cs.CRFeb 20, 2026

PenTiDef: Decentralized Federated Intrusion Detection System with Differential Privacy and Latent-Space Defense via Blockchain Coordination in IIoT

Authors: Phan The Duy, Nghi Hoang Khoa, Nguyen Tran Anh Quan, Luong Ha Tien, Ngo Duc Hoang Son, Van-Hau Pham

Organizations: Information Security Lab, University of Information Technology, Ho Chi Minh City, Vietnam. · Vietnam National University, Ho Chi Minh City, Vietnam. · VNU-HCM Information Security Center, Ho Chi Minh City, Vietnam.

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

This paper proposes PenTiDef, a fully decentralized, privacy-preserving, and poisoning-resilient framework for decentralized federated IDS (DFL-IDS). PenTiDef synergistically integrates three key components: (i) client-side Distributed Differential Privacy (DDP) with stochastic Gaussian noise to protect gradient leakage, (ii) a lightweight latent-space defense module that extracts and compresses penultimate-layer representations (PLRs) into stable Latent Semantic Representations (LSRs) via AutoEncoder, followed by Centered Kernel Alignment (CKA) and K-Means clustering for robust malicious update detection without auxiliary datasets, and (iii) a permissioned blockchain layer with smart contracts that orchestrates on-chain validation, secure FedAvg aggregation, and immutable auditability, eliminating any central server. Extensive experiments on CIC-IDS2018 and Edge-IIoTSet under both IID and realistic non-IID settings, with adversary ratios up to 40%, demonstrate that PenTiDef consistently outperforms state-of-the-art baselines (FLARE and FedCC) in detection accuracy and F1-score while maintaining lower training overhead. By jointly addressing privacy, robustness, and decentralization in a unified secure aggregation protocol, PenTiDef provides a practical and scalable solution for trustworthy collaborative intrusion detection in heterogeneous, adversarial IIoT environments.

Figures & tables

DP typeTrusted aggregatorAdding noise byPrivacy Guarantee
CDPYesAggregatorAggregated value
LDPNoUserLocally released value
DDPNoUserAggregated value
Table 1: Comparison between DP categorizations
Figure 1: The PenTiDef architecture for privacy-preserving and anti-poisoning attack mechanisms in DFL-based IDS
Figure 2: Details of layers in CNN
Figure 3: Data distribution of the Edge-IIoTSet dataset under IID and non-IID scenarios
Figure 4: Data distribution of the CIC-IDS2018 dataset under IID and non-IID scenarios
Figure 5: Comparison of training results with and without applying DDP
Figure 6: Similarity comparison between the latent space of the global model and each local model’s latent space using CKA scores in PenTiDef and FedCC.
Figure 7: Compare the similarity between the latent space of the global model and each local model’s latent space using CKA scores in PenTiDef and FedCC.
DatasetModelTraining Time
CIC-IDS2018Flare3,907s ∼\sim 01h 06m
FedCC2,460s ∼\sim 41m
PenTiDef1,988s ∼\sim 33m
Edge-IIoTsetFlare4,395s ∼\sim 1h 15m
FedCC2,825s ∼\sim 47m
PenTiDef1,869s ∼\sim 31m
Table 7: Average training time on different datasets with different models
More in this topic

Blockchain-Based Federated Learning

See all 9 papers

Explore similar work

CardsList

  1. PVBRF-ID: Private, Verifiable and Byzantine-Robust Federated Intrusion Detection for Oilfield Enterprise Networks

    Oct 8, 2026Mustafa S. Aljumaily, Ali Ataeemh Allami, Mustafa N. Mnati +2Privacy-Preserving Federated LearningFederated Learning

    Daw Alfada Company, Amarah, Maysan, Iraq · University of Misan, Amarah, Maysan, Iraq · University of Southern Queensland, Toowoomba, QLD, Australia

  2. Privacy, Robustness, and Fairness Trade-offs in Federated Intrusion Detection: Geometric Indistinguishability at the Aggregation Interface

    Sep 3, 2026Adrita Rahman Tory, ABM Shawkat Ali, Md Abu Layek +1Class-Imbalanced LearningFederated Learning Aggregation

    Bangladesh University of Business and Technology (BUBT), Mirpur-2, Dhaka-1216, Bangladesh · Jagannath University, Dhaka, Bangladesh · University of New South Wales (UNSW), ACT 2601, Australia

  3. FBID: Adaptive Personalized Federated Learning for Robust Out-of-Distribution Attack Detection in IoT Networks

    Aug 4, 2026An Khanh Bui, Cong Thanh Nguyen, Hoang-Anh Pham +2Network Intrusion DetectionInternet of Things

    UTS-HCMUT JTIRC, Ho Chi Minh City University of Technology (HCMUT), Ho Chi Minh City 700000, Vietnam. · Vietnam National University Ho Chi Minh City (VNU-HCM), Ho Chi Minh City 700000, Vietnam. · School of Electrical and Data Engineering, University of Technology Sydney, Sydney, NSW 2007, Australia.