Citation: Sternharz, G.; Skackauskas,
J.; Elhalwagy, A.; Grichnik, A.J.;
Kalganova, T.; Huda, M.N.
Self-Protected Virtual Sensor
Network for Microcontroller Fault
Detection. Sensors 2022, 22, 454.
https://doi.org/10.3390/s22020454
Academic Editor: Nunzio Cennamo
Received: 29 November 2021
Accepted: 29 December 2021
Published: 7 January 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Self-Protected Virtual Sensor Network for Microcontroller
Fault Detection
German Sternharz
1
, Jonas Skackauskas
1
, Ayman Elhalwagy
1,
*, Anthony J. Grichnik
2
, Tatiana Kalganova
1
and Md Nazmul Huda
1
1
Department of Electronic and Electrical Engineering, College of Engineering, Design and Physical Sciences,
Brunel University London, Uxbridge UB8 3PH, UK; german.sternharz@brunel.ac.uk (G.S.);
jonas.skackauskas2@brunel.ac.uk (J.S.); tatiana.kalganova@brunel.ac.uk (T.K.);
mdnazmul.huda@brunel.ac.uk (M.N.H.)
2
Blue Roof Labs, 759 N Main St., Eureka, IL 61530-1035, USA; Tony.Grichnik@BlueRoofLabs.com
* Correspondence: ayman.elhalwagy@brunel.ac.uk
Abstract:
This paper introduces a procedure to compare the functional behaviour of individual units
of electronic hardware of the same type. The primary use case for this method is to estimate the
functional integrity of an unknown device unit based on the behaviour of a known and proven
reference unit. This method is based on the so-called virtual sensor network (VSN) approach, where
the output quantity of a physical sensor measurement is replicated by a virtual model output. In the
present study, this approach is extended to model the functional behaviour of electronic hardware by
a neural network (NN) with Long-Short-Term-Memory (LSTM) layers to encapsulate potential time-
dependence of the signals. The proposed method is illustrated and validated on measurements from
a remote-controlled drone, which is operated with two variants of controller hardware: a reference
controller unit and a malfunctioning counterpart. It is demonstrated that the presented approach
successfully identifies and describes the unexpected behaviour of the test device. In the presented
case study, the model outputs a signal sample prediction in 0.14 ms and achieves a reconstruction
accuracy of the validation data with a root mean square error (RMSE) below 0.04 relative to the data
range. In addition, three self-protection features (multidimensional boundary-check, Mahalanobis
distance, auxiliary autoencoder NN) are introduced to gauge the certainty of the VSN model output.
Keywords:
Virtual Sensor Network; Digital Twin; Mahalanobis Distance; Neural Network; LSTM;
uncertainty estimation; cybersecurity; Industrial Control System
1. Introduction
Rapid development of electronic hardware is in high demand due to the competitive
market space, set product life cycles and customer demand [
1
]. At the same time, it is
important to ensure that the manufactured hardware operates as intended and is tested
in the development process based on prototypes. Otherwise, the result could lead to
additional customer support, costly recalls or warranty claims.
For safety-critical applications, the requirement for functional integrity usually has
implications on the health and safety of humans [
2
] or critical infrastructures of whole
societies [
3
,
4
]. An unexpected breach of the functional integrity of electronic hardware can
occur due to a multitude of reasons. Internal factors include errors in the manufacturing
process, software bugs, defective components, upgrades or redesigns of hardware/software.
Similarly, external influences can also interfere with the proper operation of electronic
devices. This includes the use of new or changed components due to a changed supply
from manufacturers or change of the supplier. Hardware degradation is another factor,
which can cause avoidable costly failures or downtime [
5
,
6
]. Finally, malicious activities
such as hacking attacks or sabotage should be considered [
3
,
4
]. The validation of proper
Sensors 2022, 22, 454. https://doi.org/10.3390/s22020454 https://www.mdpi.com/journal/sensors
