摘要:
In massive multi-user multi-input–multi-output (MU-MIMO) downlink transmission systems, the total power consumption of the base station is mainly attributed to high-resolution digital-to-analog converters (DACs). Therefore, utilizing one-bit DACs can effectively reduce the system power consumption. This work investigates the one-bit DACs precoding problem in massive MU-MIMO systems. Typically, the one-bit DACs precoding problem based on the minimum mean square error criterion is an NP-hard problem, making it challenging to solve. By exploiting the structural properties of nonconvex constraints, the nonconvex optimization problem is transformed into an equivalent continuous-domain mathematical programming problem with equilibrium constraints. The projected gradient method is used for obtaining the quantized precoding signals, and the method is optimized by introducing a step-size adjustment coefficient. Simulation results demonstrate that the improved projected gradient method can effectively reduce the uncoded bit error rate and achieve a performance gain of approximately $\text{2}\,\text{dB}$ with 16 quadrature amplitude modulation signaling. The simulations also demonstrate the better robustness of the suggested method with imperfect channel state information. Specifically, we evaluate the power consumption of the proposed algorithm to that of the zero-forcing precoder with different resolution DACs, illustrating its superior power efficiency. Furthermore, we demonstrate the effectiveness of the proposed algorithm through a theoretical analysis of the convergence, complexity, and exact property of the optimal solution.
作者机构:
[Kang, Xiatao; Li, Ping; Yao, Jiayi] Changsha Univ Sci & Technol, Sch Comp & Commun Engn, Changsha, Peoples R China.;[Li, Chengxi] Xidian Univ, Sch Comp, Xian, Peoples R China.;[Li, Ping] Hunan Prov Key Lab Intelligent Proc Big Data Tran, Changsha, Peoples R China.
会议名称:
16th Asian Conference on Computer Vision (ACCV)
会议时间:
DEC 04-08, 2022
会议地点:
Macao, PEOPLES R CHINA
会议主办单位:
[Kang, Xiatao;Li, Ping;Yao, Jiayi] Changsha Univ Sci & Technol, Sch Comp & Commun Engn, Changsha, Peoples R China.^[Li, Chengxi] Xidian Univ, Sch Comp, Xian, Peoples R China.^[Li, Ping] Hunan Prov Key Lab Intelligent Proc Big Data Tran, Changsha, Peoples R China.
会议论文集名称:
Lecture Notes in Computer Science
关键词:
Deep learning;Reinforcement learning;Network pruning;Pruning before training
摘要:
Pruning on neural networks before training not only compresses the original models, but also accelerates the network training phase, which has substantial application value. The current work focuses on fine-grained pruning, which uses metrics to calculate weight scores for weight screening, and extends from the initial single-order pruning to iterative pruning. Through these works, we argue that network pruning can be summarized as an expressive force transfer process of weights, where the reserved weights will take on the expressive force from the removed ones for the purpose of maintaining the performance of original networks. In order to achieve optimal expressive force scheduling, we propose a pruning scheme before training called Neural Network Panning which guides expressive force transfer through multi-index and multi-process steps, and designs a kind of panning agent based on reinforcement learning to automate processes. Experimental results show that Panning performs better than various available pruning before training methods. Our code is made public at: https://github.com/kangxiatao/RLPanning.
通讯机构:
[Wang, X ] C;Changsha Univ Sci & Technol, Sch Comp & Commun Engn, Changsha 410114, Hunan, Peoples R China.
关键词:
Deep learning;Convolution neural network;Transformer;Hierarchical attention
摘要:
Melanoma is a malignant tumor condition which can be successfully cured with early detection and treatment. Currently, deep learning is one of the research hotspots in automatic diagnosis, with two primary categories of medical image segmentation applications: Convolution Neural Network (CNN) methods such as U-Net and Transformer-based methods. Despite achieving remarkable performance in image segmentation, both above approaches have some inherent shortcomings that cannot be overlooked. The convolution operation in the CNN structure cannot adequately capture global dependencies, leading to a major negative impact on segmentation performance. While the Multi-head Self-Attention mechanism in Transformer can efficiently extract extensive global features, its large computational complexity and lack of local induction bias cannot be ignored. These factors result in a significant reduction in the precision of clinical diagnosis using automated segmentation. To address the aforementioned issues, this paper proposes a U-shaped network structure called PHCU-Net which comprises global feature block, local feature block and dual-branch hierarchical attention mechanism that is designed specifically for the segmentation network of skin lesions. In addition, we argue that the traditional skip connection can be further improved to acquire stronger contexts by simply incorporating convolution attention, so that the decoder can fully collect the global and local feature information. The aforementioned innovative work empowers our model to generate superior skin lesion segmentation results compared to other classical algorithms and deep learning-based models. Extensive experimental results on three open-source datasets (ISIC2017, ISIC2018, and PH2) demonstrate the effectiveness of the proposed PHCU-Net in melanoma lesion segmentation.
摘要:
Independence and sparsity are proved to be two basic features for spatial activations of functional magnetic resonance imaging (fMRI) data, and have shown efficiency in analysis of magnitude-only fMRI data. Since complex-valued fMRI data contains additional brain activity information beyond magnitude-only fMRI data, we propose to incorporate sparsity constraint into complex independent vector analysis (IVA) to take advantages of the two features in analyzing multi-subject complexvalued fMRI data. Specifically, we propose to improve a complexvalued IVA algorithm named AFIVA (adaptive fixed-point IVA) to add a phase sparsity constraint on spatial maps. Based on the cost function of AFIVA, we further implement the phase sparsity constraint using smoothed L0 norm, and utilize noncircularity of spatial maps as well in the second update of phase sparsity to extract meaningful activations. The results from experimental complex-valued fMRI datasets show that the proposed method yields higher accuracy than AFIVA in terms of true positive rates, confirming the advantage of sparsity in de-noising the independent spatial maps.
摘要:
Data classification of users' electricity consumption provides an in-depth analysis for users' electricity consumption status, which plays a vital role in the management and distribution of electric energy. So, some data classification methods have been proposed to solve the classification problem of electricity consumption data. However, plaintext-based data classification may bring about the privacy leakage of electricity consumption data. In this paper, we propose a privacy-preserving classification scheme for electricity consumption data under fog computing-based smart metering system, which is based on convolutional neural network (CNN) model with fully homomorphic method (CKKS). The target of our proposed scheme is to solve the leakage problem of private electricity consumption data during the classification procedure. In our scheme, an improved K-means-based labeling algorithm is constructed to process historical electricity consumption data, which is used as the sample data to train the CNN classification model by cloud server. Also, the fog nodes are only permitted to obtain the related ciphertext parameters of the trained CNN model, and perform the classification of ciphertext-based electricity consumption data generated by fully homomorphic method. Based on the classical testing data, the experimental results show that our proposed classification scheme can provide the high classification accuracy of electricity data while protecting the privacy of electricity data.
摘要:
In recent years, there has been a great development in the research of automated detection of diabetic retinopathy, and deep learning algorithms have been more and more widely used in this field. In this paper, we propose a channel cross enhancement network based on a two-stream model for diabetic retinopathy severity grading for the detailed performance of diabetic retinopathy images on different channels (RGB). The model takes the features of the full-channel input image as global features and the features extracted from the green channel of the original image as local features, and the local features complement the global features to enhance the model's ability to extract the global channel information of the image. In addition, a channel cross-attention module (CCAM) is designed to achieve the effective extraction of global channel features and the interaction of local channel features with global channel features. The proposed method is validated on the Messidor-2 dataset, and the experimental results show that the proposed method outperforms the existing methods in terms of accuracy and AUC values. After experimental validation, the method proposed in this paper can be effectively used for the auxiliary diagnosis of diabetic retinopathy, helping doctors to provide an effective basis for early clinical treatment.
摘要:
Radar-based human activity sensing possesses the advantage of penetrating detection, making it highly promising for applications in the fields of security, rescue, and medical treatment. With the advancement of deep learning technology, there is a growing interest in the field of human activity sensing with radars, including tasks such as action recognition and pose reconstruction. However, existing research usually treats these two problems as separate tasks. In this work, we propose a dual-task framework for jointly reconstructing human 3-D pose and classifying human action from 4-D radar images. Our proposed framework uses an ultra wideband multi-input multi-output (MIMO) radar as the detection sensor to obtain the range-azimuth-height-time 4-D imaging data of human targets. A human pose reconstruction network based on 3-D convolutional neural network (CNN) is then used to reconstruct the 3-D human pose, while a dual-branch network based on multiframe 3-D human poses and 4-D radar image is used to classify the human action. To evaluate the performance of the proposed framework, a dual-task dataset is constructed by merging 4-D imaging radar and camera data. Experiments and multiscenario measurements are constructed to validate the effectiveness of the proposed dual-task framework. The results demonstrate that the proposed dual-task network significantly improves the accuracy of both tasks, while providing a single solution for human action recognition and pose reconstruction.
期刊:
IEEE Transactions on Instrumentation and Measurement,2023年72:1-10 ISSN:0018-9456
通讯作者:
Li, G.
作者机构:
[Wang, You; Li, Huiyan; Wang, Hengyang; Li, Guang; Gao, Han; Liu, Li; Zhao, Shuo] Institute of Cyber-Systems and Control, Zhejiang University, State Key Laboratory of Industrial Control Technology, Hangzhou, 310027, China;[Luo, Zhiyuan] University of London, Royal Holloway, Department of Computer Science, Egham, TW20 0EX, United Kingdom;[Zhang, Jin] Changsha University of Science and Technology, School of Computer and Communication Engineering, Changsha, 410114, China
通讯机构:
[Li, G.] I;Institute of Cyber-Systems and Control, China
作者机构:
[Cao, Dun; Wang, Jin; Ru, Jia; Qin, Jian] Changsha Univ Sci & Technol, Coll Comp & Commun Engn, Changsha 410114, Peoples R China.;[Tolba, Amr] King Saud Univ, Community Coll, Dept Comp Sci, Riyadh 11437, Saudi Arabia.;[Zhu, Min] Zhejiang Shuren Univ, Coll Informat Sci & Technol, Hangzhou 310015, Peoples R China.
通讯机构:
[Zhu, M ] Z;Zhejiang Shuren Univ, Coll Informat Sci & Technol, Hangzhou 310015, Peoples R China.
关键词:
Internet of vehicles;road networks;3D road model;structure recognition;GIS
摘要:
Internet of Vehicles (IoV) is a new system that enables individual vehicles to connect with nearby vehicles, people, transportation infrastructure, and networks, thereby realizing a more intelligent and efficient transportation system. The movement of vehicles and the three-dimensional (3D) nature of the road network cause the topological structure of IoV to have the high space and time complexity. Network modeling and structure recognition for 3D roads can benefit the description of topological changes for IoV. This paper proposes a 3D general road model based on discrete points of roads obtained from GIS. First, the constraints imposed by 3D roads on moving vehicles are analyzed. Then the effects of road curvature radius (Ra), longitudinal slope (Slo), and length (Len) on speed and acceleration are studied. Finally, a general 3D road network model based on road section features is established. This paper also presents intersection and road section recognition methods based on the structural features of the 3D road network model and the road features. Real GIS data from a specific region of Beijing is adopted to create the simulation scenario, and the simulation results validate the general 3D road network model and the recognition method. Therefore, this work makes contributions to the field of intelligent transportation by providing a comprehensive approach to modeling the 3D road network and its topological changes in achieving efficient traffic flow and improved road safety.
通讯机构:
[Long, M ] G;Guangzhou Univ, Sch Elect & Commun Engn, Guangzhou 510006, Peoples R China.
关键词:
3D mesh models;integrity authentication;semi-fragile reversible watermarking;spherical crown volume division
摘要:
Aiming at the large distortion and low tampering localization accuracy of the existing semi-fragile reversible watermarking for 3D mesh models, a novel semi-fragile reversible watermarking for 3D models using spherical crown volume division is proposed. The crown volume of a sphere is divided to reduce the embedding distortion. The possible geometric and topological transformations are separately considered in the watermark generation, and the vertices of the one-ring neighbourhood are grouped to improve the tampering localization accuracy. Experimental results show that the proposed scheme can achieve better localization accuracy and lower embedding distortion than some state-of-the-art algorithms. It has good potential for the applications in integrity authentication for 3D mesh models.
通讯机构:
[Huang, D ] C;Cent South Univ, Sch Comp Sci & Engn, Changsha 410083, Peoples R China.
摘要:
With rapid progress and significant successes in quantum hacking attacks, machine learning is being applied in continuous-variable quantum key distribution (CVQKD) systems. However, most machine-learning networks are shown to be vulnerable to adversarial examples. Thus, we propose a defense scheme with artificial key fingerprints for CVQKD systems. We first embed artificial key fingerprints into quadrature values, then exploit changes of fingerprints in the communication process to probe whether the samples have been manipulated. The paper verifies the effectiveness of our proposal based on four groups of samples that consist of normal measurement values and abnormal measurement values under four attacks. Simulation results show that the model can monitor sample changes in real time with high precision, and our scheme stays robust against adversarial examples attack. This proposal provides a different idea to address the potential vulnerability between machine-learning models used in CVQKD systems and their possible attacked training and test data by adversarial examples.
摘要:
We investigate the dynamic evolution of resonant radiation (RR) emitted from modulated Airy pulses in an optical fiber with emphasis on the third-order dispersion. We show that the process of RRs emission strongly depends on the parameters of the predesigned spectral phase imposed on the Airy pulses, which is attributed to its linear focusing behaviors. A localized pulse formed at the distance-controlled focusing can effectively generate a large amount of RR after a prescribed propagation distance. At variance with the case of fundamental soliton, the minimum value of third-order dispersion required for the onset of RRs emission becomes much smaller for the modulated Airy pulse. The conversion efficiency of RRs increases with an increasing linear focusing point, but decreases with an increasing truncated coefficient. There is an optimal value of the truncated coefficient for the RRs having highest peak intensity and largest frequency shift. The impact of Raman effects on the RRs is also revealed. Our results not only provide a simply route to actively manipulate the efficient emission of RRs in conventional optical fiber, also could have significant implications of pulse shaping technology in novel highly efficient light sources based on the RRs emission for a variety of application.