Idea
Lightweight AI model detecting CAN cyberattacks in real-time to secure connected vehicles and prevent safety risks.
Research Paper
Core Innovation
This paper introduces DAIRE, a lightweight artificial neural network architecture with layer sizes proportional to attack classes, optimized for real-time CAN attack detection. It balances high detection accuracy and low false positives with minimal computational requirements, outperforming existing models in inference speed and suitability for embedded vehicular environments.
Why It Matters
Connected vehicles rely on Controller Area Network communications that are vulnerable to cyberattacks, risking safety and operational integrity. DAIRE offers fast, accurate detection of these attacks with minimal computational overhead, enabling scalable real-time protection in automotive systems. This improves vehicle security and supports safer, more reliable Internet of Vehicles deployments.
Market Size (TAM)
$10–20B TAM for automotive cybersecurity; $2–5B SAM from connected vehicle manufacturers and fleet operators. Driven by increasing IoV adoption and rising cyberattack threats.
Potential Customers & Pain Points
- Automotive manufacturers – Need robust real-time cybersecurity for connected vehicles
- Fleet operators – Require efficient detection of vehicle network attacks to avoid downtime
- IoV platform providers – Demand scalable security solutions with low latency
- Automotive cybersecurity firms – Seek advanced detection models to enhance product offerings.
Business Model
Licensing the DAIRE AI model as an embedded software module to automotive OEMs, fleet management companies, and IoV platform providers, with options for subscription-based updates and support services.
Competitive Landscape
- Argus Cyber Security
- Karamba Security
- Upstream Security
- GuardKnox
Implementation Challenges
- Integration with diverse vehicle CAN architectures
- Regulatory compliance and certification requirements
- Adoption resistance due to legacy system constraints
Validation Strategy
- Pilot deployments with automotive manufacturers for real-world CAN network testing
- Collaboration with cybersecurity firms for independent performance benchmarking
- Field trials in connected vehicle fleets to measure operational impact and reliability
Research Paper Overview
DAIRE: A lightweight AI model for real-time detection of Controller Area Network attacks in the Internet of Vehicles
Summary
DAIRE is a lightweight machine learning framework designed for real-time detection and classification of cyberattacks on Controller Area Network communications in the Internet of Vehicles. It achieves high accuracy and low false positive rates while significantly reducing computational demands, enabling practical deployment in vehicular systems to enhance automotive cybersecurity.