row1
Abstract
Fingerprinting-based indoor localization has been a research focus for GPS-denied areas. The development of neural networks has greatly promoted its application in indoor localization systems. However, recent studies showed that machine learning models, including state-of-the-art neural networks, are vulnerable to adversarial examples, and thus, neural network-based indoor localization systems are also under the threat of adversarial attacks.
To investigate the effect of adversarial attacks on indoor localization systems and to make such systems resilient to adversarial attacks, we propose AdvLoc, an adversarial deep learning indoor localization system. With the proposed AdvLoc system, the effect of adversarial attacks on indoor localization is studied under six types of adversarial attack methods in both black-box attack and white-box attack scenarios. Furthermore, adversarial training is utilized in offline training of the proposed AdvLoc system, which is effective against first-order adversarial attacks.
The proposed AdvLoc system is implemented with commodity WiFi devices and evaluated with extensive experiments in two representative indoor environments. The experimental results verify the robustness of the proposed system against first-order adversarial attacks in representative indoor environments.
Summary
The white paper details the effects of adversarial attacks on indoor localization systems. To combat this, a deep learning system was proposed named AdvLoc, which is an Adversarial deep learning for indoor Localization system. AdvLoc operates in two stages, an offline training stage and an online location estimation stage. The AdvLoc system was trained to combat the six common adversarial attacks used against WiFi Channel State Information (CSI) based indoor localization systems. The attacks were deployed in black-box and white-box scenarios, which are used for outer and inner structure implementation respectively. The extensive experimental study exposed the threat of adversarial attacks to indoor localization systems and validated performance of the AdvLoc system in defending against first-order adversarial attacks.
