Shuangyu Xu

Shuangyu Xu

Auto-generated profile, based on publicly available information

Shuangyu Xu

School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu, China

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
19 December 2023 in IEEE Sensors Journal
Verified
Journal Article
Closed access
SXShuangyu Xu
Shuangyu Xu
ZWZhihang Wang
Zhihang Wang
WZWei Zhang
Wei Zhang
ZHZishu He
Zishu He
In this paper, a regression neural network is proposed instead of the commonly used classification neural network to address the direction of arrival (DOA) estimation problem. Unlike the method treating the DOA estimation problem as a classification task, our proposed method utilizes a regression neural network to directly estimate the signal angles by mapping the input data to the high-dimensional features. It is an end-to-end method where the input data is the signal covariance matrix estimated from the received signals by the uniform linear array. The network output consists of two branches: one branch provides the estimated angles of signals, and the other branch provides the confidences corresponding to angles, indicating the probabilities that signals are incident at the estimated angles. Owing to the dual-branch output of our proposed method, it is adept at accommodating a variable number of sources without the requirement to alter the network structure. Compared to classification neural networks, our proposed method circumvents the issue of quantization errors and has the capability to handle the imbalance of positive and negative samples. Simulation results validate the superior performance of our proposed method in terms of estimation success rate, angular resolution, and estimation accuracy, especially in scenarios with a variable and unknown number of coherent signals.

13 August 2025 in IEEE Sensors Journal
Verified
Journal Article
Closed access
SXShuangyu Xu
Shuangyu Xu
ZWZhihang Wang
Zhihang Wang
ZHZishu He
Zishu He
Detecting small targets within intricate sea clutter presents a formidable challenge. In previous methods, convolutional neural network (CNN)-based detectors have relied on handcrafted features extracted through the manual data processing, which may not fully capture the discriminative features necessary to distinguish targets from sea clutter. This article introduces a novel method of target detection that utilizes CNN-based data processing to directly handle raw data. The proposed CNN-based data processing can automatically extract higher level features from signals, which are often more discriminative and valuable for subsequent detection. The two-stage design of our method allows for the easy replacement of more advanced CNN-based detectors in future applications, providing flexibility for future improvements. Experimental results demonstrate that our method achieves probabilities of detection (PDs) of 0.9008 and 0.8433 on the IPIX and SDRDSP datasets, respectively, with a probability of false alarm (PFA) of 0.001, substantially surpassing other methods. The total FLOPs of our method are 206.42M, making it suitable for real-time applications. Further experiments confirm that our proposed CNN-based data processing can enhance various CNN-based detectors across different datasets, showcasing robustness and effectiveness. Moreover, our method maintains high detection performance even with a limited number of pulses.