# Fast Deep Learning for Automatic Modulation Classification

arXiv: https://arxiv.org/abs/1901.05850 · HTML: https://mlsanity.com/paper/1901.05850/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Jan 2019.

## Built on by (most cited first, 21 of 21)

- [Distributed Learning in Wireless Networks: Recent Progress and Future Challenges](https://mlsanity.com/paper/2104.02151/) (2021)
- [Autoregressive Belief Propagation for Decoding Block Codes](https://mlsanity.com/paper/2103.11780/) (2021)
- [A Novel Automatic Modulation Classification Scheme Based on Multi-Scale Networks](https://mlsanity.com/paper/2105.15037/) (2021)
- [NMformer: A Transformer for Noisy Modulation Classification in Wireless Communication](https://mlsanity.com/paper/2411.02428/) (2024)
- [SimPSI: A Simple Strategy to Preserve Spectral Information in Time Series Data Augmentation](https://mlsanity.com/paper/2312.05790/) (2023)
- [Neural Decoding with Optimization of Node Activations](https://mlsanity.com/paper/2206.00786/) (2022)
- [Enhancing Automatic Modulation Recognition for IoT Applications Using Transformers](https://mlsanity.com/paper/2403.15417/) (2024)
- [Gradient-based Adversarial Deep Modulation Classification with Data-driven Subsampling](https://mlsanity.com/paper/2104.06375/) (2021)
- [A Hybrid Training-time and Run-time Defense Against Adversarial Attacks in Modulation Classification](https://mlsanity.com/paper/2407.06807/) (2024)
- [RF Signal Classification with Synthetic Training Data and its Real-World Performance](https://mlsanity.com/paper/2206.12967/) (2022)
- [EMC2-Net: Joint Equalization and Modulation Classification based on Constellation Network](https://mlsanity.com/paper/2303.10934/) (2023)
- [Revolution of Wireless Signal Recognition for 6G: Recent Advances, Challenges and Future Directions](https://mlsanity.com/paper/2503.08091/) (2025)
- [Plug-and-Play AMC: Context Is King in Training-Free, Open-Set Modulation with LLMs](https://mlsanity.com/paper/2505.03112/) (2025)
- [Attention-based Adversarial Robust Distillation in Radio Signal Classifications for Low-Power IoT Devices](https://mlsanity.com/paper/2506.11892/) (2025)
- [Vision Transformer with Adversarial Indicator Token against Adversarial Attacks in Radio Signal Classifications](https://mlsanity.com/paper/2507.00015/) (2025)
- [Deep neural network goes lighter: A case study of deep compression techniques on automatic RF modulation recognition for Beyond 5G networks](https://mlsanity.com/paper/2204.04390/) (2022)
- [Using Early Exits for Fast Inference in Automatic Modulation Classification](https://mlsanity.com/paper/2308.11100/) (2023)
- [Federated Learning for UAV-Based Spectrum Sensing: Enhancing Accuracy Through SNR-Weighted Model Aggregation](https://mlsanity.com/paper/2411.11159/) (2024)
- [Signal Processing Based Deep Learning for Blind Symbol Decoding and Modulation Classification](https://mlsanity.com/paper/2106.10543/) (2021)
- [Hybrid HMM Decoder For Convolutional Codes By Joint Trellis-Like Structure and Channel Prior](https://mlsanity.com/paper/2210.14749/) (2022)
- [Data-Driven Subsampling in the Presence of an Adversarial Actor](https://mlsanity.com/paper/2401.03488/) (2024)

## Abstract (arXiv)

In this work, we investigate the feasibility and effectiveness of employing deep learning algorithms for automatic recognition of the modulation type of received wireless communication signals from subsampled data. Recent work considered a GNU radio-based data set that mimics the imperfections in a real wireless channel and uses 10 different modulation types. A Convolutional Neural Network (CNN) architecture was then developed and shown to achieve performance that exceeds that of expert-based approaches. Here, we continue this line of work and investigate deep neural network architectures that deliver high classification accuracy. We identify three architectures - namely, a Convolutional Long Short-term Deep Neural Network (CLDNN), a Long Short-Term Memory neural network (LSTM), and a deep Residual Network (ResNet) - that lead to typical classification accuracy values around 90% at high SNR. We then study algorithms to reduce the training time by minimizing the size of the training data set, while incurring a minimal loss in classification accuracy. To this end, we demonstrate the performance of Principal Component Analysis in significantly reducing the training time, while maintaining good performance at low SNR. We also investigate subsampling techniques that further reduce the training time, and pave the way for online classification at high SNR. Finally, we identify representative SNR values for training each of the candidate architectures, and consequently, realize drastic reductions of the training time, with negligible loss in classification accuracy.