Adaptive feature extraction and fine‐grained modulation recognition of multi‐function radar under small sample conditions

Zhai, Qihang; Li, Yan; Zhang, Zilin; Li, Yunjie; Wang, Shafei

doi:10.1049/rsn2.12273

Cited by 7 publications

(6 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the task of radar mode recognition, we should pay more attention to the functional differences generated by the signals, rather than the differences between the signals. Therefore, we learn from the authors of [ 28 , 36 ], to encode the radar information, which maps the radar performance indicators into the machine learning model. An integrated recognition framework is used to integrate the advantages of data-driven and knowledge-driven methods.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Working Mode Recognition of Non-Specific Radar Based on ResNet-SVM Learning Framework

Pan

Xiong

Zhuo

2023

Sensors

View full text Add to dashboard Cite

Mode recognition is a basic task to interpret the behavior of multi-functional radar. The existing methods need to train complex and huge neural networks to improve the recognition ability, and it is difficult to deal with the mismatch between the training set and the test set. In this paper, a learning framework based on residual neural network (ResNet) and support vector machine (SVM) is designed, to solve the problem of mode recognition for non-specific radar, called multi-source joint recognition framework (MSJR). The key idea of the framework is to embed the prior knowledge of radar mode into the machine learning model, and combine the manual intervention and automatic extraction of features. The model can purposefully learn the feature representation of the signal on the working mode, which weakens the impact brought by the mismatch between training and test data. In order to solve the problem of difficult recognition under signal defect conditions, a two-stage cascade training method is designed, to give full play to the data representation ability of ResNet and the high-dimensional feature classification ability of SVM. Experiments show that the average recognition rate of the proposed model, with embedded radar knowledge, is improved by 33.7% compared with the purely data-driven model. Compared with other similar state-of-the-art reported models, such as AlexNet, VGGNet, LeNet, ResNet, and ConvNet, the recognition rate is increased by 12%. Under the condition of 0–35% leaky pulses in the independent test set, MSJR still has a recognition rate of more than 90%, which also proves its effectiveness and robustness in the recognition of unknown signals with similar semantic characteristics.

show abstract

Section: Related Workmentioning

confidence: 99%

“…The previous research on behavior pattern recognition [ 36 , 39 , 40 ] mainly considered a single application scenario. The recognition model was established under the condition that the radar model was specific, as well as the fluctuation range of the signal was small, so it had little scalability.…”

Section: Problem Formulationmentioning

confidence: 99%

Working Mode Recognition of Non-Specific Radar Based on ResNet-SVM Learning Framework

Pan

Xiong

Zhuo

2023

Sensors

View full text Add to dashboard Cite

show abstract

“…Few-shot learning (FSL) methods can perform DL-based AMC tasks on small-scale datasets, which has been in the spotlight [20][21][22][23][24][25]. FSL, a learning paradigm inspired by biological systems, aims to surpass the limitations of conventional DL networks regarding their capacity for generalization and adaptability across diverse scenarios.…”

Section: Introductionmentioning

confidence: 99%

Prototypical Network with Residual Attention for Modulation Classification of Wireless Communication Signals

Zang,

Gou,

Zhu

et al. 2023

Electronics

View full text Add to dashboard Cite

Automatic modulation classification (AMC) based on data-driven deep learning (DL) can achieve excellent classification performance. However, in the field of electronic countermeasures, it is difficult to extract salient features from wireless communication signals under scarce samples. Aiming at the problem of modulation classification under scarce samples, this paper proposes a few-shot learning method using prototypical network (PN) with residual attention (RA), namely PNRA, to achieve the AMC. Firstly, the RA is utilized to extract the feature vector of wireless communication signals. Subsequently, the feature vector is mapped to a new feature space. Finally, the PN is utilized to measure the Euclidean distance between the feature vector of the query point and each prototype in this space, determining the type of the signals. In comparison to mainstream few-shot learning (FSL) methods, the proposed PNRA can achieve effective and robust AMC under the data-hungry condition.

show abstract

“…However, GAN is mainly used for SAR images rather than general radar signals. In literature [28], a few shot learning neural network is proposed for adaptive feature extraction and fine‐grained modulation recognition of multi‐function radar with small samples. The network can automatically select the most appropriate data parameters, but the recognition performance of different signal features has not been studied.…”

Section: Introductionmentioning

confidence: 99%

“…The recognition rate and robustness of deep learning methods depend heavily on sufficient training samples, while it is quite difficult to obtain much non-cooperative radar waveform data in practice [27]. A controversial and challenging issue in the field of radar waveform recognition is the recognition with small samples [28]. Currently, a mainstream method to solve the issue of insufficient training samples is to use deep learning generative adversarial network (GAN) to enhance the data of radar signal feature images [29][30][31][32].…”

Section: Introductionmentioning

confidence: 99%

A radar waveform recognition method based on ambiguity function generative adversarial network data enhancement under the condition of small samples

Wang

Xu³

et al. 2022

IET Radar Sonar & Navi

View full text Add to dashboard Cite

This study proposes a small sample recognition method for radar waveforms based on data enhancement of ambiguity function generative adversarial networks (AFGAN) to address the issue of low recognition rate and unbalanced class recognition rate. First, the concept of ambiguity function efficient contour lines (AFECL) and AFECL subject resolution constant are proposed in order to effectively extract radar waveform features. Based on this, the AFGAN model is proposed for radar waveform data enhancement. Simulation experiments classified and identified six types of radar waveforms, which verifies the effectiveness of the model. The experimental results show that the overall recognition rate is increased by 18.64%–99.25% and the class recognition unbalance is reduced by 40.67%–1.33% under the condition of small samples. Compared with short‐time Fourier transform (STFT) and Wigner‐Ville distribution (WVD), the proposed AFECL features are more suitable for recognition in small samples, and the AFGAN model performs significantly better than that of the classical data enhancement models GAN and general WGAN‐GP. The signal recognition rate reached 94.28% when the SNR was −8 dB. The effectiveness of the AFGAN data enhancement method is verified using comparative experiments.

show abstract

Adaptive feature extraction and fine‐grained modulation recognition of multi‐function radar under small sample conditions

Cited by 7 publications

References 29 publications

Working Mode Recognition of Non-Specific Radar Based on ResNet-SVM Learning Framework

Working Mode Recognition of Non-Specific Radar Based on ResNet-SVM Learning Framework

Prototypical Network with Residual Attention for Modulation Classification of Wireless Communication Signals

A radar waveform recognition method based on ambiguity function generative adversarial network data enhancement under the condition of small samples

Contact Info

Product

Resources

About