An Ultra Lightweight Neural Network for Automatic Modulation Classification in Drone Communications
Authors:
Preprint
- 19 April 2024
- preprint
- Published by Springer Nature in Research Square
Abstract
Automatic Modulation Classification (AMC) is crucial for unmanned aerial vehicle (UAV) systems in non-cooperative communications. It enables UAVs to effectively identify and track signals transmitted by other communication devices. Deep Learning (DL) has been successfully applied to AMC to improve the accuracy of signal classification. Despite this, many DL-based AMC methods, due to their large number of parameters and high computational complexity, cannot be directly applied to UAV platforms with limited computing power and storage space. To address this challenge, we propose an ultra-lightweight neural network (ULNN). This network incorporates a lightweight convolutional structure, attention mechanism, and cross-channel feature fusion technique. Additionally, we introduce data augmentation (DA) based on signal phase offsets during the model training process, aimed at improving the model’s generalization ability and preventing overfitting. Through experimental validation on the public dataset RML2016.10A, our proposed ULNN network achieves an average precision of 62.83% with only 8,815 parameters and reaches a peak classification accuracy of 92.11% at SNR = 10dB. This demonstrates that our proposed ULNN network maintains high recognition accuracy while keeping the model lightweight, making it highly suitable for deployment in resource constrained environments.Keywords
All Related Versions
- Version 1, 2024-04-19, Research Square
- Published version: Scientific Reports, 14 (1), 1.
This publication has 28 references indexed in Scilit:
- Time and phase features network model for automatic modulation classificationComputers and Electrical Engineering, 2023
- A real-time transformer discharge pattern recognition method based on CNN-LSTM driven by few-shot learningElectric Power Systems Research, 2023
- Spectrum interference-based two-level data augmentation method in deep learning for automatic modulation classificationNeural Computing and Applications, 2020
- Data Augmentation for Deep Learning-Based Radio Modulation ClassificationIEEE Access, 2019
- Deep long short-term memory networks-based automatic recognition of six different digital modulation types under varying noise conditionsNeural Computing and Applications, 2019
- Radio transformer networks: Attention models for learning to synchronize in wireless systems2016
- Convolutional Radio Modulation Recognition Networks2016
- Signal Identification for Multiple-Antenna Wireless Systems: Achievements and ChallengesIEEE Communications Surveys & Tutorials, 2016
- Maximum-likelihood classification for digital amplitude-phase modulationsIEEE Transactions on Communications, 2000
- Algorithms for automatic modulation recognition of communication signalsIEEE Transactions on Communications, 1998