FedPrIDS: Privacy-Preserving Federated Learning for Collaborative Network Intrusion Detection in IoT

Abstract

With the rapid proliferation of Internet of Things (IoT) devices, network security has become increasingly critical. This paper introduces FedPrIDS (Federated Privacy-Preserving Intrusion Detection System), a cutting-edge framework that leverages federated learning to enable collaborative threat detection across distributed IoT networks while preserving data privacy.

Key Contributions

• Privacy-Preserving Architecture: Implements advanced federated learning techniques that allow multiple IoT networks to collaboratively train intrusion detection models without sharing raw data
• Enhanced Security: Provides robust protection against various network attacks while maintaining the confidentiality of sensitive network traffic data
• Scalable Solution: Designed to handle the massive scale and heterogeneity of modern IoT deployments
• Real-World Applicability: Demonstrates practical implementation in IoT environments with minimal performance overhead

Research Impact

This work addresses the critical challenge of balancing security and privacy in IoT ecosystems. By enabling collaborative learning without centralized data collection, FedPrIDS paves the way for more secure and privacy-conscious IoT deployments across industries including smart cities, healthcare, and industrial automation.

Authors

• Sameer Mankotia - Department of Computer Science, University of Idaho
• Daniel Conte de Leon* - Department of Computer Science, University of Idaho (Corresponding Author)
• Bhaskar P. Rimal - Department of Computer Science, University of Idaho

*Corresponding author

Read the full paper on MDPI

Recommended citation: Mankotia, Sameer, Conte de Leon, Daniel*, and Rimal, Bhaskar P. "FedPrIDS: Privacy-Preserving Federated Learning for Collaborative Network Intrusion Detection in IoT." Department of Computer Science, University of Idaho, Moscow, ID 83844, USA.
Download Paper