A825-HWIL: Hardware simulation platform for ARINC 825 cybersecurity analysis
Bibliographic record
Abstract
CAN, which stands for Controller Area Network, is an ubiquitous link-layer communication protocol in the automotive, aerospace and manufacturing industry.Invented in the mid-1980s, the CAN protocol has since kept a dominant position for real-time systems thanks to the simplicity and the low cost of its implementation.However, it was created before cybersecurity was a concern, and has many vulnerabilities that could be maliciously exploited to extract passenger data or to sabotage control systems.With cars being easily accessible, researchers have been able to demonstrate and validate numerous attack vectors against CAN in the automotive field.In the aerospace industry, the Aeronautical Radio, Incorporated 825 (ARINC 825) standard proposed a communication protocol that relies on CAN as the link layer.Airborne systems which follow the ARINC 825 standard are thus also exposed to the vulnerabilities of the underlying CAN protocol, but research in aerospace cybersecurity has been a lot less common due to the lack of access to link-layer data.Current data collection strategies are mostly related to application-level data and cannot replicate the required link-layer-level behaviour for such research.This work presents A825-HWIL, a modular and fully hardware-in-the-loop simulation platform that allows researchers to collect and analyze realistic ARINC-encoded CAN data on a physical CAN bus.Contrary to its software-based predecessor, ARINC825-TBv2, the link-layer simulation in A825-HWIL is completely hardware-based, which results in a better representation of the simulated system.This work essentially constitutes an evolution towards a more realistic test bench platform.While A825-HWIL is able to achieve great timing accuracy, it is also an excellent validation tool for showing the effectiveness of Gaslighter, a predictive attacker, and for demonstrating the applicability in real-time of two time-based intrusion detection systems, Delta-T and Z-Score.I cannot start this thesis without thanking my supervisor, Professor Brett H. Meyer, for your continued support and guidance.You have been a great source of motivation throughout this major undertaking, and I would not have been able to complete this journey without your dedication and your enthusiasm.Our relationship was one of respect and learning, and I would like to express my deepest gratitude for your contribution to my growth not only as a student, but also as a person.I am also thankful to the other members of the Reliable Silicon Systems Lab, both former and current.Jarul, Derek, Loren; your valuable insights were always appreciated.I enjoyed working alongside you during my first year, and I thank you for your help and feedback during my second.Next, I would like to ackowledge McGill Formula Electric, where I learned perseverance and resilience, and where I obtained the technical knowledge that inspired me to shape A825-HWIL into what is presented in this thesis.Enfin, merci maman, merci papa, et merci Alexandra, de m'avoir aiguillé sur le bon chemin et d'avoir eu confiance en moi.Vous m'avez poussé à me surpasser et vous avez toujours su trouver les bons mots, aux bons moments, pour me guider vers la réussite.Je tiens donc à vous remercier du fond du coeur pour votre présence et votre soutien, et je vous en serai éternellement reconnaissant.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.016 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".