Events & Webcasts

Deploying Machine-Learning based Anomaly Detection For Embedded Mixed-Criticality Systems: a Feedback

Embedded AI Presentation in Paris

Presenter: Marine Kadar

Phd Researcher at SYSGO


Abstract

Original Equipment Manufacturers now embed hardware virtualization in critical equipments such as aircrafts or automotive platforms to reduce costs and hardware complexity, while allowing more functionalities, such as connectivity. This evolution forces the cohabitation of distinct criticality domains on the same hardware, reaffirming the need for security. Such embedded platforms supporting different criticality levels are called embedded mixed-criticality sytems (MCS).

Because of the trade-off between performance and system overall complexity, deploying security becomes a challenging balancing act. Host Intrusion Detection Systems (HIDS) security protects the behavior of a program at runtime: it monitors the program execution flow to distinguish threats from benign activity. Many HIDS solutions leverage Machine-Learning to model normal and anomalous program executions, notably using system calls or hardware performance counters traces.

In this presentation, we propose to evaluate the deployability of machine-learning based HIDS into an embedded MCS. For this, we develop an online monitoring framework on an industrial embedded MCS platform to detect anomalies tracing hardware performance counters and [...] Read the complete Abstract 


Register here

  • Next

    Integrating Hardware-Assisted Anomaly Detection in Safety-Critical Systems Read more