Enhancing Intrusion detection: Leveraging Federated Learning and Hybrid Machine Learning Algorithms On ToN_IoT Dataset.

An efficient and productive IoT application may ease some real-time tasks however, it is at risk of cyber-attack. Intrusion Detection Systems (IDS) are of significant importance for the security measures of IoT applications. IoT/IIoT devices that deal with large data volumes are at risk of malicious attacks and as a result, anomaly-based IDS are developed. Anomaly-based intrusion detection systems perform more efficaciously than other methods. But, the question that arises is whether the performance of models meets the required standards. The research intends to improve the efficiency of IDS with a focus on the Telemetry data of IoT/IIoT sensors data from the ToN_IoT dataset. It includes data about seven IoT/IIoT devices. Federated Learning based on Deep auto-encoder is adopted to design a model with efficiently identifies attacks while solving the issue of privacy concerns. Hybrid models use Machine Learning algorithms with increased detection rates. The algorithms used for the Hybrid model are Random Forest and XGBoost. The XGBoost algorithm improves the accuracy of the Hybrid model with better predictions. The hybrid model ensures efficient pre-processing and feature selection for intrusion detection. The accuracy percentage of the Federated model is 97% on the dataset of Garage Door and 88% on the Motion Light dataset while the Hybrid model outperforms on all IoT/IIoT devices with an average accuracy score of 99.99%.