# Radio Frequency Signal Recognition of Unmanned Aerial Vehicle Based on Complex-Valued Convolutional Neural Network

## TL;DR

This paper introduces a complex-valued CNN for recognizing UAVs using RF signals, which performs better than traditional methods in low-noise conditions.

## Contribution

A novel complex-valued CNN is proposed that uses both magnitude and phase information for improved UAV signal recognition.

## Key findings

- The CV-CNN achieves 100% accuracy under noise-free conditions on the DroneRFa dataset.
- At −20 dB SNR, the CV-CNN outperforms a dual-channel RV-CNN by over seven times in accuracy.
- The model shows significantly better robustness in low-SNR environments compared to real-valued CNNs.

## Abstract

The rapid development of unmanned aerial vehicle (UAV) technology necessitates reliable recognition methods. Radio frequency (RF)-based recognition is promising, but conventional real-valued CNNs (RV-CNNs) typically discard phase information from RF spectrograms, leading to degraded performance under low-signal-to-noise ratio (SNR) conditions. To address this, this paper proposes a complex-valued CNN (CV-CNN) that operates on a constructed complex representation, where the real part is the logarithmic power spectral density (PSD) and the imaginary part is derived from Sobel edge detection. This enables genuine complex convolutions that fuse magnitude and structural cues, enhancing noise resilience. As complex-valued networks are known to be sensitive to architectural choices, we conduct comprehensive ablation experiments to investigate the impact of key hyperparameters on model performance, revealing critical stability constraints (e.g., performance collapse beyond 4–5 network depth). Evaluated on the 25-class DroneRFa dataset, the proposed model achieves 100.00% accuracy under noise-free conditions. Crucially, it demonstrates significantly superior robustness in low-SNR regimes: at −20 dB SNR, it attains 15.58% accuracy, over seven times higher than a dual-channel RV-CNN (2.20%) with identical inputs; at −15 dB, it reaches 45.86% versus 14.03%. These results demonstrate that the CV-CNN exhibits potentially superior robustness and interference resistance in comparison to its real-valued counterpart, maintaining high recognition accuracy even under low-SNR conditions.