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.