检索结果-内蒙古大学图书馆

Effect of Deep Learning Algorithm Incorporating Attention Module Optimization on Assisted Training for Youth Running Sports

引用

IEEE ACCESS 2024年 12卷 113960-113971页

作者： Han, Chunlong Liu, Pengyu Zhuhai 1 High Sch Pingsha Campus Zhuhai 519000 Peoples R China Guangzhou Acad Fine Arts Guangzhou 510006 Peoples R China

Given the limitations of traditional methods in automatic feature extraction, the study aims to improve the accuracy of posture recognition in adolescent running training through deep learning techniques. This study first constructed a human pose motion model to capture complex pose changes during the motion process. Furthermore, an algorithm framework combining convolutional neural networks and long short-term memory networks was proposed, and optimized by integrating attention modules to enhance the model's ability to capture key action details and improve recognition accuracy. The outcomes indicated that the recognition accuracy of the raw data processed using the convolutional neural network algorithm is 0.93, which is improved to 0.96 after preprocessing, while the accuracy of the long-short-term memory network algorithm is 0.91 on the raw data and 0.95 after preprocessing. The Inception combined with long-short-term memory network algorithm optimised by incorporating the attention module has the highest accuracy of 0.98 on the preprocessed data, while that on the 0.95 on the raw data, and 1.0 on specific actions such as running and riding. It can be found that the attention module significantly improves the recognition accuracy of the algorithm and its ability to recognise complex actions. This study not only improves the way that teenagers train to run, but it also creates new opportunities for the use of deep learning methods in the context of sports-assisted training.

关键词： Training Accuracy Sports Long short term memory Deep learning signal processing algorithms Optimization Pose estimation Assisted training attention deep learning motion pose recognition

来源：评论

学校读者我要写书评

暂无评论

Sparse Recovery-Based Target Detection for LFMCW Radar: An Estimation-Before-Detection Strategy

引用

IEEE SENSORS JOURNAL 2024年第16期24卷 25949-25959页

作者： Xu, Xinju Wang, Yan Jin, Benzhou Shen, Yutong Zhang, Xiaofei Wu, Qihui Nanjing Univ Aeronaut & Astronaut Coll Elect & Informat Engn Nanjing 211106 Peoples R China

To suppress the sidelobes of strong targets, conventional radar signal processing techniques typically use data windows, which results in large performance losses for the detection of weak targets. Sparse recovery (SR) algorithms can suppress sidelobes without using data windows. However, these algorithms still face challenges in target detection, such as dealing with grid mismatch and the difficulty in designing a constant false alarm rate (CFAR) detector. To solve the above problems, we propose an estimate-before-detection (EBD) algorithm for linear frequency-modulated continuous wave (LFMCW) radar. First, parameter estimation is implemented before target detection based on the original signal. Remarkably, it has been proved that the fast Fourier transform (FFT) algorithm can be used to perform estimation of target parameters efficiently. Second, leveraging the sparsity of the received signal, the Gaussian generalized approximate message-passing-based low-complexity sparse Bayesian learning (GGAMP-SBL) algorithm is used to reconstruct targets. Then, target detection is executed using the reconstruction results. Numerical results demonstrate that the proposed detector maintains the CFAR property and outperforms the conventional signal processing (CSP)-based algorithm.

关键词： Object detection signal processing algorithms Vectors Radar Radar detection Estimation Detectors Constant false alarm rate (CFAR) estimate-before-detection Gaussian generalized approximate message-passing-based low-complexity sparse Bayesian learning (GGAMP-SBL) sparse recovery (SR) target detection

来源：评论

学校读者我要写书评

暂无评论

Robust Variable Projection Algorithm for the Identification of Separable Nonlinear Models

引用

IEEE TRANSACTIONS ON AUTOMATIC CONTROL 2024年第9期69卷 6293-6300页

作者： Chen, Guang-Yong Su, Xiang-Xiang Gan, Min Guo, Wenzhong Chen, C. L. Philip Fuzhou Univ Coll Comp & Data Sci Fuzhou 350116 Peoples R China Key Lab Intelligent Metro Univ Fujian Fujian Key Lab Network Comp & Intelligent Informat Fuzhou 350108 Peoples R China Minist Educ Engn Res Ctr Big Data Intelligence Fuzhou 350108 Peoples R China South China Univ Technol Sch Comp Sci & Engn Guangzhou 510006 Peoples R China

Robust nonlinear regression frequently arises in data analysis that is affected by outliers in various application fields such as system identification, signal processing, and machine learning. However, it is still quite challenge to design an efficient algorithm for such problems due to the nonlinearity and nonsmoothness. Previous researches usually ignore the underlying structure presenting in the such nonlinear regression models, where the variables can be partitioned into a linear part and a nonlinear part. Inspired by the high efficiency of variable projection algorithm for solving separable nonlinear least squares problems, in this article, we develop a robust variable projection (RoVP) method for the parameter estimation of separable nonlinear regression problem with L-1 norm loss. The proposed algorithm eliminates the linear parameters by solving a linear programming subproblem, resulting in a reduced problem that only involves nonlinear parameters. More importantly, we derive the Jacobian matrix of the reduced objective function, which tackles the coupling between the linear parameters and nonlinear parameters. Furthermore, we observed an intriguing phenomenon in the landscape of the original separable nonlinear objective function, where some narrow valleys frequently exist. The RoVP strategy can effectively diminish the likelihood of the algorithm getting stuck in these valleys and accelerate its convergence. Numerical experiments confirm the effectiveness and robustness of the proposed algorithm.

关键词： Optimization Jacobian matrices Linear programming Parameter estimation Autoregressive processes signal processing algorithms Predictive models Radial basis function network based autoregressive (RBF-AR) model robust parameter estimation system identification variable projection (VP) algorithm

来源：评论

学校读者我要写书评

暂无评论

A State-of-the-Art Survey on IRS-NOMA for Integrated Sensing and Communication

引用

IEEE ACCESS 2024年 12卷 186087-186123页

作者： Sur, Samarendra Nath Singh, Arun Kumar Tran, Huu Q. Vishwakarma, Pradeep Imoize, Agbotiname Lucky Li, Chun-Ta Sikkim Manipal Univ Sikkim Manipal Inst Technol Dept Elect & Commun Engn Gangtok 737136 Sikkim India Ind Univ Ho Chi Minh City FET Ho Chi Minh City 70000 Vietnam CNIT I-43124 Parma Italy Politecn Torino DET I-10129 Turin Italy Fu Jen Catholic Univ Grad Inst Appl Sci & Engn Bachelors Program Artificial Intelligence & Inform New Taipei City 24205 Taiwan

In recent years, the rapid advancements in wireless communication technologies have necessitated more efficient and integrated systems. An emerging paradigm addressing this need is the integration of Intelligent Reflecting Surfaces (IRS) with Non-Orthogonal Multiple Access (NOMA) for Integrated Sensing and Communication (ISAC). This paper offers a comprehensive survey of the state-of-the-art IRS-NOMA technologies, emphasizing their potential to enhance ISAC systems. We examine the fundamental principles of IRS and NOMA, investigating how their combined integration can markedly improve spectral and energy efficiencies, system capacity, and overall performance. The survey methodically categorizes and reviews recent research contributions, highlighting key innovations, design methodologies, advancements in the sophisticated algorithms, and performance metrics. Furthermore, we address the practical challenges and implementation issues associated with IRS-NOMA in ISAC, such as hardware limitations, signal processing complexities, and security concerns. Additionally, we outline open research directions and future prospects, aiming to stimulate further advancements in this promising field. Also, we highlight the interplay between the IRS assisted ISAC NOMA with the other emerging technologies towards the realization of 6G. Lastly, we present and summarize the lessons learned from this study. This survey serves as an invaluable resource for researchers and practitioners interested in understanding and advancing IRS-NOMA for next-generation integrated sensing and communication systems.

关键词： Integrated sensing and communication NOMA Surveys Interference cancellation Wireless networks signal processing algorithms Mathematical models Location awareness Downlink 6G mobile communication 6G ISAC IRS optimization sensing wireless communication

来源：评论

学校读者我要写书评

暂无评论

Noise Reduction in Automotive Pulse Radar Using signal Subspace and Presumed Ambiguity Function

引用

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2024年第7期73卷 10708-10713页

作者： Zhu, Luoyan Liu, Yinsheng Vorobyov, Sergiy A. He, Danping Guan, Ke Zhong, Zhangdui Chang, Liang Beijing Jiaotong Univ Sch Elect & Informat Engn Beijing 100044 Peoples R China Beijing Jiaotong Univ State Key Lab Adv Rail Autonomous Operat Beijing 100044 Peoples R China Aalto Univ Dept Signal Proc & Acoust Espoo 02330 Finland State Radio Monitoring Ctr China Beijing Peoples R China

Increasing demand for target detection accuracy under complex road conditions is a challenging issue for Intelligent Transport System (ITS). Weak target signals suffer heavily from heavy noise, especially at low signal-to-noise ratio (SNR). Therefore, effective noise reduction is crucial to enable better analysis of target signals. In pulse radar applications, the ambiguity function (AF) characterizes the time response of a filter matched to a given finite energy signal and resembles the form of a sinc function for several types of transmitted signal waveforms. This paper addresses the problem of noise reduction for pulse radar with such waveforms. In particular, the pulse shape is designed in terms of the AF function. Furthermore, inspired by the fact that the signal subspace spanned by the received signals is dominated by the target echoes, a low-rank filter is developed to reduce noise for pulse echoes in the eigen-domain. Theoretical analysis and simulation results are both presented to demonstrate the improvement of SNR and detection probability.

关键词： Noise reduction Radar signal to noise ratio signal processing algorithms Correlation Standards Program processors Pulse radar noise reduction target detection ambiguity function signal subspace

来源：评论

学校读者我要写书评

暂无评论

A Novel Interpretable Model via Algorithm Unrolling for Intelligent Fault Diagnosis of Machinery

引用

IEEE SENSORS JOURNAL 2024年第1期24卷 495-505页

作者： Rao, Fengpei Zeng, Ming Cheng, Yiwei China Univ Geosci Sch Mech Engn & Elect Informat Wuhan 430074 Peoples R China China Univ Geosci Shenzhen Res Inst Shenzhen 518057 Peoples R China

Identifying the health state of machinery using sensor information is one of the key tasks for intelligent operation and maintenance of machinery. In recent years, deep learning models have been widely used in this field. However, these models usually lack interpretability. To this end, we use the algorithm unrolling technique to unroll the convolutional version of iterative shrinkage-thresholding algorithm (CISTA) into a neural network called CISTA-Net. The CISTA-Net's structure is determined by the iterative algorithm and thus has a clear theoretical basis. By adding a classification module and a reconstruction module in parallel to its output end, the CISTA-Net is interpretable in terms of both its structure and output. Interpretability analysis through the well-known bearing dataset from the Case Western Reserve University (CWRU) shows that the reconstruction module can enhance the interpretability. In addition, a simple residual block can be optionally added to the input end of the CISTA-Net for feature pre-extraction, resulting in an alternative CISTA-Net called ResCISTA-Net. The ResCISTA-Net is compared with some conventional and interpretable fault diagnosis models on another two mechanical fault diagnosis tasks. Experimental results suggest that: 1) the residual block can not only improve diagnostic accuracy but also prevent performance degradation for some tasks;2) the reconstruction module can also improve diagnostic accuracy;and 3) the ResCISTA-Net performs excellently in terms of diagnostic accuracy, training sample size requirement, and noise robustness.

关键词： Convolution Convolutional codes Feature extraction Encoding Kernel signal processing algorithms Fault diagnosis Algorithm unrolling convolutional sparse coding intelligent fault diagnosis interpretable neural network

来源：评论

学校读者我要写书评

暂无评论

Spatial-Temporal-Based Underdetermined Near-Field 3-D Localization Employing a Nonuniform Cross Array

引用

IEEE JOURNAL OF SELECTED TOPICS IN signal processing 2024年第4期18卷 561-571页

作者： Chen, Hua Yi, Zelong Jiang, Zhiwei Liu, Wei Tian, Ye Wang, Qing Wang, Gang Ningbo Univ Fac Elect Engn & Comp Sci Ningbo 315211 Peoples R China Zhejiang Key Lab Mobile Network Applicat Technol Ningbo 315211 Peoples R China Queen Mary Univ London Sch Elect Engn & Comp Sci London E1 4NS England Tianjin Univ Sch Elect Informatin Engn Tianjin 300072 Peoples R China

In this paper, an underdetermined three-dimensional (3-D) near-field source localization method is proposed, based on a two-dimensional (2-D) symmetric nonuniform cross array. Firstly, by utilizing the symmetric coprime array along the x-axis, a fourth-order cumulant (FOC) based matrix is constructed, followed by vectorization operation to form a single virtual snapshot, which is equivalent to the received data of a virtual array observing from virtual far-field sources, generating an increased number of degrees of freedom (DOFs) compared to the original physical array. Meanwhile, multiple delay lags, named as pseudo snapshots, are introduced to address the single snapshot issue. Then, the received data of the uniform linear array along the y-axis is similarly processed to form another virtual array, followed by a cross-correlation operation on the virtual array observations constructed from the coprime array. Finally, the 2-D angles of the near-field sources are jointly estimated by employing the recently proposed sparse and parametric approach (SPA) and the Vandermonde decomposition technique, eliminating the need for parameter discretization. To estimate the range term, the conjugate symmetry property of the signal's autocorrelation function is used to construct the second-order statistics based received data with the whole array elements, and subsequently, the one-dimensional (1-D) MUSIC algorithm is applied. Moreover, some properties of the proposed array are analyzed. Compared with existing algorithms, the proposed one has better estimation performance given the same number of sensor elements, which can work in an underdetermined and mixed sources situation, as shown by simulation results with 3-D parameters automatically paired.

关键词： Arrays Location awareness Array signal processing Estimation Direction-of-arrival estimation signal processing algorithms Symmetric matrices Near-field source localization cross array spatial-temporal underdetermined estimation

来源：评论

学校读者我要写书评

暂无评论

Communication-Avoiding Fusion of GEMM-Based Convolutions for Deep Learning in the RISC-V GAP8 MCU

引用

IEEE INTERNET OF THINGS JOURNAL 2024年第21期11卷 35640-35653页

作者： Ramirez, Cristian Castello, Adrian Martinez, Hector Quintana-Orti, Enrique S. Univ Politecn Valencia DISCA Valencia 46022 Spain Univ Cordoba Elect & Comp Engn Cordoba 21007 Spain

Incorporating deep learning (DL) technologies to the edge is crucial for improving the security, privacy, and energy efficiency of the Internet of Things (IoT). In this scenario, the limitations of edge devices in terms of power dissipation, memory capacity, and processing power require a careful selection and optimization of algorithms for IoT DL applications. In this line, our work focuses on the convolution operator, a key component in deep neural networks for signal processing and computer vision. Specifically, the work aims at the efficient implementation of the lowering-based implementation of this operator, on the GAP8 parallel ultralow power platform (PULP), with the goal of mitigating the data transfer costs across the memory hierarchy. Our contributions include 1) an analytical model for estimating the parallel execution time, 2) the exploration of different configuration options, and 3) four variants of the algorithm that fuse several components to address memory bottlenecks in the method. Overall, our best fused variant provides a speedup of up to 1.25x over the baseline algorithm when applied to infer MobileNet-v1+ImageNet and VGG9+CIFAR10 using eight threads, and up to 1.34x for ResNet18+ImageNet.

关键词： Convolution Tensors Transforms Internet of Things signal processing algorithms Memory management Matrix decomposition convolutional neural networks (CNNs) cost analysis deep learning (DL) edge processors high performance

来源：评论

学校读者我要写书评

暂无评论

A Parkinson's Auxiliary Diagnosis Algorithm Based on a Hyperparameter Optimization Method of Deep Learning

引用

IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS 2024年第4期21卷 912-923页

作者： Wang, Xingbo Li, Shujuan Pun, Chi-Man Guo, Yijing Xu, Feng Gao, Hao Lu, Huimin Nanjing Univ Posts & Commun Coll Automat Nanjing 210049 Peoples R China Nanjing Univ Posts & Commun Coll Artificial Intelligence Nanjing 210049 Peoples R China Hangzhou Zhuoxi Inst Brain & Intelligence Hangzhou 311100 Peoples R China Univ Macau Dept Comp & Informat Sci Macau 999078 Peoples R China Southeast Univ Dept Neurol Zhongda Hosp Nanjing 210009 Peoples R China Tsinghua Univ Sch Software Beijing 100084 Peoples R China Kyushu Inst Technol Dept Mech & Control Engn Kitakyushu 8048550 Japan

Parkinson's disease is a common mental disease in the world, especially in the middle-aged and elderly groups. Today, clinical diagnosis is the main diagnostic method of Parkinson's disease, but the diagnosis results are not ideal, especially in the early stage of the disease. In this paper, a Parkinson's auxiliary diagnosis algorithm based on a hyperparameter optimization method of deep learning is proposed for the Parkinson's diagnosis. The diagnosis system uses ResNet50 to achieve feature extraction and Parkinson's classification, mainly including speech signal processing part, algorithm improvement part based on Artificial Bee Colony algorithm (ABC) and optimizing the hyperparameters of ResNet50 part. The improved algorithm is called Gbest Dimension Artificial Bee Colony algorithm (GDABC), proposing "Range pruning strategy" which aims at narrowing the scope of search and "Dimension adjustment strategy" which is to adjust gbest dimension by dimension. The accuracy of the diagnosis system in the verification set of Mobile Device Voice Recordings at King's College London (MDVR-CKL) dataset can reach more than 96%. Compared with current Parkinson's sound diagnosis methods and other optimization algorithms, our auxiliary diagnosis system shows better classification performance on the dataset within limited time and resources.

关键词： Feature extraction Deep learning Residual neural networks Neural networks signal processing algorithms Diseases Data models Artificial bee colony algorithm deep learning hyperparameter optimization parkinson's auxiliary speech diagnosis

来源：评论

学校读者我要写书评

暂无评论

Momentum-Based Iterative Hard Thresholding Algorithm for Sparse signal Recovery

引用

IEEE signal processing LETTERS 2025年 32卷 1346-1350页

作者： Jin, Wen Xie, Lie-Jun Ningbo Univ Sch Math & Stat Ningbo 315211 Peoples R China

The iterative hard thresholding (IHT) algorithm is widely used for recovering sparse signals in compressed sensing. Despite the development of numerous variants of this effective algorithm, its convergence rate and accuracy in finding the optimal solution still have room for enhancement. Aiming at this issue, we propose a momentum-based iterative hard thresholding (MIHT) algorithm by introducing a new iterative search direction derived from the momentum method, which uses historical iteration information to refine the search direction and thereby accelerate convergence. We establish a sufficient condition, in terms of (3s)-order restricted isometry constant, to guarantee the convergence of MIHT. Excitingly, numerical experiments demonstrate that MIHT possesses an excellent recovery success rate and outperforms a wide range of existing IHT variants.

关键词： signal processing algorithms Vectors Iterative algorithms Convergence Thresholding (Imaging) Indexes Optimization Training Data mining Compressed sensing iterative hard thresholding momentum method restricted isometry constant search direction

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：