A High Resolution Convolutional Neural Network with Squeeze and Excitation Module for Automatic Modulation Classification

Automatic modulation classification(AMC) technology is one of the cutting-edge technologies in cognitive radio communications. AMC based on deep learning has recently attracted much attention due to its superior performances in classification accuracy and robustness. In this paper, we propose a novel, high resolution and multi-scale feature fusion convolutional neural network model with a squeeze-excitation block, referred to as HRSENet,to classify different kinds of modulation signals.The proposed model establishes a parallel computing mechanism of multi-resolution feature maps through the multi-layer convolution operation, which effectively reduces the information loss caused by downsampling convolution. Moreover, through dense skipconnecting at the same resolution and up-sampling or down-sampling connection at different resolutions, the low resolution representation of the deep feature maps and the high resolution representation of the shallow feature maps are simultaneously extracted and fully integrated, which is benificial to mine signal multilevel features. Finally, the feature squeeze and excitation module embedded in the decoder is used to adjust the response weights between channels, further improving classification accuracy of proposed model.The proposed HRSENet significantly outperforms existing methods in terms of classification accuracy on the public dataset “Over the Air” in signal-to-noise(SNR) ranging from-2dB to 20dB. The classification accuracy in the proposed model achieves 85.36% and97.30% at 4dB and 10dB, respectively, with the improvement by 9.71% and 5.82% compared to LWNet.Furthermore, the model also has a moderate computation complexity compared with several state-of-the-art methods.

A Few-Shot Learning-Based Automatic Modulation Classification Method for Internet of Things

Due to the limited computational capability and the diversity of the Internet of Things devices working in different environment,we consider fewshot learning-based automatic modulation classification(AMC)to improve its reliability.A data enhancement module(DEM)is designed by a convolutional layer to supplement frequency-domain information as well as providing nonlinear mapping that is beneficial for AMC.Multimodal network is designed to have multiple residual blocks,where each residual block has multiple convolutional kernels of different sizes for diverse feature extraction.Moreover,a deep supervised loss function is designed to supervise all parts of the network including the hidden layers and the DEM.Since different model may output different results,cooperative classifier is designed to avoid the randomness of single model and improve the reliability.Simulation results show that this few-shot learning-based AMC method can significantly improve the AMC accuracy compared to the existing methods.

Deep learning algorithm featuring continuous learning for modulation classifications in wireless networks

Although modulation classification based on deep neural network can achieve high Modulation Classification(MC)accuracies,catastrophic forgetting will occur when the neural network model continues to learn new tasks.In this paper,we simulate the dynamic wireless communication environment and focus on breaking the learning paradigm of isolated automatic MC.We innovate a research algorithm for continuous automatic MC.Firstly,a memory for storing representative old task modulation signals is built,which is employed to limit the gradient update direction of new tasks in the continuous learning stage to ensure that the loss of old tasks is also in a downward trend.Secondly,in order to better simulate the dynamic wireless communication environment,we employ the mini-batch gradient algorithm which is more suitable for continuous learning.Finally,the signal in the memory can be replayed to further strengthen the characteristics of the old task signal in the model.Simulation results verify the effectiveness of the method.

A Convolutional and Transformer Based Deep Neural Network for Automatic Modulation Classification

Automatic modulation classification(AMC)aims at identifying the modulation of the received signals,which is a significant approach to identifying the target in military and civil applications.In this paper,a novel data-driven framework named convolutional and transformer-based deep neural network(CTDNN)is proposed to improve the classification performance.CTDNN can be divided into four modules,i.e.,convolutional neural network(CNN)backbone,transition module,transformer module,and final classifier.In the CNN backbone,a wide and deep convolution structure is designed,which consists of 1×15 convolution kernels and intensive cross-layer connections instead of traditional 1×3 kernels and sequential connections.In the transition module,a 1×1 convolution layer is utilized to compress the channels of the previous multi-scale CNN features.In the transformer module,three self-attention layers are designed for extracting global features and generating the classification vector.In the classifier,the final decision is made based on the maximum a posterior probability.Extensive simulations are conducted,and the result shows that our proposed CTDNN can achieve superior classification performance than traditional deep models.

Incremental Learning of Radio Modulation Classification Based on Sample Recall

Radio modulation classification has always been an important technology in the field of communications.The difficulty of incremental learning in radio modulation classification is that learning new tasks will lead to catastrophic forgetting of old tasks.In this paper,we propose a sample memory and recall framework for incremental learning of radio modulation classification.For data with different signal-to-noise ratios,we use a partial memory strategy by selecting appropriate samples for memorizing.We compare the performance of our proposed method with three baselines through a large number of simulation experiments.Results show that our method achieves far higher classification accuracy than finetuning method and feature extraction method.Furthermore,it performs closely to joint training method which uses all old data in terms of classification accuracy which validates the effectiveness of our method against catastrophic forgetting.

Automatic modulation classification using modulation fingerprint extraction

An automatic method for classifying frequency shift keying(FSK),minimum shift keying(MSK),phase shift keying(PSK),quadrature amplitude modulation(QAM),and orthogonal frequency division multiplexing(OFDM)is proposed by simultaneously using normality test,spectral analysis,and geometrical characteristics of in-phase-quadrature(I-Q)constellation diagram.Since the extracted features are unique for each modulation,they can be considered as a fingerprint of each modulation.We show that the proposed algorithm outperforms the previously published methods in terms of signal-to-noise ratio(SNR)and success rate.For example,the success rate of the proposed method for 64-QAM modulation at SNR=11 dB is 99%.Another advantage of the proposed method is its wide SNR range;such that the probability of classification for 16-QAM at SNR=3 dB is almost 1.The proposed method also provides a database for geometrical features of I-Q constellation diagram.By comparing and correlating the data of the provided database with the estimated I-Q diagram of the received signal,the processing gain of 4 dB is obtained.Whatever can be mentioned about the preference of the proposed algorithm are low complexity,low SNR,wide range of modulation set,and enhanced recognition at higher-order modulations.

Modulation classification of MPSK signals based on nonparametric Bayesian inference

A nonparametric Bayesian method is presented to classify the MPSK （M-ary phase shift keying） signals. The MPSK signals with unknown signal noise ratios （SNRs） are modeled as a Gaussian mixture model with unknown means and covariances in the constellation plane, and a clustering method is proposed to estimate the probability density of the MPSK signals. The method is based on the nonparametric Bayesian inference, which introduces the Dirichlet process as the prior probability of the mixture coefficient, and applies a normal inverse Wishart （NIW） distribution as the prior probability of the unknown mean and covariance. Then, according to the received signals, the parameters are adjusted by the Monte Carlo Markov chain （MCMC） random sampling algorithm. By iterations, the density estimation of the MPSK signals can be estimated. Simulation results show that the correct recognition ratio of 2/4/8PSK is greater than 95% under the condition that SNR 〉5 dB and 1 600 symbols are used in this method.

Semi-Supervised Learning with Generative Adversarial Networks on Digital Signal Modulation Classification

Deep Learning(DL)is such a powerful tool that we have seen tremendous success in areas such as Computer Vision,Speech Recognition,and Natural Language Processing.Since Automated Modulation Classification(AMC)is an important part in Cognitive Radio Networks,we try to explore its potential in solving signal modulation recognition problem.It cannot be overlooked that DL model is a complex model,thus making them prone to over-fitting.DL model requires many training data to combat with over-fitting,but adding high quality labels to training data manually is not always cheap and accessible,especially in real-time system,which may counter unprecedented data in dataset.Semi-supervised Learning is a way to exploit unlabeled data effectively to reduce over-fitting in DL.In this paper,we extend Generative Adversarial Networks(GANs)to the semi-supervised learning will show it is a method can be used to create a more dataefficient classifier.

Tracking performance of large margin classifier in automatic modulation classification with a software radio environment

Automatic modulation classification is the process of identification of the modulation type of a signal in a general environment. This paper proposes a new method to evaluate the tracking performance of large margin classifier against signal-tonoise ratio （SNR）, and classifies all forms of primary user＇s signals in a cognitive radio environment. For achieving this objective, two structures of a large margin are developed in additive white Gaussian noise （AWGN） channels with priori unknown SNR. A combination of higher order statistics and instantaneous characteristics is selected as effective features. Simulation results show that the classification rates of the proposed structures are well robust against environmental SNR changes.

Digital modulation classification using multi-layer perceptron and time-frequency features

Considering that real communication signals corrupted by noise are generally nonstationary, and timefrequency distributions are especially suitable for the analysis of nonstationary signals, time-frequency distributions are introduced for the modulation classification of communication signals： The extracted time-frequency features have good classification information, and they are insensitive to signal to noise ratio （SNR） variation. According to good classification by the correct rate of a neural network classifier, a multilayer perceptron （MLP） classifier with better generalization, as well as, addition of time-frequency features set for classifying six different modulation types has been proposed. Computer simulations show that the MLP classifier outperforms the decision-theoretic classifier at low SNRs, and the classification experiments for real MPSK signals verify engineering significance of the MLP classifier.

Modulation classification based on spectrogram

The aim of modulation classification （MC） is to identify the modulation type of a commtmication signal. It plays an important role in many cooperative or noncooperative communication applications. Three spectrogram-based modulation classification methods are proposed. Their recognition scope and performance are investigated or evaluated by theoretical analysis and extensive simulation studies. The method taking moment-like features is robust to frequency offset while the other two, which make use of principal component analysis （PCA） with different transformation inputs, can achieve satisfactory accuracy even at low SNR （as low as 2 dB）. Due to the properties of spectrogram, the statistical pattern recognition techniques, and the image preprocessing steps, all of our methods are insensitive to unknown phase and frequency offsets, timing errors, and the arriving sequence of symbols.

Realizable Hardware-Based Method for Digital Modulation Classification

A new method suited for hardware implementation is developed to classify 8 different digital modulation types with raised cosine base-band impulse without knowing the carrier frequency and symbol timing. The normalized histogram of stagnation points for instantaneous parameters is used to recognize both ideal rectangular and raised cosine base-band digital signals. Carrier frequency estimation is used to enhance the recognition rate of phase-modulated signals. In the condition of 10 dB signal noise ratio （SNR）, the recognizing rate is over 80% . The new algorithm is suited for hardware implementation.

A NEW LIKELIHOOD-BASED MODULATION CLASSIFICATION ALGORITHM USING MCMC

In this paper,a new likelihood-based method for classifying phase-amplitude-modulated signals in Additive White Gaussian Noise (AWGN) is proposed.The method introduces a new Markov Chain Monte Carlo (MCMC) algorithm,called the Adaptive Metropolis (AM) algorithm,to directly generate the samples of the target posterior distribution and implement the multidimensional integrals of likelihood function.Modulation classification is achieved along with joint estimation of unknown parameters by running an ergodic Markov Chain.Simulation results show that the proposed method has the advantages of high accuracy and robustness to phase and frequency offset.

Automatic modulation classification based on the combination of clustering and neural network

In this paper, we propose a new modulation classification method based on the combination of clustering and neural network, in which a new algorithm is introduced to extract key features. In order to recognize modulation types based on the constellation diagram such as phase shift keying （PSK） and quadrature amplitude modulation （QAM）, fuzzy C-means （FCM） clustering is adopted for recovering the constellation under different number of clusters. Then cluster validity measure is applied to extract key features which discriminate between different modulation types. The features are sent to neural network so that modulation types can be recognized. In order to conquer the disadvantages of standard back propagation （BP） neural network, conjugate gradient learning algorithm of Polak-Ribiere update is employed to improve the speed of convergence and the performance of modulation recognition. Simulation results show that classification rates of the algorithm proposed in this paper are much higher than those of clustering algorithm.

Classification using wavelet packet decomposition and support vector machine for digital modulations

To make the modulation classification system more suitable for signals in a wide range of signal to noise rate （SNR）, a feature extraction method based on signal wavelet packet transform modulus maxima matrix （WPTMMM） and a novel support vector machine fuzzy network （SVMFN） classifier is presented. The WPTMMM feature extraction method has less computational complexity, more stability, and has the preferable advantage of robust with the time parallel moving and white noise. Further, the SVMFN uses a new definition of fuzzy density that incorporates accuracy and uncertainty of the classifiers to improve recognition reliability to classify nine digital modulation types （i.e. 2ASK, 2FSK, 2PSK, 4ASK, 4FSK, 4PSK, 16QAM, MSK, and OQPSK）. Computer simulation shows that the proposed scheme has the advantages of high accuracy and reliability （success rates are over 98% when SNR is not lower than 0dB）, and it adapts to engineering applications.

An effective approach for low-complexity maximum likelihood based automatic modulation classification of STBC-MIMO systems

A low-complexity likelihood methodology is proposed for automatic modulation classification of orthogonal space-time block code(STBC)based multiple-input multiple-output(MIMO)systems.We exploit the zero-forcing equalization technique to modify the typical average likelihood ratio test(ALRT)function.The proposed ALRT function has a low computational complexity compared to existing ALRT functions for MIMO systems classification.The proposed approach is analyzed for blind channel scenarios when the receiver has imperfect channel state information(CSI).Performance analysis is carried out for scenarios with different numbers of antennas.Alamouti-STBC systems with 2×2 and 2×1 and space-time transmit diversity with a 4×4 transmit and receive antenna configuration are considered to verify the proposed approach.Some popular modulation schemes are used as the modulation test pool.Monte-Carlo simulations are performed to evaluate the proposed methodology,using the probability of correct classification as the criterion.Simulation results show that the proposed approach has high classification accuracy at low signal-to-noise ratios and exhibits robust behavior against high CSI estimation error variance.

Automatic modulation classification based on Alex Net with data augmentation

Deep learning(DL) requires massive volume of data to train the network. Insufficient training data will cause serious overfitting problem and degrade the classification accuracy. In order to solve this problem, a method for automatic modulation classification(AMC) using AlexNet with data augmentation was proposed. Three data augmentation methods is considered, i.e., random erasing, CutMix, and rotation. Firstly, modulated signals are converted into constellation representations. And all constellation representations are divided into training dataset and test dataset. Then, training dataset are augmented by three methods. Secondly, the optimal value of execution probability for random erasing and CutMix are determined. Simulation results show that both of them perform optimally when execution probability is 0.5. Thirdly, the performance of three data augmentation methods are evaluated. Simulation results demonstrate that all augmentation methods can improve the classification accuracy. Rotation improves the classification accuracy by 13.04% when signal noise ratio(SNR) is 2 dB. Among three methods, rotation outperforms random erasing and CutMix when SNR is greater than-6 dB. Finally, compared with other classification algorithms, random erasing, CutMix, and rotation used in this paper achieved the performance significantly improved. It is worth mentioning that the classification accuracy can reach 90.5% with SNR at 10 dB.

Modulation classification based on the collaboration of dual-channel CNN-LSTM and residual network

Deep learning has recently been progressively introduced into the field of modulation classification due to its wide application in image, vision, and other areas. Modulation classification is not only the priority of cognitive radio and spectrum sensing, but also the link during signal demodulation. Combining the advantages of convolutional neural network(CNN), long short-term memory(LSTM), and residual network(ResNet), a modulation classification method based on dual-channel CNN-LSTM and ResNet is proposed to automatically classify the modulation signal more accurately. Specifically, CNN and LSTM are initially used to form a dual-channel structure to effectively explore the spatial and temporal features of the original complex signal. It solves the problem of only focusing on temporal or spatial aspects, and increases the diversity of features. Secondly, the features extracted from CNN and LSTM are fused, making the extracted features richer and conducive to signal classification. In addition, a convolutional layer is added within the residual unit to deepen the network depth. As a result, more representative features are extracted, improving the classification performance. Finally, simulation results on the radio machine learning(RadioML) 2018.01 A dataset signify that the network’s classification performance is superior to many classifiers in the literature.

A novel adaptive classification scheme for digital modulations in satellite communication

To make the modulation classification system more suitable for signals in a wide range of signal to noise ratios （SNRs）, a novel adaptive modulation classification scheme is presented in this paper. Differ-ent from traditional schemes, the proposed scheme employs a new SNR estimation algorithm for small samples before modulation classification, which makes the modulation classifier work adaptively according to estimated SNRs. Furthermore, it uses three efficient features and support vector machines （SVM） in modulation classification. Computer simulation shows that the scheme can adaptively classify ten digital modulation types （i.e. 2ASK, 4ASK, 2FSK, 4FSK, 2PSK, 4PSK, 16QAM, TFM, π/4QPSK and OQPSK） at SNRS ranging from 0dB to 25dB and success rates are over 95% when SNR is not lower than 3dB. Accuracy, efficiency and simplicity of the proposed scheme are obviously improved, which make it more adaptive to engineering applications.

Hierarchical Digital Modulation Classification Using Cascaded Convolutional Neural Network

Automatic modulation classification(AMC)aims to identify the modulation format of the received signals corrupted by the noise,which plays a major role in radio monitoring.In this paper,we propose a novel cascaded convolutional neural network(CasCNN)-based hierarchical digital modulation classification scheme,where M-ary phase shift keying(PSK)and M-ary quadrature amplitude modulation(QAM)modulation formats are considered to be classified.In CasCNN,two-block convolutional neural networks are cascaded.The first block network is utilized to classify the different classes of modulation formats,namely PSK and QAM.The second block is designed to identify the indexes of the modulations in the same PSK or QAM class.Moreover,it is noted that the gird constellation diagram extracted from the received signal is utilized as the inputs to the CasCNN.Extensive simulations demonstrate that CasCNN yields performance gain and performs stronger robustness to frequency offset compared with other recent methods.Specifically,CasCNN achieves 90%classification accuracy at 4 dB signal-to-noise ratio when the symbol length is set as 256.