Wenjie Xu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Wenjie Xu has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Shichao Yi
Jiangsu University of Science and Technology
6 shared publications - Haitao Zou
School of Computer, Jiangsu University of Science and Technology, Zhenjiang, China
1 shared publication - Mindong Liu
School of Naval Architecture and Ocean Engineering, Jiangsu University of Science and Technology, Zhenjiang 212003, China
1 shared publication - Siyu Wu
School of Software, Nanjing University of Information Science and Technology, Nanjing, China
1 shared publication - Xiaofang Guo
School of Computer, Jiangsu University of Science and Technology, Zhenjiang, China
1 shared publication - Yifan Zheng
School of Computer, Jiangsu University of Science and Technology, Zhenjiang, China
1 shared publication
Latest Publications
- 21 September 2026
- Published by Elsevier in Digital Signal Processing
- Vol. 184
Two-stage DOA estimation with LSTM-CNN under impulsive noise
- 21 September 2026
- Published by Elsevier in Digital Signal Processing
- Vol. 184
Two-stage DOA estimation with LSTM-CNN under impulsive noise
- 7 March 2026
- Published by Springer Nature in Circuits, Systems, and Signal Processing
- Vol. 45 (9) , 1-24
A Simple Method for Accelerating Neural Network Convergence and Improving Accuracy in DOA Estimation
- 7 March 2026
- Published by Springer Nature in Circuits, Systems, and Signal Processing
- Vol. 45 (9) , 1-24
A Simple Method for Accelerating Neural Network Convergence and Improving Accuracy in DOA Estimation
- Open Access
A GAN-CNN Fusion Framework for Deep Learning-Based DOA Estimation in Low-SNR Environments
Direction of Arrival (DOA) estimation faces significant performance degradation under low Signal-to-Noise Ratio (SNR) conditions, where traditional algorithms and deep learning models struggle due to corrupted spatial information and limited training data. To address these challenges, this paper introduces a novel two-stage framework that integrates a Generative Adversarial Network (GAN) for signal enhancement with a complex-valued Convolutional Neural Network (CNN) for DOA estimation. The proposed GAN incorporates an attention mechanism and a dedicated phase-consistent loss function to suppress noise while preserving spatial phase information critical for accurate direction finding. Enhanced signals are transformed into covariance matrices and processed by a complex-valued CNN designed to extract robust spatial features. Extensive experiments demonstrate that the proposed method achieves a DOA accuracy of 72.2% and a Root Mean Square Error (RMSE) of at —10 dB SNR with 500 snapshots, substantially outperforming conventional and deep learning baselines. The framework also shows strong robustness to limited data, maintaining 93.8% accuracy with only 50 snapshots. The framework offers a practical solution for reliable DOA estimation in low-SNR and data-scarce environments.Open AccessA GAN-CNN Fusion Framework for Deep Learning-Based DOA Estimation in Low-SNR Environments
Direction of Arrival (DOA) estimation faces significant performance degradation under low Signal-to-Noise Ratio (SNR) conditions, where traditional algorithms and deep learning models struggle due to corrupted spatial information and limited training data. To address these challenges, this paper introduces a novel two-stage framework that integrates a Generative Adversarial Network (GAN) for signal enhancement with a complex-valued Convolutional Neural Network (CNN) for DOA estimation. The proposed GAN incorporates an attention mechanism and a dedicated phase-consistent loss function to suppress noise while preserving spatial phase information critical for accurate direction finding. Enhanced signals are transformed into covariance matrices and processed by a complex-valued CNN designed to extract robust spatial features. Extensive experiments demonstrate that the proposed method achieves a DOA accuracy of 72.2% and a Root Mean Square Error (RMSE) of at —10 dB SNR with 500 snapshots, substantially outperforming conventional and deep learning baselines. The framework also shows strong robustness to limited data, maintaining 93.8% accuracy with only 50 snapshots. The framework offers a practical solution for reliable DOA estimation in low-SNR and data-scarce environments. - 29 January 2026
- Published by Springer Nature in International Journal of Machine Learning and Cybernetics
- Vol. 17 (2) , 54
LRCN-STAN: micro-expression recognition via dual-stream LRCN-enhanced spatiotemporal attention network
- 29 January 2026
- Published by Springer Nature in International Journal of Machine Learning and Cybernetics
- Vol. 17 (2) , 54
LRCN-STAN: micro-expression recognition via dual-stream LRCN-enhanced spatiotemporal attention network
- Open Access
DACL-Net: A Dual-Branch Attention-Based CNN-LSTM Network for DOA Estimation
While deep learning methods are increasingly applied in the field of DOA estimation, existing approaches generally feed the real and imaginary parts of the covariance matrix directly into neural networks without optimizing the input features, which prevents classical attention mechanisms from improving accuracy. This paper proposes a spatio-temporal fusion model named DACL-Net for DOA estimation. The spatial branch applies a two-dimensional Fourier transform (2D-FT) to the covariance matrix, causing angles to appear as peaks in the magnitude spectrum. This operation transforms the original covariance matrix into a dark image with bright spots, enabling the convolutional neural network (CNN) to focus on the bright-spot components via an attention module. Additionally, a spectrum attention mechanism (SAM) is introduced to enhance the extraction of temporal features in the time branch. The model learns simultaneously from two data branches and finally outputs DOA results through a linear layer. Simulation results demonstrate that DACL-Net outperforms existing algorithms in terms of accuracy, achieving an RMSE of 0.04° at an SNR of 0 dB.
Open AccessDACL-Net: A Dual-Branch Attention-Based CNN-LSTM Network for DOA Estimation
While deep learning methods are increasingly applied in the field of DOA estimation, existing approaches generally feed the real and imaginary parts of the covariance matrix directly into neural networks without optimizing the input features, which prevents classical attention mechanisms from improving accuracy. This paper proposes a spatio-temporal fusion model named DACL-Net for DOA estimation. The spatial branch applies a two-dimensional Fourier transform (2D-FT) to the covariance matrix, causing angles to appear as peaks in the magnitude spectrum. This operation transforms the original covariance matrix into a dark image with bright spots, enabling the convolutional neural network (CNN) to focus on the bright-spot components via an attention module. Additionally, a spectrum attention mechanism (SAM) is introduced to enhance the extraction of temporal features in the time branch. The model learns simultaneously from two data branches and finally outputs DOA results through a linear layer. Simulation results demonstrate that DACL-Net outperforms existing algorithms in terms of accuracy, achieving an RMSE of 0.04° at an SNR of 0 dB.
- 24 September 2025
- Published by Springer Nature in Circuits, Systems, and Signal Processing
- Vol. 45 (4) , 2806-2824
DOA Estimation Using Complex-Valued Neural Networks with Generative Probability Wave
- 24 September 2025
- Published by Springer Nature in Circuits, Systems, and Signal Processing
- Vol. 45 (4) , 2806-2824
DOA Estimation Using Complex-Valued Neural Networks with Generative Probability Wave
- Open Access
- 31 January 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (3) , 1473
Spectrum Attention Mechanism-Based Acoustic Vector DOA Estimation Method in the Presence of Colored Noise
In the field of direction of arrival (DOA) estimation, a common assumption is that array noise follows a uniform Gaussian white noise model. However, practical systems often encounter non-ideal noise conditions, such as non-uniform or colored noise, due to sensor characteristics and external environmental factors. Traditional DOA estimation techniques experience significant performance degradation in the presence of colored noise, necessitating the exploration of specialized DOA estimation methods for such environments. This study introduces a DOA estimation method for acoustic vector arrays based on a spectrum attention mechanism (SAM). By employing SAM as an adaptive filter and constructing a double-branch model combining a convolutional neural network (CNN) and long short-term memory (LSTM), the method extracts the spatial and temporal features of signals, and effectively reduces the frequency components of colored noise, enhancing DOA estimation accuracy in colored noise scenarios. At an SNR of −5 dB, it achieves an accuracy rate of 85% while maintaining a low RMSE of only 2.03°.
- 31 January 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (3) , 1473
Open AccessSpectrum Attention Mechanism-Based Acoustic Vector DOA Estimation Method in the Presence of Colored Noise
In the field of direction of arrival (DOA) estimation, a common assumption is that array noise follows a uniform Gaussian white noise model. However, practical systems often encounter non-ideal noise conditions, such as non-uniform or colored noise, due to sensor characteristics and external environmental factors. Traditional DOA estimation techniques experience significant performance degradation in the presence of colored noise, necessitating the exploration of specialized DOA estimation methods for such environments. This study introduces a DOA estimation method for acoustic vector arrays based on a spectrum attention mechanism (SAM). By employing SAM as an adaptive filter and constructing a double-branch model combining a convolutional neural network (CNN) and long short-term memory (LSTM), the method extracts the spatial and temporal features of signals, and effectively reduces the frequency components of colored noise, enhancing DOA estimation accuracy in colored noise scenarios. At an SNR of −5 dB, it achieves an accuracy rate of 85% while maintaining a low RMSE of only 2.03°.