Anomaly detection in an embedded system

Woo, Lai Leng, Zwolinski, Mark and Halak, Basel (2021) Anomaly detection in an embedded system. In, Hardware Supply Chain Security: Threat Modelling, Emerging Attacks and Countermeasures. Springer, pp. 167-211. (doi:10.1007/978-3-030-62707-2_6).

Record type: Book Section

Abstract

Embedded systems, especially those that are mission-critical or safety-critical, require a higher level of dependability. Error detection is first step and a vital aspect in fault tolerance because a processor cannot tolerate a problem that it is not aware of. Even if the processor cannot recover from a detected fault, it can still alert the user that an error has occurred and halt. Thus, error detection provides, at the minimum, a measure of safety. Online error detection is the ability to detect any form of violation of system specifications during runtime. One of the techniques that has been applied for online error detection is anomaly detection. This section will discuss the techniques for anomaly detection and a case study on using a single hardware performance counter for early detection and prediction of failure.

This record has no associated files available for download.

More information

Published date: 4 February 2021

Additional Information: Publisher Copyright: © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021.

Keywords: Anomaly detection, Fault tolerance, Hardware performance counter, Online detection, Prediction, Statistical forecasting, Time-series