A Novel Micro-Doppler Coherence Loss for Deep Learning Radar Applications
This addresses a specific issue in radar signal processing for micro-Doppler applications, but it is incremental as it modifies an existing deep learning framework rather than proposing a new paradigm.
The paper tackles the problem of deep learning models for micro-Doppler applications not prioritizing relevant features due to unadjusted objective functions, and introduces a micro-Doppler coherence loss that results in models more resilient to noise on real data.
Deep learning techniques are subject to increasing adoption for a wide range of micro-Doppler applications, where predictions need to be made based on time-frequency signal representations. Most, if not all, of the reported applications focus on translating an existing deep learning framework to this new domain with no adjustment made to the objective function. This practice results in a missed opportunity to encourage the model to prioritize features that are particularly relevant for micro-Doppler applications. Thus the paper introduces a micro-Doppler coherence loss, minimized when the normalized power of micro-Doppler oscillatory components between input and output is matched. The experiments conducted on real data show that the application of the introduced loss results in models more resilient to noise.