Abstract:Automatic modulation classification is of crucial importance in wireless communication networks. Deep learning based automatic modulation classification schemes have attracted extensive attention due to the superior accuracy. However, the data-driven method relies on a large amount of training samples and the classification accuracy is poor in the low signal-to-noise radio (SNR). In order to tackle these problems, a novel data-and-knowledge dual-driven automatic modulation classification scheme based on radio frequency machine learning is proposed by exploiting the attribute features of different modulations. The visual model is utilized to extract visual features. The attribute learning model is used to learn the attribute semantic representations. The transformation model is proposed to convert the attribute representation into the visual space. Extensive simulation results demonstrate that our proposed automatic modulation classification scheme can achieve better performance than the benchmark schemes in terms of the classification accuracy, especially in the low SNR. Moreover, the confusion among high-order modulations is reduced by using our proposed scheme compared with other traditional schemes.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the challenges of automatic modulation classification (AMC) in wireless communication networks, especially the problem of poor classification performance in low signal - to - noise ratio (SNR) environments. Although traditional data - driven methods perform well at high SNR, in practical applications, especially when the SNR is low, the classification accuracy of these methods drops significantly. In addition, these methods require a large number of training samples, which are often difficult to obtain in actual communication scenarios. To solve these problems, the paper proposes a new data - and - knowledge - dual - driven automatic modulation classification scheme based on radio - frequency machine learning (RFML). This scheme utilizes the attribute characteristics of different modulation methods, combines the visual features extracted by the visual model, and the semantic representations learned by the attribute learning model, and converts the attribute representations into the visual space through the transformation model, thereby achieving more accurate modulation classification. Experimental results show that the proposed scheme has better classification performance in low - SNR environments and reduces the confusion between high - order modulations. Specifically, the main contributions of the paper include: 1. **Introducing semantic information for the first time**: By introducing semantic information such as class attributes, the number of required training samples is reduced, which is especially important for actual complex and dynamic wireless networks. 2. **Improved residual network**: An improved residual network for attribute learning is proposed. 3. **Data - and - knowledge - dual - driven framework**: A new data - and - knowledge - dual - driven framework is constructed to learn multi - dimensional representations of different modulation methods from I/Q signals. 4. **Performance verification**: Simulation results show that the proposed scheme has better classification performance in low - SNR environments, especially significantly reducing the confusion between 16QAM and 64QAM. In summary, this paper aims to improve the performance of automatic modulation classification in low - SNR environments by combining data - driven and knowledge - driven methods, reduce the dependence on a large number of training samples, and improve the classification accuracy.

Data-and-Knowledge Dual-Driven Automatic Modulation Recognition for Wireless Communication Networks

An Improved Modulation Recognition Algorithm Based on Fine-Tuning and Feature Re-Extraction

Data and Knowledge Dual-Driven Automatic Modulation Classification for 6G Wireless Communications

A Novel Automatic Modulation Classification Scheme Based on Multi-Scale Networks

Attention Mechanism Based ResNeXt Network for Automatic Modulation Classification.

Automatic Modulation Classification of Signals Using Wavelet Transform and Pattern Recognition

A deep learning method based on convolutional neural network for automatic modulation classification of wireless signals

Automatic modulation classification of digital modulation signals with stacked autoencoders

Automatic Modulation Recognition Based on Deep-Learning Features Fusion of Signal and Constellation Diagram

Deep Learning-Based Automatic Modulation Recognition in OTFS and OFDM Systems.

Deep Convolutional Neural Network with Wavelet Decomposition for Automatic Modulation Classification

An efficient automatic modulation recognition using time–frequency information based on hybrid deep learning and bagging approach

Deep Learning-based Automatic Modulation Recognition Algorithm in Non-Cooperative Communication Systems

Data-Transform Multi-Channel Hybrid Deep Learning for Automatic Modulation Recognition

Machine Learning Based Automatic Modulation Recognition for Wireless Communications: A Comprehensive Survey

A Novel Attention Cooperative Framework for Automatic Modulation Recognition

Automatic Recognition of Communication Signal Modulation Based on the Multiple-Parallel Complex Convolutional Neural Network

A Modulation Recognition Algorithm via Hybrid Feature Analysis in Aeronautical Wireless Channel

Automatic Modulation Classification Using a Deep Multi-Stream Neural Network

Low-SNR Modulation Recognition based on Deep Learning on Software Defined Radio

Radio Modulation Classification Optimization Using Combinatorial Deep Learning Technique