Federated Learning for Distributed IoT Security: A Privacy-Preserving Approach to Intrusion Detection

Billions of resource-constrained IoT devices now stream security-critical yet privacy-sensitive traffic across the Internet. To detect intrusions without exposing raw data, we propose a federated learning (FL) framework that trains collaboratively on the edge while preserving privacy. At its core is...

Full description

Saved in:
Bibliographic Details
Main Authors: Chandu Gutti, Karthik Thumula, Parag Balbudhe
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11095679/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849239083028578304
author Chandu Gutti
Karthik Thumula
Parag Balbudhe
author_facet Chandu Gutti
Karthik Thumula
Parag Balbudhe
author_sort Chandu Gutti
collection DOAJ
description Billions of resource-constrained IoT devices now stream security-critical yet privacy-sensitive traffic across the Internet. To detect intrusions without exposing raw data, we propose a federated learning (FL) framework that trains collaboratively on the edge while preserving privacy. At its core is Hybrid Adaptive-Weight Aggregation (HADA), a new server-side rule that weights client updates by their SHAP-based feature stability and individual differential-privacy (DP) budgets. HADA down-weights noisy or low-privacy updates, speeding convergence and hardening the global model against poisoning. Evaluated on the CIC-BCCC-NRC TabularIoTAttack-2024 and Edge-IIoTset benchmarks with 100 simulated ARM-class devices, the proposed FL-IDS achieves 85&#x2013;89 % detection accuracy&#x2014;on par with centralized training&#x2014;while reducing per-round uplink traffic by an order of magnitude. Under strong adversarial perturbations (FGSM, PGD-10, label-flip) it retains 66&#x2013;73 % accuracy, outperforming vanilla FedAvg by up to 22 percentage points. A formal convergence theorem is proved for non-IID, DP-noisy updates, and the privacy&#x2013;utility curve shows <inline-formula> <tex-math notation="LaTeX">$\lt \!1.5$ </tex-math></inline-formula> pp accuracy loss at <inline-formula> <tex-math notation="LaTeX">$\varepsilon = 1.0$ </tex-math></inline-formula>. These results demonstrate that HADA-FL delivers a scalable, privacy-preserving, and attack-resilient intrusion-detection system suitable for real-world heterogeneous IoT deployments.
format Article
id doaj-art-2da0ea0d3ab14c18baf328fe593b6442
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-2da0ea0d3ab14c18baf328fe593b64422025-08-20T04:01:15ZengIEEEIEEE Access2169-35362025-01-011313586313587510.1109/ACCESS.2025.359248111095679Federated Learning for Distributed IoT Security: A Privacy-Preserving Approach to Intrusion DetectionChandu Gutti0https://orcid.org/0009-0001-5968-5898Karthik Thumula1https://orcid.org/0009-0001-2690-8901Parag Balbudhe2https://orcid.org/0009-0005-7594-5767Independent Researcher, San Jose, CA, USAIndependent Researcher, San Jose, CA, USAIndependent Researcher, Boston, MA, USABillions of resource-constrained IoT devices now stream security-critical yet privacy-sensitive traffic across the Internet. To detect intrusions without exposing raw data, we propose a federated learning (FL) framework that trains collaboratively on the edge while preserving privacy. At its core is Hybrid Adaptive-Weight Aggregation (HADA), a new server-side rule that weights client updates by their SHAP-based feature stability and individual differential-privacy (DP) budgets. HADA down-weights noisy or low-privacy updates, speeding convergence and hardening the global model against poisoning. Evaluated on the CIC-BCCC-NRC TabularIoTAttack-2024 and Edge-IIoTset benchmarks with 100 simulated ARM-class devices, the proposed FL-IDS achieves 85&#x2013;89 % detection accuracy&#x2014;on par with centralized training&#x2014;while reducing per-round uplink traffic by an order of magnitude. Under strong adversarial perturbations (FGSM, PGD-10, label-flip) it retains 66&#x2013;73 % accuracy, outperforming vanilla FedAvg by up to 22 percentage points. A formal convergence theorem is proved for non-IID, DP-noisy updates, and the privacy&#x2013;utility curve shows <inline-formula> <tex-math notation="LaTeX">$\lt \!1.5$ </tex-math></inline-formula> pp accuracy loss at <inline-formula> <tex-math notation="LaTeX">$\varepsilon = 1.0$ </tex-math></inline-formula>. These results demonstrate that HADA-FL delivers a scalable, privacy-preserving, and attack-resilient intrusion-detection system suitable for real-world heterogeneous IoT deployments.https://ieeexplore.ieee.org/document/11095679/Federated learningIoT securityintrusion detection systemprivacy-preservingdistributed learningcybersecurity
spellingShingle Chandu Gutti
Karthik Thumula
Parag Balbudhe
Federated Learning for Distributed IoT Security: A Privacy-Preserving Approach to Intrusion Detection
IEEE Access
Federated learning
IoT security
intrusion detection system
privacy-preserving
distributed learning
cybersecurity
title Federated Learning for Distributed IoT Security: A Privacy-Preserving Approach to Intrusion Detection
title_full Federated Learning for Distributed IoT Security: A Privacy-Preserving Approach to Intrusion Detection
title_fullStr Federated Learning for Distributed IoT Security: A Privacy-Preserving Approach to Intrusion Detection
title_full_unstemmed Federated Learning for Distributed IoT Security: A Privacy-Preserving Approach to Intrusion Detection
title_short Federated Learning for Distributed IoT Security: A Privacy-Preserving Approach to Intrusion Detection
title_sort federated learning for distributed iot security a privacy preserving approach to intrusion detection
topic Federated learning
IoT security
intrusion detection system
privacy-preserving
distributed learning
cybersecurity
url https://ieeexplore.ieee.org/document/11095679/
work_keys_str_mv AT chandugutti federatedlearningfordistributediotsecurityaprivacypreservingapproachtointrusiondetection
AT karthikthumula federatedlearningfordistributediotsecurityaprivacypreservingapproachtointrusiondetection
AT paragbalbudhe federatedlearningfordistributediotsecurityaprivacypreservingapproachtointrusiondetection