"FogSec_FedECLF_Lite_2026"

"Fog computing has evolved a lot in the last few years, since there is a growing need for real-time and scalable applications for the Internet of Things (IoT). The expansion of fog computing in IoT applications is accompanied by so many issues that need effective solutions to respond to those challenges, especially fog computing security. By employing a federated ensemble learning technique that enables several nodes to train a shared model without disclosing their local data, this work seeks to enhance security in fog computing. This helps to preserve privacy while still enabling the system to scale and retain reasonable communication overhead. Instead of presenting its contributions as discrete concepts, the paper builds them gradually to address the primary issues in fog-cloud situations. It presents a three-layer fog computing security architecture designed to connect security initiatives across various system components. Then, the paper presents an Ensemble Learning Approach (ENLA) that combines several classifiers to improve detection accuracy. However, the specific integration method is still unclear. Next, Fed-ECLF-Lite, a lightweight federated ensemble learning model, was created especially for heterogeneous fog nodes, which are typically limited by communication costs. All of these components are then combined in the study to create a deployable version of the Fed-ECLF-Lite security model, which is then evaluated at scale on various devices and datasets. Accuracy, latency, general scalability, and privacy-preserving collaboration between nodes have all significantly improved as a result of the approach."