Hybrid 1D CNN-BiLSTM with focal loss for robust intrusion detection in in-vehicle CAN networks
A Malini, Gerardine Immaculate Mary · 2026
Read Hybrid 1D CNN-BiLSTM with focal loss for robust intrusion detection in in-vehicle CAN networks by A Malini, Gerardine Immaculate Mary (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0353472.txt.
Description
A. Malini and Gerardine Immaculate Mary build a guard for the internal network of a modern car, in this 2026 PLOS ONE study. Cars coordinate brakes, steering, and engine through a message bus that was designed without security, leaving it open to injected or replayed messages. Their detector combines two neural network types: one reads the shape of each message, the other follows sequences over time in both directions.
A training trick called focal loss forces the model to concentrate on rare, hard-to-spot attacks instead of drowning in normal traffic. Tested on a public car-hacking dataset, the system scored high on accuracy, precision, and recall alike. Such onboard watchdogs will matter as cars grow more connected.
Catalog record
| CREATOR | A Malini, Gerardine Immaculate Mary |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 70034 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |