Related Experiment Video
Updated: Mar 15, 2026
09:44
Author Spotlight: Advancing Large-Scale Neural Dynamics Through HD-MEA Technology
Published on: March 8, 2024
6.1K
Communication Signal Modulation Recognition Method Based on Multi-Feature Multi-Channel ResNet and BiLSTM Neural
1College of Information Engineering, Shanghai Maritime University, Shanghai 201306, China.
Sensors (Basel, Switzerland)
|March 14, 2026
Summary
This study introduces a novel deep neural network for communication signal modulation recognition, achieving high accuracy even in low signal-to-noise ratio environments. The MF-MC ResNet-BiLSTM model significantly improves upon existing methods.
Area of Science:
- Electrical Engineering
- Computer Science
- Signal Processing
Background:
- Traditional signal modulation recognition methods suffer from insufficient accuracy.
- Deep learning offers potential for enhanced recognition capabilities.
Purpose of the Study:
- To propose a novel deep neural network for improved communication signal modulation recognition.
- To enhance recognition accuracy and robustness, especially in low SNR conditions.
Main Methods:
- A multi-feature multi-channel ResNet and BiLSTM (MF-MC ResNet-BiLSTM) neural network was developed.
- Data was converted into IQ, AP, and FFT formats for multi-channel input.
- Feature fusion was performed in a high-dimensional space, integrating ResNet-BiLSTM and an adaptive multi-head attention network.
Main Results:
- The proposed method achieved a recognition rate of 95.67% and a recall rate of 94.56% in low SNR environments (-22 dB-2 dB).
- Demonstrated superior performance compared to MMF, FGDNN, and LightMFFS networks.
- Exhibited good model generalization capabilities and robustness.
Conclusions:
- The MF-MC ResNet-BiLSTM method significantly advances communication signal modulation recognition.
- The approach is effective in challenging low SNR environments.
- This deep learning model offers a robust and accurate solution for signal modulation identification.
