Yu Guo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Yu Guo has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Yuzhen Huang
Academy of Military Sciences of PLA, Beijing, China
4 shared publications - Zhi Zhang
Key Laboratory of Universal Wireless Communications, Ministry of Education, Beijing University of Posts and Telecommunications, Beijing, China
4 shared publications - Ping Zhang
Beijing University of Posts and Telecommunications
2 shared publications - Jianhong Chu
State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications (BUPT), Beijing, China
1 shared publication Peiqi Lin
School of Modern Post, Beijing University of Posts and Telecommunications, Beijing, China
1 shared publicationWanqing Zhang
School of Modern Post, Beijing University of Posts and Telecommunications, Beijing, China
1 shared publication
Latest Publications
- 13 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 31 (10709908) , 76-80
Dual Class Token Vision Transformer for Direction of Arrival Estimation in Low SNR
In this letter, we propose a deep learning-based method for the direction of arrival (DOA) estimation in the low signal-to-noise ratio (SNR) scenario. Specifically, the DOA estimation is modeled as a multi-label classification task, and a novel dual class token Vision Transformer (DCT-ViT) is designed to fit it. Different from the classical ViT architecture with a single class token, the DCT-ViT includes two class tokens which are located at the beginning and end of the latent vector sequence, respectively. This architecture enables enhanced information mining and feature extraction from the array signal data in order to improve the accuracy of DOA estimation. Furthermore, a single DCT-ViT model can accommodate different source numbers by leveraging a training dataset with different numbers of sources. Simulation results illustrate that our proposed method outperforms existing methods in the low SNR scenario, including classical model-based and other deep learning-based methods.
- 13 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 31 (10709908) , 76-80
Dual Class Token Vision Transformer for Direction of Arrival Estimation in Low SNR
In this letter, we propose a deep learning-based method for the direction of arrival (DOA) estimation in the low signal-to-noise ratio (SNR) scenario. Specifically, the DOA estimation is modeled as a multi-label classification task, and a novel dual class token Vision Transformer (DCT-ViT) is designed to fit it. Different from the classical ViT architecture with a single class token, the DCT-ViT includes two class tokens which are located at the beginning and end of the latent vector sequence, respectively. This architecture enables enhanced information mining and feature extraction from the array signal data in order to improve the accuracy of DOA estimation. Furthermore, a single DCT-ViT model can accommodate different source numbers by leveraging a training dataset with different numbers of sources. Simulation results illustrate that our proposed method outperforms existing methods in the low SNR scenario, including classical model-based and other deep learning-based methods.
- 29 October 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Letters
- Vol. 26 (1) , 69-73
A Novel Approach of 2-D DOA Estimation by Employing Coprime Linear Array Motion
In this letter, we explore the feature of coprime linear array (CLA) motion for enhanced array aperture and higher degree of freedom (DOF) by implementing synthetic aperture processing. Since the traditional fixed CLA cannot achieve two-dimensional (2-D) DOA estimation, we propose a novel approach to construct the virtual coprime planar array (VCPA) just by the CLA motion. Specifically, the CLA moves along the vertical direction at a constant velocity and the two subarrays of CLA are controlled to snap at different time delays related to the coprime number of sensors, by which the received signals at different time delays are synthetically processed to construct a VCPA. Simulation results demonstrate that: a) our proposed approach outperforms the synthetic CLA moving along the direction of layout of the initial arrays on DOF and the dimension of DOA estimation; b) the proposed VCPA can achieve 2-D DOA estimation just by the motive CLA and achieve no worse performance than the practical coprime planar array but by using fewer sensors.
- 29 October 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Letters
- Vol. 26 (1) , 69-73
A Novel Approach of 2-D DOA Estimation by Employing Coprime Linear Array Motion
In this letter, we explore the feature of coprime linear array (CLA) motion for enhanced array aperture and higher degree of freedom (DOF) by implementing synthetic aperture processing. Since the traditional fixed CLA cannot achieve two-dimensional (2-D) DOA estimation, we propose a novel approach to construct the virtual coprime planar array (VCPA) just by the CLA motion. Specifically, the CLA moves along the vertical direction at a constant velocity and the two subarrays of CLA are controlled to snap at different time delays related to the coprime number of sensors, by which the received signals at different time delays are synthetically processed to construct a VCPA. Simulation results demonstrate that: a) our proposed approach outperforms the synthetic CLA moving along the direction of layout of the initial arrays on DOF and the dimension of DOA estimation; b) the proposed VCPA can achieve 2-D DOA estimation just by the motive CLA and achieve no worse performance than the practical coprime planar array but by using fewer sensors.
- 3 April 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 27 (10709908) , 570-574
DOA Estimation Method Based on Cascaded Neural Network for Two Closely Spaced Sources
In this letter, we explore the problem of DOA estimation using neural networks for two closely spaced sources. Since the traditional high-resolution techniques based on classical algorithms cannot achieve high-accuracy DOA estimation in the presence of two closely spaced sources, especially at low signal-to-noise ratios (SNR), we propose a novel DOA estimation method based on a cascaded neural network to address this problem. Specifically, this network comprises two parts: the SNR classification network and the DOA estimation network. The latter network contains two estimation subnetworks, which are appropriate for different SNRs by training with noisy data and activated by the output of the SNR classification network. Simulation results demonstrate that the estimation performance of our proposed method achieves much better than that of the existing algorithms under various conditions, especially for the scenes with low SNRs or small snapshot number.
- 3 April 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 27 (10709908) , 570-574
DOA Estimation Method Based on Cascaded Neural Network for Two Closely Spaced Sources
In this letter, we explore the problem of DOA estimation using neural networks for two closely spaced sources. Since the traditional high-resolution techniques based on classical algorithms cannot achieve high-accuracy DOA estimation in the presence of two closely spaced sources, especially at low signal-to-noise ratios (SNR), we propose a novel DOA estimation method based on a cascaded neural network to address this problem. Specifically, this network comprises two parts: the SNR classification network and the DOA estimation network. The latter network contains two estimation subnetworks, which are appropriate for different SNRs by training with noisy data and activated by the output of the SNR classification network. Simulation results demonstrate that the estimation performance of our proposed method achieves much better than that of the existing algorithms under various conditions, especially for the scenes with low SNRs or small snapshot number.
- 4 November 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Systems Journal
- Vol. 15 (1) , 407-410
A 2-D DOA Estimation Method With Reduced Complexity in Unfolded Coprime L-Shaped Array
In this article, we focus on investigating the two-dimensional direction of arrival (2-D DOA) estimation for the unfolded coprime L-shaped array (UCLsA). Due to 2-D spectral peak searching, the computational complexity of the existing 2-D DOA estimation methods based on the multiple signal classification (MUSIC) algorithm is too high. Motivated by this, we propose a novel low-complexity 2-D DOA estimation method based on MUSIC, where the 2-D spectral peak searching is converted to 1-D searching by constructing a transform domain. Moreover, the proposed method makes full use of the received signals to calculate the noise subspace, which can take advantages of the large array aperture and mutual information of the UCLsA and, significantly, improve the estimation accuracy. Simulation results demonstrate that: 1) the proposed method achieves similar performance to the Cramér-Rao bound (CRB); 2) the proposed method outperforms the existing 2-D DOA estimation algorithms; and 3) the implementation cost of the proposed method is lower than those of the existing methods.
- 4 November 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Systems Journal
- Vol. 15 (1) , 407-410
A 2-D DOA Estimation Method With Reduced Complexity in Unfolded Coprime L-Shaped Array
In this article, we focus on investigating the two-dimensional direction of arrival (2-D DOA) estimation for the unfolded coprime L-shaped array (UCLsA). Due to 2-D spectral peak searching, the computational complexity of the existing 2-D DOA estimation methods based on the multiple signal classification (MUSIC) algorithm is too high. Motivated by this, we propose a novel low-complexity 2-D DOA estimation method based on MUSIC, where the 2-D spectral peak searching is converted to 1-D searching by constructing a transform domain. Moreover, the proposed method makes full use of the received signals to calculate the noise subspace, which can take advantages of the large array aperture and mutual information of the UCLsA and, significantly, improve the estimation accuracy. Simulation results demonstrate that: 1) the proposed method achieves similar performance to the Cramér-Rao bound (CRB); 2) the proposed method outperforms the existing 2-D DOA estimation algorithms; and 3) the implementation cost of the proposed method is lower than those of the existing methods.
- 1 July 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 31, 1-6
The Impact of online word-of-mouth on Consumers' Purchase Intention in Social Commerce
Compared with traditional online shopping, consumer decision-making in social commerce is more susceptible to online word-of-mouth which can be published by people with weak relationship (stranger), moderate relationship (opinion leader), and strong relationship (friend). It's meaningful to study how word-of-mouth impacts consumer's purchase intention. In this paper, we designed an S1-S2 paradigm (reviewer review) using Event-Related Potentials(ERPs) technique to investigate this problem. On the basis of the given information (reviewer and review), participants had to decide whether to buy the product. The Late Positive Potential (LPP) component was successfully evoked and the order of LPP amplitudes from large to small was opinion leader-friendstranger. It indicates that participants consume more and more emotional processing resources in this order. The order of conformity rates from high to low was friend-opinion leaderstranger, inconsistent with the order of LPP amplitudes. It shows that emotional resources only play a part in decision-making process. The final decision made by the participants is more relevant to the degree of intimacy. Our findings suggest merchants that they can promote their products through opinion leaders to attract customers' attention and increase the proportion of friends' comments, thereby motivating consumers to make purchase decisions.
- 1 July 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 31, 1-6
The Impact of online word-of-mouth on Consumers' Purchase Intention in Social Commerce
Compared with traditional online shopping, consumer decision-making in social commerce is more susceptible to online word-of-mouth which can be published by people with weak relationship (stranger), moderate relationship (opinion leader), and strong relationship (friend). It's meaningful to study how word-of-mouth impacts consumer's purchase intention. In this paper, we designed an S1-S2 paradigm (reviewer review) using Event-Related Potentials(ERPs) technique to investigate this problem. On the basis of the given information (reviewer and review), participants had to decide whether to buy the product. The Late Positive Potential (LPP) component was successfully evoked and the order of LPP amplitudes from large to small was opinion leader-friendstranger. It indicates that participants consume more and more emotional processing resources in this order. The order of conformity rates from high to low was friend-opinion leaderstranger, inconsistent with the order of LPP amplitudes. It shows that emotional resources only play a part in decision-making process. The final decision made by the participants is more relevant to the degree of intimacy. Our findings suggest merchants that they can promote their products through opinion leaders to attract customers' attention and increase the proportion of friends' comments, thereby motivating consumers to make purchase decisions.