ADL-ID: Adversarial Disentanglement Learning for Wireless Device Fingerprinting Temporal Domain Adaptation

Abstract

As the journey of 5G standardization is coming to an end, academia and industry have already begun to consider the sixth-generation (6G) wireless networks, with an aim to meet the service demands for the next decade. Deep learning-based RF fingerprinting (DL-RFFP) has recently been recognized as a potential solution for enabling key wireless network applications and services, such as spectrum policy enforcement and network access control. The state-of-the-art DL-RFFP frameworks suffer from a significant performance drop when tested with data drawn from a domain that is different from that used for training data. In this paper, we propose ADL-ID, an unsupervised domain adaption framework that is based on adversarial disentanglement representation to address the temporal domain adaptation for the RFFP task. Our framework has been evaluated on real LoRa and WiFi datasets and showed about 24% improvement in accuracy when compared to the baseline CNN network on short-term temporal adaptation. It also improves the classification accuracy by up to 9% on long-term temporal adaptation. Furthermore, we release a 5-day, 2.1TB, large-scale WiFi 802.11b dataset collected from 50 Pycom devices to support the research community efforts in developing and validating robust RFFP methods.

Comprehensive RF Dataset Collection and Release: A Deep Learning-Based Device Fingerprinting Use Case

“…The underlying assumption in these frameworks is that the source and target domains exhibit slightly different distributions. One notable domain adaptation framework, ADL-ID [18], integrates disentangled representation learning with adversarial learning to tackle the challenge of short-term temporal generalization in RFFP. ADL-ID involves segregating the feature vector into two distinct components: device-specific (fingerprints) and domain-specific features.…”

“…Impairment compensation techniques, on the other hand, target specific channel impairments, limiting their generalizability across different environments and wireless channels. Recently, domain adaptation techniques have emerged as a means to reduce the presence of domain information in the feature vector through adversarial learning [17], or to decouple device-specific and domain-specific features through adversarial disentanglement learning [18] and generative adversarial networks [19]. However, these techniques do not scale well and fail to maintain a practically acceptable performance for medium to large numbers of devices.…”