Modulation Recognition Of Signals Research Articles

Sparse Adversarial Attacks against DL-Based Automatic Modulation Classification

Automatic modulation recognition (AMR) serves as a crucial component in domains such as cognitive radio and electromagnetic countermeasures, acting as a significant prerequisite for the efficient signal processing of receivers. Deep neural networks (DNNs), despite their effectiveness, are known to be vulnerable to adversarial attacks. This vulnerability has inspired the introduction of subtle interference to wireless communication signals—interference so minuscule that it is difficult for the human eye to discern. Such interference can mislead eavesdroppers into erroneous modulation pattern recognition when using DNNs, thereby camouflaging communication signal modulation patterns. Nonetheless, the majority of current camouflage methods used for electromagnetic signal modulation recognition rely on a global perturbation of the signal. They fail to consider the local agility of signal disturbance and the concealment requirements for bait signals that are intercepted by the interceptor. This paper presents a generator framework designed to produce perturbations with sparse properties. Furthermore, we introduce a method to reduce spectral loss, which minimizes the spectral difference between adversarial perturbation and the original signal. This method makes perturbation more challenging to monitor, thereby deceiving enemy electromagnetic signal modulation recognition systems. The experimental results validated that the proposed method significantly outperformed existing methods in terms of generation time. Moreover, it can generate adversarial signals characterized by high deceivability and transferability even under extremely sparse conditions.

Automatic Modulation Recognition Based on a New Deep K-SVD Denoising Algorithm

Automatic modulation recognition has a wide range of applications in wireless communication. To solve the problem that the previous methods convert signal modulation recognition into image recognition, leading to increased time costs and information loss, an automatic modulation recognition approach consisting of the improved deep K-SVD denoising algorithm is suggested. First, the effectiveness of the model for random sine wave denoising is demonstrated by simulation. Second, the original I/Q signals are fed into the modified deep K-SVD model for denoising, skipping the complicated image processing steps. Finally, the noise-reduced signals are input into a multi-channel convolutional long short-term neural network to complete the modulation recognition. To solve the slow convergence problem of Iterative Shrinkage Thresholding Algorithms in sparse coding, the Fast Iterative Shrinkage Thresholding Algorithm is adopted to improve the computational efficiency and obtain a better denoising effect. The experiments show that the improved model has an average recognition accuracy of 91.26% at different SNRs from -2 dB to 18 dB, which is better than the state-of-the-art modulation recognition models.

Stacking Ensemble Learning for the Modulation Recognition of Underwater Acoustic Communication Signals

Effective detection and recognition of underwater acoustic communication signals is of great significance in the field of ocean exploration. In this paper, a novel deep-learning recognition method of modulation modes of underwater acoustic communication signals is proposed. The stacking ensemble learning framework is established on the basis of several single deep-learning classifiers with different network structures and parameter configurations. The proposed method achieved an average recognition accuracy rate of more than 95% on the dataset generated under a simulation environment. The performance is also verified by means of a comparison with the classic SVM method.

MIDW‐Net: A multi‐tasking network architecture for radar intra‐pulse parameter description

AbstractThe automatic modulation recognition (AMR) of radar signals has become a popular research topic in recent years. However, most algorithms focus on the type of signal modulation and lack further understanding of the signal. To address this gap, a network architecture for multi‐tasking intra‐pulse description words (MIDW‐Net) is proposed herein. In this framework, the denoising algorithm employs a convolutional denoising autoencoder, which is an effective method for suppressing noise interference and preserving signal information. The multiscale feature‐extraction capability of a feature pyramid network (FPN) is utilized to expand the perceptual domain without losing the high‐frequency features of the image. Finally, AMR and modulation parameter estimation are accomplished via multitask learning. Experiments performed on simulated radar signals using four intra‐pulse descriptors verified the effectiveness of the proposed algorithm.

EMRNet: Efficient Modulation Recognition Networks for Continuous-Wave Radar Signals

Automatic modulation recognition (AMR) of radar signals is a currently active area, especially in electronic reconnaissance, where systems need to quickly identify the intercepted signal and formulate corresponding interference measures on computationally limited platforms. However, previous methods generally have high computational complexity and considerable network parameters, making the system unable to detect the signal timely in resource-constrained environments. This letter firstly proposes an efficient modulation recognition network (EMRNet) with tiny and lowlatency models to match the requirements for mobile reconnaissance equipments. One-dimensional residual depthwise separable convolutions block (1D-RDSB) with an adaptive size of receptive fields is developed in EMRNet to replace the traditional convolution block. With 1D-RDSB, EMRNet achieves a high classification accuracy and dramatically reduces computation cost and network paraments. The experiment results show that EMRNet can achieve higher precision than existing 2D-CNN methods, while the computational cost and parament amount of EMRNet are reduced by about 13.93× and 80.88×, respectively.

Multi-task neural network for solving the problem of recognizing the type of QAM and PSK modulation under parametric a priori uncertainty

Objectives. Automatic modulation recognition of unknown signals is an important task for various fields oftechnology such as radio control, radio monitoring, and identification of interference and sources of radio emission. The paper aims to develop a method for recognizing the types of signal modulation under conditions of parametric a priori uncertainty, including the uncertainty of carrier frequency- and initial signal phase values. An additional task consists in estimating the offset values of the carrier frequency or signal phase at the initial stage of the recognition process.Methods. A multi-task learning with artificial neural network and the theory of cumulants of random variables are used.Results. For signals with a carrier frequency and initial phase shift, cumulant approaches for QAM-8, APSK-16, QAM-64, and PSK-8 modulations are calculated. A multi-task learning with artificial neural network using cumulant features and a data standardization algorithm is presented. The results of the experiment show that using multi-task learning with an artificial neural network provides high accuracy of recognizing QAM-8 and APSK-16, QAM-64 and PSK-8 modulations with small mismatches of the carrier frequency or initial phase. The accuracy of determining the offset values from the carrier frequency or the initial phase for QAM-8, APSK-16, QAM-64, and PSK-8 modulation is high.Conclusions. The multi-task learning with neural network using high-order signal cumulants makes it possible not only to recognize modulation types with high accuracy under conditions of a priori uncertainty of signal parameters, but also to determine the offset values of carrier frequency or initial signal phase from expected values.

Modulation Recognition of Low-SNR UAV Radar Signals Based on Bispectral Slices and GA-BP Neural Network

In this paper, we address the challenge of low recognition rates in existing methods for radar signals from unmanned aerial vehicles (UAV) with low signal-to-noise ratios (SNRs). To overcome this challenge, we propose the utilization of the bispectral slice approach for accurate recognition of complex UAV radar signals. Our approach involves extracting the bispectral diagonal slice and the maximum bispectral amplitude horizontal slice from the bispectrum amplitude spectrum of the received UAV radar signal. These slices serve as the basis for subsequent identification by calculating characteristic parameters such as convexity, box dimension, and sparseness. To accomplish the recognition task, we employ a GA-BP neural network. The significant variations observed in the bispectral slices of different signals, along with their robustness against Gaussian noise, contribute to the high separability and stability of the extracted bispectral convexity, bispectral box dimension, and bispectral sparseness. Through simulations involving five radar signals, our proposed method demonstrates superior performance. Remarkably, even under challenging conditions with an SNR as low as −3 dB, the recognition accuracy for the five different radar signals exceeds 90%. Our research aims to enhance the understanding and application of modulation recognition techniques for UAV radar signals, particularly in scenarios with low SNRs.

Automatic modulation recognition of radiation source signals based on two-dimensional data matrix and improved residual neural network

Automatic modulation recognition (AMR) of radiation source signals is a research focus in the field of cognitive radio. However, the AMR of radiation source signals at low SNRs still faces a great challenge. Therefore, the AMR method of radiation source signals based on two-dimensional data matrix and improved residual neural network is proposed in this paper. First, the time series of the radiation source signals are reconstructed into two-dimensional data matrix, which greatly simplifies the signal preprocessing process. Second, the depthwise convolution and large-size convolutional kernels based residual neural network (DLRNet) is proposed to improve the feature extraction capability of the AMR model. Finally, the model performs feature extraction and classification on the two-dimensional data matrix to obtain the recognition vector that represents the signal modulation type. Theoretical analysis and simulation results show that the AMR method based on two-dimensional data matrix and improved residual network can significantly improve the accuracy of the AMR method. The recognition accuracy of the proposed method maintains a high level greater than 90% even at −14 dB SNR.

Modulation signal recognition based on feed‐forward attention mechanism

AbstractIn recent years, research on modulated signal recognition using deep learning (DL) has achieved remarkable success, allowing automatic modulation recognition (AMR) to play a crucial role in modern communication systems. The emergence of the attention mechanism has then rapidly led to a wide range of applications in image classification and speech recognition, which proves the effectiveness of the attention mechanism. In this paper, the authors propose a network model feed‐forward Attention mechanism with Residuals networks and Long Short‐Term Memory (RLADNN) based on feed‐forward attention mechanism with Residuals networks (Resnet) and Long Short‐Term Memory (LSTM), which takes the advantage that the Attention mechanism can effectively solve some long‐term memory problems in [−20:18] signal‐to‐noise ratio (SNR) for the recognition of 11 modulated signals with different SNR, and effectively improves the recognition rate.

Modulation Recognition Method for Underwater Acoustic Communication Signals Based on Passive Time Reversal-Autoencoder with the Synchronous Signals.

In the process of the modulation recognition of underwater acoustic communication signals, the multipath effect seriously interferes with the signal characteristics, reducing modulation recognition accuracy. The existing methods passively improve the accuracy from the perspective of selecting appropriate signal features, lacking specialized preprocessing for suppressing multipath effects. So, the accuracy improvement of the designed modulation recognition models is limited, and the adaptability to environmental changes is poor. The method proposed in this paper actively utilizes common synchronous signals in underwater acoustic communication as detection signals to achieve passive time reversal without external signals and designs a passive time reversal-autoencoder to suppress multipath effects, enhance signals' features, and improve modulation recognition accuracy and environmental adaptability. Firstly, synchronous signals are identified and estimated. Subsequently, a passive time reversal-autoencoder is designed to enhance power spectrum and square spectrum features. Finally, a modulation classification is performed using a convolutional neural network. The model is trained in simulation channels generated by Bellhop and tested in actual channels which are different from the training period. The average recognition accuracy of the six modulated signals is improved by 10% compared to existing passive modulation recognition methods, indicating good environmental adaptability as well.

Automatic modulation recognition of radar signals based on histogram of oriented gradient via improved principal component analysis

Automatic modulation recognition (AMR) of radar signals plays a critical role in electronic reconnaissance. Current AMR algorithms are mainly based on convolutional neural networks (CNN), which can learn the feature hierarchy by establishing high-level features from low-level features. However, for time–frequency analysis-based methods, distinct low-level features in the time–frequency spectrum can already reflect modulation characteristics. Thus, this study develops a novel approach based on low-level shape descriptors via histograms of oriented gradients (HOG) and support vector machine (SVM). Comparison studies with classic CNN-based methods have also been done to reveal the superiority of the designed approach. Experimental results demonstrate that the HOG-SVM approach has a more efficient performance. To further enhance the classification precision under low signal-to-noise ratios, an improved principal component analysis denoising algorithm is developed to improve signal quality under intense noise background. Experiments based on simulated and measured signals demonstrate that the proposed algorithm can accurately distinguish signals under intense noise environments.

Modulation Signal Recognition of Underwater Acoustic Communication Based on Archimedes Optimization Algorithm and Random Forest

This paper researches the recognition of modulation signals in underwater acoustic communication, which is the fundamental prerequisite for achieving noncooperative underwater communication. In order to improve the accuracy of signal modulation mode recognition and the recognition effects of traditional signal classifiers, the article proposes a classifier based on the Archimedes Optimization Algorithm (AOA) and Random Forest (RF). Seven different types of signals are selected as recognition targets, and 11 feature parameters are extracted from them. The decision tree and depth obtained by the AOA algorithm are calculated, and the optimized random forest after the AOA algorithm is used as the classifier to achieve the recognition of underwater acoustic communication signal modulation mode. Simulation experiments show that when the signal-to-noise ratio (SNR) is higher than -5dB, the recognition accuracy of the algorithm can reach 95%. The proposed method is compared with other classification and recognition methods, and the results show that the proposed method can ensure high recognition accuracy and stability.

Enhancing the generalization ability of deep learning model for radio signal modulation recognition

Enhancing the generalization ability of deep learning model for radio signal modulation recognition

Research on modulation recognition of underwater acoustic communication signal based on deep learning

Marine information technology plays a important role in the development of marine resources development, marine climate early warning, and other industries. Underwater acoustic communication technology can help us better access marine information. The performance of the underwater acoustic signal modulation recognition algorithm depends on the accuracy of feature extraction. however, due to excessive underwater noise, many traditional algorithms can not recognize features well. For this reason, this paper proposes a modulation recognition network for aquatic communication signals based on deep learning. ResNet can capture the characteristics of deep features and combines ResNet with a modulation recognition network. Finally, the experiment proves the effectiveness of this method.

Dynamic open set specific emitter identification via multi‐channel reconstructive discriminant network

AbstractSpecific emitter identification (SEI) is an emerging device authentication technology, which depends on the inherent hardware characteristics of wireless devices. By analysing the received signal, the hardware characteristics of a specific emitter can be extracted at the receiver and used for device authentication and association. Most of the existing SEI schemes focus on the identification under closed sets. In view of the explosive growth of the number of IoT devices, based on generative adversarial networks, this study proposes a method that considers the identification of unknown emitters. The reconstruction network to reconstruct the signals of the known class and fully train the feature space of the signals of the known class is designed. In the discriminator, two channels are specifically designed to perform anomaly detection for unknown signals and end‐to‐end closed‐set classification for known signals. In order to better reject unknown signals and accept known signals, Receiver operating characteristic curve and Youden index are used to determine the optimal threshold for anomaly detection. Under the given optimal threshold conditions, the identified threshold points have larger True Positive Rate. In addition, the time‐frequency feature combination vector is designed to reveal the essential characteristics of emitters. The experimental results on the real‐world datasets collected from universal software radio peripherals in short‐range communication scenario show that compared with the existing open‐set recognition methods such as C2AE, Openmax and SoftMax, the average recognition accuracy of the proposed framework improved by 0.08, 0.15, and 0.2 respectively, and the Marco‐F1 score improved by 0.05, 0.1, and 0.16, respectively, which proves the superiority of the proposed framework. In addition to identifying unknown and known Universal software radio peripherals in this article, some potential and widespread applications of the proposed framework include access user security authentication in IoT, such as Hack RF and Blue tooth devices. In addition, as a relatively general signal processing framework, the model can be used for air safety management and maritime ship identity authentication in civil aspects. In the military aspect, it can be used in electronic support and various signal reconnaissance scenarios, such as automatic signal modulation recognition in open set scenarios, individual identity determination of radar emitter and unknown working state identification of transmitters, which proves reference for subsequent research on open‐set signal recognition.

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

In signal communication based on a non-cooperative communication system, the receiver is an unlicensed third-party communication terminal, and the modulation parameters of the transmitter signal cannot be predicted in advance. After the RF signal passes through the RF band-pass filter, low noise amplifier, and image rejection filter, the intermediate frequency signal is obtained by down-conversion, and then the IQ signal is obtained in the baseband by using the intermediate frequency band-pass filter and down-conversion. In this process, noise and signal frequency offset are inevitably introduced. As the basis of subsequent analysis and interpretation, modulation recognition has important research value in this environment. The introduction of deep learning also brings new feature mining tools. Based on this, this paper proposes a signal modulation recognition method based on multi-feature fusion and constructs a deep learning network with a double-branch structure to extract the features of IQ signal and multi-channel constellation, respectively. It is found that through the complementary characteristics of different forms of signals, a more complete signal feature representation can be constructed. At the same time, it can better alleviate the influence of noise and frequency offset on recognition performance, and effectively improve the classification accuracy of modulation recognition.

Automatic Modulation Recognition of Radiation Source Signals Based on Data Rearrangement and the 2D FFT

It is a challenge for automatic modulation recognition (AMR) methods for radiation source signals to work in environments with low signal-to-noise ratios (SNRs). This paper proposes a modulation feature extraction method based on data rearrangement and the 2D fast Fourier transform (FFT) (DR2D), and a DenseNet feature extraction network with early fusion is constructed to recognize the extracted modulation features. First, the input signal is preprocessed by DR2D to obtain three types of joint frequency feature bins with multiple time scales. Second, the feature fusion operation is performed on the inputs of the different layers of the proposed network. Finally, feature recognition is completed in the subsequent layers. The theoretical analysis and simulation results show that DR2D is a fast and robust preprocessing method for extracting the features of modulated radiation source signals with less computational complexity. The proposed DenseNet feature extraction network with early fusion can identify the extracted modulation features with less spatial complexity than other types of convolutional neural networks (CNNs) and performs well in low-SNR environments.

Intelligent Modulation Recognition of Communication Signal for Next-Generation 6G Networks

In recent years, the need for a fast, efficient and a reliable wireless network has increased dramatically. Numerous 5G networks have already been tested while a few are in the early stages of deployment. In non-cooperative communication scenarios, the recognition of digital signal modulations assists people in identifying the communication targets and ensures an effective management over them. The recent advancements in both Machine Learning (ML) and Deep Learning (DL) models demand the development of effective modulation recognition models with self-learning capability. In this background, the current research article designs a Deep Learning enabled Intelligent Modulation Recognition of Communication Signal (DLIMR-CS) technique for next-generation networks. The aim of the proposed DLIMR-CS technique is to classify different kinds of digitally-modulated signals. In addition, the fractal feature extraction process is applied with the help of the Sevcik Fractal Dimension (SFD) approach. Then, the extracted features are fed into the Deep Variational Autoencoder (DVAE) model for the classification of the modulated signals. In order to improve the classification performance ofthe DVAE model, the Tunicate Swarm Algorithm (TSA) is used to fine-tune the hyperparameters involved in DVAE model. A wide range of simulations was conducted to establish the enhanced performance of the proposed DLIMR-CS model. The experimental outcomes confirmed the superior recognition rate of the DLIMR-CS model over recent state-of-the-art methods under different evaluation parameters.

A Method of Noise Reduction for Radio Communication Signal Based on RaGAN

Radio signals are polluted by noise in the process of channel transmission, which will lead to signal distortion. Noise reduction of radio signals is an effective means to eliminate the impact of noise. Using deep learning (DL) to denoise signals can reduce the dependence on artificial domain knowledge, while traditional signal-processing-based denoising methods often require knowledge of the artificial domain. Aiming at the problem of noise reduction of radio communication signals, a radio communication signal denoising method based on the relativistic average generative adversarial networks (RaGAN) is proposed in this paper. This method combines the bidirectional long short-term memory (Bi-LSTM) model, which is good at processing time-series data with RaGAN, and uses the weighted loss function to construct a noise reduction model suitable for radio communication signals, which realizes the end-to-end denoising of radio signals. The experimental results show that, compared with the existing methods, the proposed algorithm has significantly improved the noise reduction effect. In the case of a low signal-to-noise ratio (SNR), the signal modulation recognition accuracy is improved by about 10% after noise reduction.

Signal Modulation Recognition System Based on Different Signal Noise Rate Using Artificial Intelligent Approach

Everyone has paid much attention to modulation-type recognition in the past few years. There are many ways to find the modulation type, but only a few good ways to deal with signals with a lot of noise. This study comes up with a way to test how well different machine learning algorithms can handle noise when detecting digital and analogue modulations. This study looks at the four most common digital and analogue modulations: Phase Shift Keying, Quadrature Phase Shift Keying, Amplitude Modulation, and Morse Code. A signal noise rate from -10dB to +25dB is used to find these modulations. We used machine learning algorithms to determine the modulation type like Decision Tree, Random Forest, Support Vectors Machine, and k-nearest neighbours. After the IQ samples had been converted to the amplitude of samples and radio frequency format, the accuracy of each method looked good. Still, in the format of the sample phase, each algorithm's accuracy was less. The results show that the proposed method works to find the signals that have noises. When there is less noise, the random forest algorithm gives better results than SVM, but SVM gives better results when there is more noise.