A Blockchain-Integrated Federated Learning Model and Autoencoder-Based Feature Reduction for Improving IoT Intrusion Detection

Abstract

The rapid growth of Internet of Things (IoT) environments has brought forth a wealth of security challenges in detecting network intrusions in diverse and resource-restricted systems. Privacy, scalability, and single point of failure issues plague traditional centralized intrusion detection solutions. To address these challenges, the study proposes a secure and adaptive intrusion detection model using Federated Learning (FL) and Blockchain, augmented with autoencoder-based feature reduction. The ToN-IoT dataset is pre-processed, and then an unsupervised autoencoder is used to build informative low-dimensional feature representations. The processed data is deployed to various clients to mimic a real federated situation. Every client will train a local Long Short-Term Memory (LSTM) model on its own private data to preserve data privacy.Then a blockchain-based mechanism is utilized to enhance the security and integrity of model aggregation. SHA-256 hashed local model weights are recorded on the blockchain to avoid tampering and provide traceability. Federated averaging is then implemented to refresh the global model along with blockchain-based verification of the aggregation process. Our findings show the performance of the proposed framework, leading to an accuracy of 99.96%, precision of 99.99%, recall of 99.95%, and F1-score of 99.97%. These findings show that combining FL, blockchain, and deep feature extraction offers a viable and secure solution for intrusion detection systems in IoT.