The increasing demand for efficient Internet of Things (IoT) solutions has accelerated the adoption of Digital Twin (DT) technologies for real-time monitoring and data-driven decision-making. A key challenge in IoT is the design of Electronic DTs (EDTs) for modeling intentional hardware tampering also providing realistic attack scenarios for Machine Learning (ML)-based anomaly detection. This paper proposes an integrated edge-fog DT architecture, combining EDTs, Device DTs (DDTs), and ML forecasting techniques where the EDT replicates the electrical characteristics of the sensing circuit, enabling circuit-level threat modeling of Hardware Trojan induced perturbations. The proposed framework has been implemented using open-source tools and off-the-shelf IoT devices and validated through a reference use case based on a rainwater harvesting system, composed of multiple storage tanks, monitored by the IoT infrastructure. A liquid level sensor is adopted as a representative device to model the interaction between physical nodes, circuitlevel behavior, and application-level functionalities. Experimental evaluation, conducted on a dataset reflecting rainfall patterns in Apulia region in Italy, demonstrates the effectiveness of the forecasting models in capturing sensor dynamics and identifying deviations from expected behavior. Early results also confirmed that the adopted residual-based statistical analysis provides a feasible and lightweight solution for HT detection enhancing the security and reliability of smart IoT systems.

An Integrated Device and Electronic Digital Twin for IoT Monitoring and Hardware Threat Analysis: A Liquid Level Sensor Application

Massaro, Alessandro;Loseto, Giuseppe
;
Epicoco, Nicola;Gramegna, Filippo
2026-01-01

Abstract

The increasing demand for efficient Internet of Things (IoT) solutions has accelerated the adoption of Digital Twin (DT) technologies for real-time monitoring and data-driven decision-making. A key challenge in IoT is the design of Electronic DTs (EDTs) for modeling intentional hardware tampering also providing realistic attack scenarios for Machine Learning (ML)-based anomaly detection. This paper proposes an integrated edge-fog DT architecture, combining EDTs, Device DTs (DDTs), and ML forecasting techniques where the EDT replicates the electrical characteristics of the sensing circuit, enabling circuit-level threat modeling of Hardware Trojan induced perturbations. The proposed framework has been implemented using open-source tools and off-the-shelf IoT devices and validated through a reference use case based on a rainwater harvesting system, composed of multiple storage tanks, monitored by the IoT infrastructure. A liquid level sensor is adopted as a representative device to model the interaction between physical nodes, circuitlevel behavior, and application-level functionalities. Experimental evaluation, conducted on a dataset reflecting rainfall patterns in Apulia region in Italy, demonstrates the effectiveness of the forecasting models in capturing sensor dynamics and identifying deviations from expected behavior. Early results also confirmed that the adopted residual-based statistical analysis provides a feasible and lightweight solution for HT detection enhancing the security and reliability of smart IoT systems.
2026
Edge Computing, Digital Twin, Hardware Trojan, Liquid Level Sensor, Forecasting Models
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12572/39429
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
social impact