← Back to all news
arXiv Machine Learning By Konstantinos Gyftodimos, Kyriakos Chiotis, Elena Politi, George Dimitrakopoulos, Eirini Liotou

Temporal transformer CAN encoder with federated lightweight heads for anomaly detection

Read the original on arXiv Machine Learning →

The paper introduces a privacy‑preserving anomaly detection framework for in‑vehicle networks that uses a Temporal Transformer CAN Encoder with Federated Lightweight Heads. It captures subtle temporal and contextual irregularities in CAN bus traffic by learning signal evolution over time, while federated learning allows multiple vehicles or ECUs to collaboratively improve a shared model without sharing raw data. Experiments on open‑source datasets demonstrate robust detection performance with maintained efficiency and privacy.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Machine Learning
Aug 19

Digital Twin-Based Intrusion Detection for Vehicle Powertrain CAN Bus Systems

The paper presents a digital‑twin (DT) based intrusion detection system (IDS) for vehicle powertrain CAN bus traffic, modeling physical relationships among decoded signals to detect payload‑manipulation attacks that preserve normal timing and sequencing. Using a shared‑encoder LSTM trained on 17 Hyundai/Kia CAN signals, the DT flags anomalies when residuals exceed a threshold, achieving high detection rates (up to 94.6%) for stealthy attacks such as continuous drift and masquerade, while a range‑and‑plausibility baseline fails to detect them. The study demonstrates that learning coupled vehicle dynamics enables detection of payload‑level attacks that evade traditional timing‑based IDSs, though false positives remain a challenge.

By Araf Rahman, M Sabbir Salek, Mashrur Chowdhury
More like this →
arXiv AI
Jun 12

ASTER: Latent Pseudo-Anomaly Generation for Unsupervised Time-Series Anomaly Detection

arXiv:2604. 13924v3 Announce Type: replace-cross Abstract: Time-series anomaly detection (TSAD) is critical in domains such as industrial monitoring, healthcare, and cybersecurity, but it remains challenging due to rare and heterogeneous anomalies and the scarcity of labelled data.

By Romain Hermary, Samet Hicsonmez, Dan Pineau, Abd El Rahman Shabayek, Djamila Aouada
llmsragdiffusionbenchmarks
More like this →