Doppler Invariant CNN for Signal Classification

TL;DR

This paper introduces a complex-valued CNN architecture with adaptive pooling that achieves Doppler shift invariance in signal classification, reducing reliance on data augmentation and improving robustness.

Contribution

The proposed CNN architecture exploits shift equivariance and uses adaptive polyphase sampling to establish provable frequency shift invariance for signal classification.

Findings

Maintains consistent accuracy with and without Doppler shifts

Outperforms vanilla CNN in robustness to Doppler effects

Provides a framework with provable invariance against real-world frequency shifts

Abstract

Radio spectrum monitoring in contested environments motivates the need for reliable automatic signal classification technology. Prior work highlights deep learning as a promising approach, but existing models depend on brute-force Doppler augmentation to achieve real-world generalization, which undermines both training efficiency and interpretability. In this paper, we propose a convolutional neural network (CNN) architecture with complex-valued layers that exploits convolutional shift equivariance in the frequency domain. To establish provable frequency bin shift invariance, we use adaptive polyphase sampling (APS) as pooling layers followed by a global average pooling layer at the end of the network. Using a synthetic dataset of common interference signals, experimental results demonstrate that unlike a vanilla CNN, our model maintains consistent classification accuracy with and without random Doppler shifts despite being trained on no Doppler-shifted examples. Overall, our method establishes an invariance-driven framework for signal classification that offers provable robustness against real-world effects.