AUDRON: A Deep Learning Framework with Fused Acoustic Signatures for Drone Type Recognition

2025-12-23

AI Summary

Low-cost, non-intrusive acoustic identification of unmanned aerial vehicles (UAVs) is critical for effective safety regulation but remains challenging under real-world noisy and variable environmental conditions. Method: This paper proposes an end-to-end deep learning framework featuring a novel multi-granularity acoustic feature-level fusion mechanism—jointly modeling static spectral characteristics via MFCCs, dynamic time-frequency evolution via STFT spectrograms, and compact low-dimensional representations learned by an autoencoder. The architecture integrates CNNs for local spectral pattern extraction and RNNs for capturing long-range temporal dependencies. Contribution/Results: Evaluated under cross-environmental noise conditions, the model achieves 98.51% accuracy in binary UAV classification and 97.11% in multi-class identification, demonstrating superior generalization and robustness. Its lightweight design and inference efficiency indicate strong potential for real-time deployment in practical UAV monitoring systems.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Multi-modal Vision

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Unmanned aerial vehicles (UAVs), commonly known as drones, are increasingly used across diverse domains, including logistics, agriculture, surveillance, and defense. While these systems provide numerous benefits, their misuse raises safety and security concerns, making effective detection mechanisms essential. Acoustic sensing offers a low-cost and non-intrusive alternative to vision or radar-based detection, as drone propellers generate distinctive sound patterns. This study introduces AUDRON (AUdio-based Drone Recognition Network), a hybrid deep learning framework for drone sound detection, employing a combination of Mel-Frequency Cepstral Coefficients (MFCC), Short-Time Fourier Transform (STFT) spectrograms processed with convolutional neural networks (CNNs), recurrent layers for temporal modeling, and autoencoder-based representations. Feature-level fusion integrates complementary information before classification. Experimental evaluation demonstrates that AUDRON effectively differentiates drone acoustic signatures from background noise, achieving high accuracy while maintaining generalizability across varying conditions. AUDRON achieves 98.51 percent and 97.11 percent accuracy in binary and multiclass classification. The results highlight the advantage of combining multiple feature representations with deep learning for reliable acoustic drone detection, suggesting the framework's potential for deployment in security and surveillance applications where visual or radar sensing may be limited.

Problem

• Recognizes drone types using acoustic signatures
• Detects drones with fused deep learning features
• Differentiates drone sounds from background noise accurately

Innovation

• Deep learning framework with fused acoustic signatures
• Combines MFCC, STFT spectrograms, CNNs, and recurrent layers
• Feature-level fusion for classification and noise differentiation