Abstract

Federated learning can be a promising solution for enabling IoT cybersecurity (i.e., anomaly detec- tion in the IoT environment) while preserving data privacy and mitigating the high communication/s- torage overhead (e.g., high-frequency data from time-series sensors) of centralized over-the-cloud approaches. In this paper, to further push forward this direction with a comprehensive study in both algorithm and system design, we build FedIoT platform that contains a synthesized dataset us- ing N-BaIoT, FedDetect algorithm, and a sys- tem design for IoT devices. Furthermore, the proposed FedDetect learning framework im- proves the performance by utilizing an adaptive optimizer (e.g., Adam) and a cross-round learn- ing rate scheduler. In a network of realistic IoT devices (Raspberry PI), we evaluate FedIoT plat- form and FedDetect algorithm in both model and system performance. Our results demonstrate the efficacy of federated learning in detecting a large range of attack types. The system efficiency analysis indicates that both end-to-end training time and memory cost are affordable and promis- ing for resource-constrained IoT devices. The source code is publicly available.