SENTINEL: Securing Indoor Localization Against Adversarial Attacks With Capsule Neural Networks

• Published in: IEEE transactions on computer-aided design of integrated circuits and systems 2024-11, Vol.43 (11), p.4021-4032
• Main Authors: Gufran, Danish, Anandathirtha, Pooja, Pasricha, Sudeep
• Format: Article
• Language: English
• Subjects: Adversarial attacks; adversarial training; capsule neural networks; Datasets; device heterogeneity; Error reduction; evil twin attacks; Fingerprint recognition; Fingerprinting; Fluctuations; Heterogeneity; Indoor environment; Indoor environments; Localization; Location awareness; Location based services; Machine learning; man-in-the-middle attacks; Neural networks; Real-time systems; Resilience; rogue access points (APs); Signal strength; Training; Wi-Fi received signal strength (RSS) fingerprinting; Wireless fidelity; Working environment noise
• ISSN: 0278-0070; 1937-4151
• DOI: 10.1109/TCAD.2024.3446717

With the increasing demand for edge device-powered location-based services in indoor environments, Wi-Fi received signal strength (RSS) fingerprinting has become popular, given the unavailability of GPS indoors. However, achieving robust and efficient indoor localization faces several challenges, due to RSS fluctuations from dynamic changes in indoor environments and heterogeneity of edge devices, leading to diminished localization accuracy. While advances in machine learning (ML) have shown promise in mitigating these phenomena, it remains an open problem. Additionally, emerging threats from adversarial attacks on ML-enhanced indoor localization systems, especially those introduced by malicious or rogue access points (APs), can deceive ML models to further increase localization errors. To address these challenges, we present SENTINEL, a novel embedded ML framework utilizing modified capsule neural networks to bolster the resilience of indoor localization solutions against adversarial attacks, device heterogeneity, and dynamic RSS fluctuations. We also introduce RSSRogueLoc, a novel dataset capturing the effects of rogue APs from several real-world indoor environments. Experimental evaluations demonstrate that SENTINEL achieves significant improvements, with up to 3.5\times reduction in mean error and 3.4\times reduction in worst-case error compared to state-of-the-art frameworks using simulated adversarial attacks. SENTINEL also achieves improvements of up to 2.8\times in mean error and 2.7\times in worst-case error compared to state-of-the-art frameworks when evaluated with the real-world RSSRogueLoc dataset.