检索结果-内蒙古大学图书馆

30th IEEE signal processing and Communications Applications Conference (SIU)

作者： Dogan, Eren Ugurdag, H. Fatih Unlu, Hasan Ozyegin Univ Istanbul Turkey

ISBN: (数字)9781665450928

ISBN: (纸本)9781665450928

Applications of artificial neural networks on low-cost embedded systems and microcontrollers (MCUs), has recently been attracting more attention than ever. Since MCUs have limited memory capacity as well as limited compute-speed compared to workstations, employment of current deep learning algorithms on MCUs becomes more practical with the help of model compression. This makes MCUs common and practical alternative solution for autonomous systems. In this paper, we add model compression, specifically Deep Compression, to an existing work, which efficiently deploys PyTorch models on MCUs, in order to increase neural network speed and save electrical power. First, we prune the weight values close to zero in convolutional and fully connected layers. Secondly, the remaining weights and activations are quantized to 8-bit integers from 32-bit floating-point. Finally, forward pass functions are compressed using special data structures for sparse matrices, which store only nonzero weights. In the case of the LeNet-5 model, the memory footprint was reduced by 12.5x, and the inference speed was boosted by 2.6x.

关键词： Deep learning Embedded systems Autonomous systems Microcontrollers Computational modeling Employment signal processing algorithms

来源：评论

学校读者我要写书评

暂无评论

CapNet: A Deep Learning-based Framework for Estimation of Capnograph signal from PPG 44

CapNet: A Deep Learning-based Framework for Estimation of Ca...

引用

44th Annual International Conference of the IEEE-Engineering-in-Medicine-and-Biology-Society (EMBC)

作者： Ahmed, Shahed Islam, Md Tariqul Biswas, Soumav Samrat, Rayhan Hayther Akash, Tafhimul Islam Subhana, Arik Shahnaz, Celia Bangladesh Univ Engn & Technol BUET Dept Elect & Elect Engn Dhaka Bangladesh

ISBN: (数字)9781728127828

ISBN: (纸本)9781728127828

Ambulatory respiration signal extraction system is required to maintain continuous surveillance of a patient with respiratory deficiency. The capnograph signal has received a lot of attention in recent years as a valuable indicator of respiratory conditions. However, the typical capnograph signal extraction method is quite expensive and also unpleasant to the patient due to the involvement of a nasal cannula. With the advent of wearable sensor technology, there has been significant research on the use of photoplethysmogram (PPG) signals as a less expensive alternative to extract respiratory information. In this paper, we propose CapNet, a novel deep learning-based framework which takes the regular PPG signal as input, and estimates the capnograph signal as output. Training, validation and testing of the proposed networks in CapNet is done using the IEEE TMBE Respiratory Rate Benchmark dataset by utilizing reference capnograph respiration signals. With a lower MSE and higher cross-correlation values, CapNet outperforms two traditional signal processing algorithms and another recently proposed deep neural network, RespNet. The proposed framework expectantly can be implementable and feasible for constant supervising of patients undergoing respiratory ailments.

关键词： Deep learning Training Surveillance Neural networks Estimation signal processing algorithms Nose

来源：评论

学校读者我要写书评

暂无评论

A Robotic Ultrasound System Controlled by Human sEMG signals 7

A Robotic Ultrasound System Controlled by Human sEMG Signals

引用

7th IEEE International Conference on Advanced Robotics and Mechatronics

作者： Xue, Boyang Xue, Wei Huang, Qinghua Northwestern Polytech Univ Sch Mech Engn Xian 710072 Shaanxi Peoples R China Northwestern Polytech Univ Ctr Opt Imagery Anal & Learning OPTMAL Xian 710072 Shaanxi Peoples R China

ISBN: (数字)9781665483063

ISBN: (纸本)9781665483063

With the rapid development of bioelectric signal processing and corresponding equipment, the processing and recognition of electromyography (EMG) signals are more and more widely used in the field of robotics, helping to achieve seamless communication and interaction between humans and machines. Taking the normative movement of human arms as the research object, this paper studies the classification and recognition method of multichannel surface EMG signals and the control problems of robots during exercise. Based on this, a human-computer interaction control system between the human EMG signal and the robot is designed and realized.

关键词： Ultrasonic imaging Trajectory planning Robot control signal processing algorithms Process control signal processing Electromyography

来源：评论

学校读者我要写书评

暂无评论

Detection, Parameters Estimation, and Imaging of Moving Targets Based on Extended Post-Doppler STAP in Multichannel WasSAR-GMTI

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2024年 62卷 1页

作者： Chen, Jingwei An, Daoxiang Ge, Beibei Zhou, Zhimin Natl Univ Def Technol Coll Elect Sci & Technol Changsha 410073 Peoples R China Sci & Technol Complex Aviat Syst Simulat Lab Beijing 100076 Peoples R China

As a new observation mode, wide-angle staring synthetic aperture radar ground moving target indication (WasSAR-GMTI), which can observe the fixed area for a long time and achieve moving target tracking, has attracted a lot of interest. However, with increasingly complex flight trajectories and application scenarios, the performance of traditional method tends to degrade. In this article, a novel approach based on post-Doppler space-time adaptive processing (PD-STAP) and imaging method is proposed. First, selecting and extending training data from repeat-pass data can obtain accuracy estimated clutter covariance matrix (CCM). Moreover, the relationship between extended quantity of training data and clutter suppression performance is further analyzed in this article, and then, the joint-pixel PD-STAP (JP PD-STAP), which utilizing auxiliary units under test cell along range direction, is performed to overcome the effect of residual range curvature after keystone transform (KT). Finally, after clutter suppression, the parameters and imaging method based on WasSAR system are proposed. The multidimensional parameter is simplified as 1-D line search problem, which has high efficiency. Considering the curvilinear trajectory, the back-projection (BP) is adopted to process. Inserting parameters into BP can be seen as coarse motion compensation (MOCO), and the further spectrum compression (SC) method and phase gradient autofocus (PGA) is performed to obtain refocus and relocate imaging of moving target. The real-measured WasSAR data processing results demonstrated the effectiveness and practicability of the proposed algorithm.

关键词： Clutter Imaging Training data Parameter estimation Vectors signal processing algorithms Target tracking Clutter covariance matrix (CCM) keystone transform (KT) post-Doppler space-time adaptive processing (PD-STAP) wide-angle staring synthetic aperture radar ground moving target indication (WasSAR-GMTI)

来源：评论

学校读者我要写书评

暂无评论

Lightweight neural network based model for real-time precise HR monitoring during high intensity workout using consumer smartwatches 44

Lightweight neural network based model for real-time precise...

引用

44th Annual International Conference of the IEEE-Engineering-in-Medicine-and-Biology-Society (EMBC)

作者： Fedorin, Illia Pohribnyi, Vitalii Sverdlov, Denys Krasnoshchok, Illia Samsung R&D Inst Ukraine UA-01032 Kiev Ukraine

ISBN: (数字)9781728127828

ISBN: (纸本)9781728127828

Mobile and wearable electronics is one of the rapidly developing areas of high technologies, which regularly appear new devices that offer new features for monitoring our health, level of physical exertion and everyday activity. From the point of view of medicine and sports, usually wearable devices are used to solve one of two tasks: round-the-clock continuous heart rate monitoring, and the control of the heart rate during workout. One of the drawbacks of such devices is the calculation errors that are caused by the motion artifacts. The most critical of them are those that arise in some cases when performing high-intensity training's, which are conjugate with intense hands movements, such as sprint running or boxing. In addition, limited resources of wearable devices do not always allow usage of very complex algorithms for processing certain events. In our work, an ultra-lightweight framework for a precise real-time heart rate monitoring during the high intensity physical exercises is developed. The model is a combination of a digital signal processing and deep convolutional and recurrent neural networks approaches. From the personalization point of view, the effect of anthropometric parameters has been studied. The model shows an average mean absolute error of 2.4 +/- 2.8 bpm during 5 fold cross-validation on an internal dataset, 2.9 +/- 3.4 bpm when evaluated on 12 IEEE SPC subjects, and 4.8 +/- 5.3 bpm when evaluated on 24 subjects from Wonkwang University dataset. Such results exceed the current state of the art solutions both in terms of the achieved accuracy of heart rate estimation and consumed computational resources.

关键词： Heart rate Performance evaluation Training Wearable computers Estimation signal processing algorithms Real-time systems

来源：评论

学校读者我要写书评

暂无评论

Fast and Efficient Single-Snapshot Range and Angle Estimation for FDA-MIMO Radar

引用

IEEE GEOSCIENCE AND REMOTE SENSING LETTERS 2024年 21卷 1页

作者： Chen, Geng Wang, Chunyang Gong, Jian Xu, Jingwei Tan, Ming Air Force Engn Univ Sch Air Def & AntiMissile Xian 710051 Peoples R China Beijing Aviat Meteorol Inst Beijing 10085 Peoples R China Xidian Univ Natl Lab Radar Signal Proc Xian 710071 Peoples R China Natl Univ Def Technol Coll Informat & Commun Wuhan 430010 Peoples R China

This letter proposes an iterative, fast, and efficient single-snapshot range and angle estimation method for frequency diverse array multiple-input multiple-output (FDA-MIMO) radar. The proposed approach transforms the angle and range estimation problem into the transmit and receive frequency estimation, and a 2-D frequency estimation algorithm is designed for multiple sinusoids (corresponding to multiple targets in radar). The frequency estimation is divided into coarse estimation and fine estimation. The coarse estimation can be achieved using the discrete Fourier transform, and fine estimation is estimated by the designed iterative structure. The designed algorithm achieves effective target parameters' estimation with low computational cost. Simulation results demonstrate the effectiveness of the algorithm under various situations.

关键词： Estimation Frequency estimation Radar Discrete Fourier transforms signal to noise ratio Vectors signal processing algorithms Fast Fourier transform frequency diverse array multiple-input multiple-output (FDA-MIMO) single-snapshot target parameters' estimation

来源：评论

学校读者我要写书评

暂无评论

2-D Autofocus for High-Squint SAR Based on Affine Coordinate Back-Projection Algorithm

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2024年 62卷

作者： Lou, Yishan Lin, Hao Chen, Quan Xing, Mengdao Zhou, Shengwei Sun, Guang-Cai Xidian Univ Natl Key Lab Radar Signal Proc Xian 710071 Peoples R China Shanghai Inst Satellite Engn Shanghai 200240 Peoples R China Xidian Univ Acad Adv Interdisciplinary Res Natl Key Lab Radar Signal Proc Xian 710071 Peoples R China

When synthetic aperture radar (SAR) works in high-squint (HS) mode, the interpolation and scaling operation of traditional frequency domain imaging algorithms will change the original structure of motion error, and lead to imaging difficulties. As a classical time-domain imaging algorithm, the back-projection (BP) algorithm is linear processing with a high tolerance for motion error. Therefore, the BP algorithm is very suitable for HS SAR imaging. To further improve the estimation accuracy of motion error, an innovative affine coordinate (AC) system is introduced into the BP algorithm. Based on this AC system, a novel 2-D autofocus algorithm is proposed, which can more accurately estimate and correct the 2-D phase error of the HS SAR BP image. The proposed algorithm has the following advantages: 1) the AC imaging grid is established according to the proposed resolution calculation method based on the BP image spectrum. Under this imaging grid, the nonsystem range cell migration (NsRCM) and the range defocus term of the BP image are significantly reduced, making the phase error estimation more accurate;2) a spectrum alignment processing for BP image in the AC system is proposed to remove the spectrum aliasing so that the azimuth phase error (APE) can be accurately estimated;and 3) the spectrum of the BP image is orthogonal in the AC system, which makes the 2-D phase error compensation based on the established prior phase error structure more accurate. Simulation and real data experiments validate the performance of the proposed algorithm.

关键词： Imaging Synthetic aperture radar Accuracy signal processing algorithms Radar polarimetry Interpolation Radar imaging Frequency-domain analysis Time-domain analysis Image resolution 2-D autofocus affine coordinate (AC) system back-projection (BP) algorithm high-squint (HS) synthetic aperture radar (SAR)

来源：评论

学校读者我要写书评

暂无评论

signal Iterative Search Method Based on Kalman Filter for Vortex Flowmeter

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2024年 73卷 1页

作者： Chen, Jie Tian, Fu-Rong Wang, Si-Cheng Li, Bin Zhang, Wei Wan, Shu-Xu Shanghai Univ Sch Mech Engn & Automat Shanghai 200444 Peoples R China

Vortex flowmeters measure the fluid flow rate using the principle of fluid oscillation, making the measurement process susceptible to external vibration. Therefore, the research on anti-vibration signal processing of vortex flowmeters is of great significance. Based on a large number of experiments, an iterative search method based on the Kalman filter for vortex flowmeter (hereinafter referred to as ISKF) is proposed innovatively. First, the mathematical models of vortex flow signal and interference are constructed. According to the mathematical model, the interference is characterized as white Gaussian noise, periodic vibration interference, and transient impact interference. Then a Kalman filter-based vortex flow system model is created to eliminate the white Gaussian noise by linearizing the vortex flow signal model. For the periodic vibration, an iterative search criterion is designed based on the amplitude-frequency relationship of the vortex flow signal. Finally, offline and online experiments are conducted on the gas experimental setup. The experimental results show that the ISKF method can better overcome periodic vibration and transient impact interference. In general, this article lays a solid theoretical foundation for the commercialization of a new type of anti-vibration flowmeter.

关键词： Kalman filters Interference Flowmeters Vibrations Mathematical models Transient analysis signal processing algorithms Iterative search Kalman filter periodic vibration transient impact vortex flowmeter

来源：评论

学校读者我要写书评

暂无评论

Randomized Simultaneous Hard Thresholding Pursuit Algorithm for Multiple Measurement Vectors 14

Randomized Simultaneous Hard Thresholding Pursuit Algorithm ...

引用

14th IEEE International Conference on signal processing and Communications (SPCOM)

作者： Bapat, Ketan Atul Chakraborty, Mrityunjoy Indian Inst Technol Dept Elect & Elect Commun Engn Kharagpur W Bengal India

ISBN: (数字)9781665482509

ISBN: (纸本)9781665482509

In this paper, we propose a new algorithm named Randomized Simultaneous Hard Thresholding Pursuit(RSHTP) for the multiple measurements vector (MMV) problem in compressed sensing. In the proposed algorithm, the gradient is calculated only with respect to few of the signals at each iteration that are chosen randomly. This reduces the computational cost which is significant when the problem size is large. A deterministic convergence analysis is carried out where we present theoretical guarantees using the restricted isometric property (RIP). Simulation studies show that the proposed algorithm enjoys at par performance even at a moderate rate of column selection in each iteration.

关键词： Conferences Computational modeling signal processing algorithms signal processing Computational efficiency Pursuit algorithms Compressed sensing

来源：评论

学校读者我要写书评

暂无评论

Singular Value Decomposition-Based Adaptive Sampling Approximate Message Passing Net for Sparse-View CT Reconstruction

IEEE PHOTONICS JOURNAL

引用

IEEE PHOTONICS JOURNAL 2024年第1期16卷 1页

作者： Wu, Zhenhua Xu, Jiafei Yang, Lixia Anhui Univ Informat Mat & Intelligent Sensing Lab Anhui Prov Hefei 230601 Peoples R China China Elect Technol Grp 38th Res Inst Hefei 230031 Peoples R China Southeast Univ State Key Lab Millimeter Waves Nanjing 210096 Peoples R China

Sparse-viewcomputed tomography (CT) imaging is a promising technique for reducing radiation dose and accelerating data acquisition in medical imaging. However, the challenges of handling a reduced number of projection views persist for both iterative estimation and deep neural reconstruction methods. In this paper, to address these challenges, we present a singular value decomposition-based adaptive sampling approximate message passing network (ASAMP-Net) sparse-view CT imaging method. To achieve multiple sparse views projection within a single scene imaging and alleviate the computational burden, our proposed ASAMP-Net method incorporates an adaptive sampling module into the AMP deep unrolling network. This module dynamically adjusts the data samples used during the learning process, making our method highly adaptable to various projection matrices. Moreover, by decomposing the projection matrix into its principal components, our approach identifies the respective contributions of independent structures. We then select the most significant principal components to construct a projection matrix model with increased orthogonality, thereby enhancing reconstruction performance. Extensive experiments on public datasets demonstrate the superiority of our method. Notably, ASAMP-Net handles various sparse projection views with just a single training process, achieving prominent imaging results compared to other methods in the literature.

关键词： Image reconstruction Computed tomography Training Matrix decomposition Imaging Iterative methods signal processing algorithms Adaptive sampling ASAMP-Net deep unfolding image reconstruction sparse-view CT

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：