ECU Identification using Neural Network Classification and Hyperparameter Tuning

TL;DR

This paper presents a neural network-based fingerprint intrusion detection system for CAN protocol that achieves high accuracy in identifying ECUs through spectral and step response features, enhanced by feature reduction and hyperparameter tuning.

Contribution

It introduces a modified Fingerprint IDS utilizing spectral and step response features with hyperparameter tuning for improved ECU identification accuracy.

Findings

Achieved 99.4% detection rate of trusted ECU traffic.

Utilized feature set reduction to enhance model performance.

Applied hyperparameter tuning to optimize neural network classification.

Abstract

Intrusion detection for Controller Area Network (CAN) protocol requires modern methods in order to compete with other electrical architectures. Fingerprint Intrusion Detection Systems (IDS) provide a promising new approach to solve this problem. By characterizing network traffic from known ECUs, hazardous messages can be discriminated. In this article, a modified version of Fingerprint IDS is employed utilizing both step response and spectral characterization of network traffic via neural network training. With the addition of feature set reduction and hyperparameter tuning, this method accomplishes a 99.4% detection rate of trusted ECU traffic.