Cybersecurity for IoT and Edge Computing Devices using AI/ML

With the widespread use of IoT and edge computing devices, these systems have become primary targets for cyber-attacks. Traditional security mechanisms require significant computing resources, rendering them unsuitable for resource-constrained IoT and edge computing devices. Recent advancements in cybersecurity of IoT and edge devices focus on addressing the challenge of security efficacy despite highly limited computational resources. In the present study, a novel integration of machine learning based security protocols and real-time intrusion detection, specifically designed to address critical resource constraints of IoT and edge computing devices, has been proposed. The approach not only provide secure data handling and network resilience but also enable fail-safe operation of IoT and edge devices. Performance evaluation in a realistic simulation environment demonstrates its effectiveness in detecting and identifying anomalies as well as mitigating security threats such as Denial of Service (DoS) attacks, achieving an F1-score of 92.1% and ROC-AUC of 96.4%. The proposed distributed framework features a lightweight, self-contained IoT edge component alongside advanced secured analytics running on server resources, offering a scalable hybrid solution to protect even large-scale IoT and edge computing networks.