Related Experiment Video
Updated: Aug 16, 2026
09:43
Transmission of Multiple Signals through an Optical Fiber Using Wavefront Shaping
Published on: March 20, 2017
A novel technique for automatic hybrid modulation classification in UWA communications using 3D constellation
Mohamed A Abdel-Moneim1, Khalil F Ramadan2, El-Sayed M El-Rabaie2
1Department of Telecommunication, Faculty of Engineering, Egyptian Russian University, Cairo, 11829, Egypt.
Scientific Reports
|August 14, 2026
Summary
This study introduces a novel deep learning method for automatic modulation classification in underwater acoustic (UWA) communications. The approach accurately identifies hybrid modulation schemes even in challenging UWA environments with low signal-to-noise ratios.
Area of Science:
- Underwater acoustic (UWA) communications
- Signal processing
- Machine learning
Background:
- UWA systems face severe channel impairments like multipath, noise, and Doppler effects.
- These impairments complicate automatic modulation classification (AMC), especially in dense networks with interference.
- Limited research exists on AMC for frequency-indexed hybrid modulations in UWA environments.
Purpose of the Study:
- Investigate AMC for frequency-indexed hybrid modulation schemes (FPK, FQAM) in UWA environments.
- Develop a robust feature extraction and modulation discrimination method.
- Evaluate the proposed framework's performance under various transmission schemes and conditions.
Main Methods:
- Utilized a three-dimensional (3D) constellation representation for signal analysis.
- Converted 3D signal representations into image-based inputs.
- Employed deep convolutional neural networks (CNNs) like AlexNet, VGG-19, and ResNet50 for classification.
Main Results:
- Achieved accurate modulation recognition in severe UWA conditions and low signal-to-noise ratio (SNR).
- Demonstrated the framework's robustness under both single-carrier (SC) and orthogonal frequency-division multiplexing (OFDM) schemes.
- Confirmed the effectiveness of deep learning for AMC in hybrid UWA systems.
Conclusions:
- The proposed deep learning-based AMC framework is effective for hybrid modulation schemes in UWA systems.
- Combining hybrid modulation with advanced CNN models enhances communication reliability in challenging underwater environments.
- This research paves the way for next-generation UWA communication systems.
