Federated intrusion detection protects oilfield networks from attacks
PVBRF-ID: Private, Verifiable and Byzantine-Robust Federated Intrusion Detection for Oilfield Enterprise Networks
Cryptography and SecurityMachine Learning
Summary
Oilfield enterprise networks need to detect cyberattacks without sharing sensitive data across many sites. The authors introduce PVBRF-ID, a method where sites share secret pieces of data to check for malicious updates without seeing the full details. Their approach is both private and robust against attackers who try to poison the system. Tests show it performs well even when some sites act maliciously, unlike traditional methods that either leak data or are easily fooled.
What this means in practice
- •For oilfield network operators: Securely train shared intrusion detectors without exposing individual site data or risking poisoned updates.
- •For industrial cybersecurity teams: Protect multi-site industrial networks by verifying updates in federated learning without sacrificing privacy or robustness.
Authors
Mustafa S. Aljumaily, Ali Ataeemh Allami, Mustafa N. Mnati, Nawar S. Alseelawi, Shahab Abdulla
Abstract
Federated learning allows dispersed oilfield sites to train a shared intrusion detector without pooling network traffic, yet its aggregation step forces a choice between privacy and robustness. Byzantine-robust rules must read each update in the clear, whereas secure aggregation hides updates from the checks that detect poisoning. We present PVBRF-ID, a two-server protocol in which sites secret-share their updates and the servers jointly test each update's norm and its cosine similarity to a reference computed on a small clean root set, revealing one accept bit per site. Commitments, Merkle roots and a homomorphic audit make the inclusion of accepted updates publicly verifiable. On a 15-class Edge-IIoTset task with ten sites holding differently distributed data and ten random seeds, PVBRF-ID matched FedAvg on clean data (macro-F1 0.708 vs. 0.706) and withstood sign-flip attacks (0.696 vs. 0.091). Against an adaptive attacker controlling 30% of sites, PVBRF-ID with secure clipping reached 0.641, whereas FedAvg fell to 0.046; plaintext multi-Krum remained higher by 0.052. An update-inference attacker recovered each site's class mix from plaintext updates (Spearman correlation 0.81) but not from one server's view. Backdoors passed the filter, and noise-based mitigation gave only heuristic protection. The implementation uses Python 3.13, NumPy 2.3, 30 rounds and 10 seeds.