Deyuan Chen
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
Deyuan Chen has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Guofang Tu
School of Electronic, Electrical and Communication Engineering, University of Chinese Academy of Sciences, Beijing 101408, China
2 shared publications - Shaoshuai Gao
School of Electronic, Electrical and Communication Engineering, University of Chinese Academy of Sciences, Beijing, China
2 shared publications Baobin Li
School of Electronic, Electrical and Communication Engineering, University of Chinese Academy of Sciences, Beijing, China
1 shared publicationDengjie Yang
School of Electronic, Electrical and Communication Engineering,University of Chinese Academy of Sciences, Beijing, China
1 shared publication- Mengyu Zhu
School of Electronic, Electrical and Communication Engineering, University of Chinese Academy of Sciences, Beijing 101408, China
1 shared publication Qing He
Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China
1 shared publication
Latest Publications
- Open Access
Multi-Access Edge Computing (MEC) Based on MIMO: A Survey
With the rapid development of wireless communication technology and the emergence of intelligent applications, higher requirements have been put forward for data communication and computing capacity. Multi-access edge computing (MEC) can handle highly demanding applications by users by sinking the services and computing capabilities of the cloud to the edge of the cell. Meanwhile, the multiple input multiple output (MIMO) technology based on large-scale antenna arrays can achieve an order-of-magnitude improvement in system capacity. The introduction of MIMO into MEC takes full advantage of the energy and spectral efficiency of MIMO technology, providing a new computing paradigm for time-sensitive applications. In parallel, it can accommodate more users and cope with the inevitable trend of continuous data traffic explosion. In this paper, the state-of-the-art research status in this field is investigated, summarized and analyzed. Specifically, we first summarize a multi-base station cooperative mMIMO-MEC model that can easily be expanded to adapt to different MIMO-MEC application scenarios. Subsequently, we comprehensively analyze the current works, compare them to each other and summarize them, mainly from four aspects: research scenarios, application scenarios, evaluation indicators and research issues, and research algorithms. Finally, some open research challenges are identified and discussed, and these indicate the direction for future research on MIMO-MEC.
Open AccessMulti-Access Edge Computing (MEC) Based on MIMO: A Survey
With the rapid development of wireless communication technology and the emergence of intelligent applications, higher requirements have been put forward for data communication and computing capacity. Multi-access edge computing (MEC) can handle highly demanding applications by users by sinking the services and computing capabilities of the cloud to the edge of the cell. Meanwhile, the multiple input multiple output (MIMO) technology based on large-scale antenna arrays can achieve an order-of-magnitude improvement in system capacity. The introduction of MIMO into MEC takes full advantage of the energy and spectral efficiency of MIMO technology, providing a new computing paradigm for time-sensitive applications. In parallel, it can accommodate more users and cope with the inevitable trend of continuous data traffic explosion. In this paper, the state-of-the-art research status in this field is investigated, summarized and analyzed. Specifically, we first summarize a multi-base station cooperative mMIMO-MEC model that can easily be expanded to adapt to different MIMO-MEC application scenarios. Subsequently, we comprehensively analyze the current works, compare them to each other and summarize them, mainly from four aspects: research scenarios, application scenarios, evaluation indicators and research issues, and research algorithms. Finally, some open research challenges are identified and discussed, and these indicate the direction for future research on MIMO-MEC.
- Open Access
- 1 February 2023
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 2425 (1) , 012011
EKConv: Compressing Convolutional Neural Networks with Evolutionary Kernel Convolution
Convolutional neural networks (CNNs) have achieved tremendous success in visual recognition tasks but mainly rely on massive learnable parameters. To solve this problem, many effective and efficient convolution operators have been proposed, such as group-wise convolution, point-wise convolution, and depth-wise convolution. However, the above convolution operations model and optimize the weight relationship within the same convolutional layer. To reduce the network parameters, we explicitly construct the relationship between convolution kernels of adjacent layers. Specifically, we propose an evolutionary kernel convolution, namely EKConv, to generate weight parameters by group-wise convolution efficiently. In particular, EKConv makes the kernel parameters of the current convolutional layer inherit from its preceding adjacent kernel, which promotes the information exchange between convolution kernels. More importantly, EKConv is a novel plug-and-play module that can be easily embedded into mainstream CNNs. Extensive experimental results show that EKConv can compress the parameters of CNNs by a large margin yet barely sacrifice image classification performance.
- 1 February 2023
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 2425 (1) , 012011
Open AccessEKConv: Compressing Convolutional Neural Networks with Evolutionary Kernel Convolution
Convolutional neural networks (CNNs) have achieved tremendous success in visual recognition tasks but mainly rely on massive learnable parameters. To solve this problem, many effective and efficient convolution operators have been proposed, such as group-wise convolution, point-wise convolution, and depth-wise convolution. However, the above convolution operations model and optimize the weight relationship within the same convolutional layer. To reduce the network parameters, we explicitly construct the relationship between convolution kernels of adjacent layers. Specifically, we propose an evolutionary kernel convolution, namely EKConv, to generate weight parameters by group-wise convolution efficiently. In particular, EKConv makes the kernel parameters of the current convolutional layer inherit from its preceding adjacent kernel, which promotes the information exchange between convolution kernels. More importantly, EKConv is a novel plug-and-play module that can be easily embedded into mainstream CNNs. Extensive experimental results show that EKConv can compress the parameters of CNNs by a large margin yet barely sacrifice image classification performance.
- Open Access
- 27 December 2022
- Published by MDPI AG in International Journal of Environmental Research and Public Health
- Vol. 20 (1) , 466
Suicide Possibility Scale Detection via Sina Weibo Analytics: Preliminary Results
Suicide, as an increasingly prominent social problem, has attracted widespread social attention in the mental health field. Traditional suicide clinical assessment and risk questionnaires lack timeliness and proactivity, and high-risk groups often conceal their intentions, which is not conducive to early suicide prevention. In this study, we used machine-learning algorithms to extract text features from Sina Weibo data and built a suicide risk-prediction model to predict four dimensions of the Suicide Possibility Scale—hopelessness, suicidal ideation, negative self-evaluation, and hostility—all with model validity of 0.34 or higher. Through this method, we can detect the symptoms of suicidal ideation in a more detailed way and improve the proactiveness and accuracy of suicide risk prevention and control.
- 27 December 2022
- Published by MDPI AG in International Journal of Environmental Research and Public Health
- Vol. 20 (1) , 466
Open AccessSuicide Possibility Scale Detection via Sina Weibo Analytics: Preliminary Results
Suicide, as an increasingly prominent social problem, has attracted widespread social attention in the mental health field. Traditional suicide clinical assessment and risk questionnaires lack timeliness and proactivity, and high-risk groups often conceal their intentions, which is not conducive to early suicide prevention. In this study, we used machine-learning algorithms to extract text features from Sina Weibo data and built a suicide risk-prediction model to predict four dimensions of the Suicide Possibility Scale—hopelessness, suicidal ideation, negative self-evaluation, and hostility—all with model validity of 0.34 or higher. Through this method, we can detect the symptoms of suicidal ideation in a more detailed way and improve the proactiveness and accuracy of suicide risk prevention and control.
- Open Access
- 2 August 2022
- Published by Frontiers Media SA in Frontiers in Behavioral Neuroscience
- Vol. 16, 901568
Reliability and validity analysis of personality assessment model based on gait video
Personality affects an individual’s academic achievements, occupational tendencies, marriage quality and physical health, so more convenient and objective personality assessment methods are needed. Gait is a natural, stable, and easy-to-observe body movement that is closely related to personality. The purpose of this paper is to propose a personality assessment model based on gait video and evaluate the reliability and validity of the multidimensional model. This study recruited 152 participants and used cameras to record their gait videos. Each participant completed a 44-item Big Five Inventory (BFI-44) assessment. We constructed diverse static and dynamic time-frequency features based on gait skeleton coordinates, interframe differences, distances between joints, angles between joints, and wavelet decomposition coefficient arrays. We established multidimensional personality trait assessment models through machine learning algorithms and evaluated the criterion validity, split-half reliability, convergent validity, and discriminant validity of these models. The results showed that the reliability and validity of the Gaussian process regression (GPR) and linear regression (LR) models were best. The mean values of their criterion validity were 0.478 and 0.508, respectively, and the mean values of their split-half reliability were all greater than 0.8. In the formed multitrait-multimethod matrix, these methods also had higher convergent and discriminative validity. The proposed approach shows that gait video can be effectively used to evaluate personality traits, providing a new idea for the formation of convenient and non-invasive personality assessment methods.
- 2 August 2022
- Published by Frontiers Media SA in Frontiers in Behavioral Neuroscience
- Vol. 16, 901568
Open AccessReliability and validity analysis of personality assessment model based on gait video
Personality affects an individual’s academic achievements, occupational tendencies, marriage quality and physical health, so more convenient and objective personality assessment methods are needed. Gait is a natural, stable, and easy-to-observe body movement that is closely related to personality. The purpose of this paper is to propose a personality assessment model based on gait video and evaluate the reliability and validity of the multidimensional model. This study recruited 152 participants and used cameras to record their gait videos. Each participant completed a 44-item Big Five Inventory (BFI-44) assessment. We constructed diverse static and dynamic time-frequency features based on gait skeleton coordinates, interframe differences, distances between joints, angles between joints, and wavelet decomposition coefficient arrays. We established multidimensional personality trait assessment models through machine learning algorithms and evaluated the criterion validity, split-half reliability, convergent validity, and discriminant validity of these models. The results showed that the reliability and validity of the Gaussian process regression (GPR) and linear regression (LR) models were best. The mean values of their criterion validity were 0.478 and 0.508, respectively, and the mean values of their split-half reliability were all greater than 0.8. In the formed multitrait-multimethod matrix, these methods also had higher convergent and discriminative validity. The proposed approach shows that gait video can be effectively used to evaluate personality traits, providing a new idea for the formation of convenient and non-invasive personality assessment methods.
- 23 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Letters
- Vol. 25 (9) , 3022-3026
Group-Based LED Selection for Generalized Spatial Modulation in Visible Light Communication
Visible light communication (VLC) systems have attracted considerable attention in recent years. However, it still faces the challenges of user mobility and random device orientation, which will result in unreliable connection and performance degradation. In this letter, we propose a group-based LED selection aided generalized spatial modulation (GLS-GSM) scheme to tackle this issue by preventing some poor LEDs from communication. Firstly, the LEDs are divided into several equal-sized groups. Then, each group conducts Euclidean distance-based LED selection (ED-LS) and spatial modulation (SM) independently. Moreover, an ED-based grouping criterion is exploited to search for an optimal grouping strategy. The bit error rate (BER) performance of the proposed scheme is evaluated for sitting and walking activities. Numerical simulation results demonstrate the effectiveness of the proposed method.
- 23 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Letters
- Vol. 25 (9) , 3022-3026
Group-Based LED Selection for Generalized Spatial Modulation in Visible Light Communication
Visible light communication (VLC) systems have attracted considerable attention in recent years. However, it still faces the challenges of user mobility and random device orientation, which will result in unreliable connection and performance degradation. In this letter, we propose a group-based LED selection aided generalized spatial modulation (GLS-GSM) scheme to tackle this issue by preventing some poor LEDs from communication. Firstly, the LEDs are divided into several equal-sized groups. Then, each group conducts Euclidean distance-based LED selection (ED-LS) and spatial modulation (SM) independently. Moreover, an ED-based grouping criterion is exploited to search for an optimal grouping strategy. The bit error rate (BER) performance of the proposed scheme is evaluated for sitting and walking activities. Numerical simulation results demonstrate the effectiveness of the proposed method.