FedSecureFormer: A Fast, Federated and Secure Transformer Framework for Lightweight Intrusion Detection in Connected and Autonomous Vehicles

TL;DR

This paper introduces FedSecureFormer, a lightweight, federated transformer framework designed for fast and secure intrusion detection in connected and autonomous vehicles, emphasizing minimal layers for efficiency.

Contribution

It proposes a novel encoder-only transformer architecture optimized for federated learning in vehicle security applications, focusing on speed and security.

Findings

Achieves high detection accuracy with minimal model complexity

Ensures data privacy through federated learning approach

Demonstrates effectiveness in real-world vehicle scenarios

Abstract

This works presents an encoder-only transformer built with minimum layers for intrusion detection in the domain of Connected and Autonomous Vehicles using Federated Learning.