Skip to content
Open access

Anomaly-Based Intrusion Detection for IoT Microcontrollers Using Power Side-Channel Fingerprinting

Sep 2026 · Italian National Conference on Sensors · Vol 26 · 0 citations · 55 references
Medicine

TL;DR

This study introduces a unified evaluation framework for device-level intrusion detection using power side-channel fingerprints under an open-set threat model to target post-enrollment device substitution, and reveals that temporal viability is strongly influenced by feature representation under the evaluated conditions.

Abstract

This study introduces a unified evaluation framework for device-level intrusion detection using power side-channel fingerprints under an open-set threat model. We target post-enrollment device substitution: an adversary replaces a legitimate Commercial Off-The-Shelf (COTS) node with a counterfeit of the same model and firmware after enrollment, during deployment or maintenance, making the substitution invisible to credential-based authentication. To address this scenario, we evaluate seven ESP32-WROOM-32 and seven Arduino Uno R3 devices across four sessions spanning 45 days. Each platform group is drawn from a single manufacturing batch so that measured differences reflect within-batch variation; results are correspondingly established at this scale and for this sourcing. Unlike most prior studies that evaluate fingerprinting under isolated assumptions, this study treats intrusion detection as a system-level problem that requires the joint consideration of temporal stability, open-set recognition, and feature representation. We conduct a two-tier evaluation. Tier 1 characterizes device uniqueness and temporal stability using handcrafted statistical features and neural embeddings trained via triplet + Additive-Margin Softmax (AM-Softmax) loss, analyzed via separation ratios, d′ scores, F1-scores, and statistical power. Tier 2 applies embeddings to an open-set intrusion-detection protocol using percentile-based thresholding. This analysis reveals that temporal viability is strongly influenced by feature representation under the evaluated conditions. Over 45 days, under an identical Leave-One-Run-Out protocol and scoring rule, neural embeddings achieve a 93.9% true positive rate at a 9.9% false acceptance rate, whereas the same rule applied to handcrafted statistical features yields 66.4% and 35.8%. When each device identity is withheld from training altogether, embeddings hold a 26.0% false acceptance rate against 39.0% for statistical features. Platform comparison shows that ESP32 achieves a 1.80× higher separation ratio than Arduino (3.06 vs. 1.70), with architecture-dependent discriminability. Statistical power analysis yields 89.1% power at d=0.774. These results identify representation-level robustness as a key requirement for practical power-based intrusion detection, while longer-term validation remains an important direction for future work.

Read PDF

Similar papers

Preprint Sep 2026

Improving the Reliability of Anomaly Detection for Encrypted OPC UA Traffic over Private 5G

Open Platform Communications Unified Architecture (OPC UA) is increasingly deployed over private 5G networks in industrial environments, where end-to-end encryption prevents payload inspection by network-based intrusion detection systems (IDSs). Although payload-agnostic statistical features extracted from encrypted tr...

Son-Ha Song, Florian Foerster, Henry Beuster et al. · 0 citations
Conference Sep 2026

Comparative Malware Behavior Analysis in a Docker-Based Wazuh Security Operations Center Testbed

Modern security monitoring requires laboratory environments that are not only effective for threat observation but also practical to deploy and reproduce. This research presents a Docker-based Wazuh Security Operations Center (SOC) testbed for comparative malware behavior analysis in an isolated Windows 10 environment...

Aurelia Sandra Darmawan, D. Maharani, Alwan Muhammad Fajar et al. · 0 citations
Open access 2026

Real-Time Endpoint Ransomware Detection and Prevention via Strategic Deception Directories and Block-Level Entropy Analysis

A lightweight host-based ransomware detection and prevention prototype for Windows endpoints that combines three mechanisms: strategically placed deception directories (1000_Trap), localized Shannon-entropy analysis, and automated process termination using psutil is presented.

Kia-Yi Tong, Muhammad Bin Amran, Megat Muhammad Ridzuan et al. · 0 citations
Conference Open access Sep 2026

DOS Classification, and Virtual Machine Detection via Passive Monitoring through Security Twins

The security twin (ST) paradigm is an emerging framework for dynamically modeling and defending complex cyber-physical systems. This paper extends NotLine, a fully automated, non-intrusive ST construction platform with three mechanisms. First, a lightweight statistical engine based on the numerically stable Welford alg...

Vincenzo Sammartino, Fabrizio Baiardi, Salvatore Ruggieri · 0 citations
Sep 2026

Secure over-the-air firmware update mechanism for embedded IoT devices using lightweight cryptographic authentication

Securing over-the-air firmware updates for resource-constrained IoT devices remains challenging, as existing solutions either rely on memory-intensive cryptographic algorithms or provide only basic integrity checks. This study aims to address this gap by developing a resource-efficient yet cryptographically comprehensi...

Xiang-Ning Meng · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.