Embedding-Assisted Attentional Deep Learning for Real-World RF
Fingerprinting of Bluetooth
Abstract
A computationally efficient framework to fingerprint real-world
Bluetooth devices is presented in this work. Despite the active research
in this topic, a generalizable framework suitable for real-world
deployment in terms of performance in a new and evolving environment as
well as hardware efficiency of the architecture is lacking. We propose
an embedding-assisted attentional framework (Mbed-ATN) suitable for
fingerprinting actual Bluetooth devices and analyze its generalization
capability in different settings and demonstrate the effect of sample
length and anti-aliasing decimation on its performance. The embedding
module serves as a dimensionality reduction unit that maps the high
dimensional 3D input tensor to a 1D feature vector for further
processing by the ATN module. Furthermore, unlike the prior research in
this field, we closely evaluate the complexity of the model and test its
fingerprinting capability with real-world Bluetooth dataset collected
under a different time frame and experimental setting while being
trained on another. Our study reveals a 7.3x and 65.2x lesser memory
usage with the proposed Mbed-ATN architecture in contrast to Oracle at
input sample lengths of M=10 kS and M=100 kS respectively. Further, the
proposed Mbed-ATN showcases a 16.9x fewer FLOPs and 7.5x lesser
trainable parameters when compared to Oracle. Finally, we show that when
subject to anti-aliasing decimation and at greater input sample lengths
of 1 MS, the proposed Mbed-ATN framework results in a 5.32x higher TPR,
37.9 % fewer false alarms, and 6.74x higher accuracy under the
challenging real-world setting.