检索结果-内蒙古大学图书馆

GLOBECOM 2022 - 2022 IEEE Global communications Conference

作者： Huaji Zhou Xu Wang Jing Bai Zhu Xiao Science and Technology on Communication Information Security Control Laboratory Jiaxing China School of Artificial Intelligence Xidian University Xi'an China College of Computer Science and Electronic Engineering Hunan University Changsha China

ISBN: (纸本)9781665435413

Deep learning-based recognition of radio signal modulation has emerged as a current research hotspot with significant practical potential. However, in practical applications, radio modulation signal data acquisition is complicated to obtain, and label samples are costly and time-consuming to meet the data dependence of deep learning. Transfer learning allows pretrained networks to be reused on large-scale datasets, making it a kind of solution for modulation signal recognition in limited data. The method of suppressing small singular values in the feature vector is employed in this paper to realize selective knowledge transfer for modulation signal recognition, while stochastic normalization is employed to replace the batch normalization layer to avoid over-fitting. We tested the stochastic normalized selective knowledge transfer method on the RML2016.10A and RML2016.04C datasets, with an SNR of 6dB signal samples, and found that it can lead to average growth of 15.77% and 10.32% when compared to direct training, and 6.1% and 2.73% when compared to vanilla fine-tuning. In addition, we check up under a variety of SNR conditions to ensure that our method is effective.

关键词： Training Deep learning Transfer learning Data acquisition Modulation Global communication Knowledge transfer

来源：评论

学校读者我要写书评

暂无评论

Lightweight Vision Transformer for Small Data Sets

Lightweight Vision Transformer for Small Data Sets

引用

Youth Academic Annual Conference of Chinese Association of Automation (YAC)

作者： Wenxing Fang Yanxu Su Institutes of Physical Science and Information Technology Anhui University Hefei China School of Artificial Intelligence Anhui University Hefei P. R. China Key Laboratory of Measurement and Control of Complex System of Engineering Ministry of Education Southeast University Nanjing P. R. China

With the rapid development of Transformers in the field of computer vision, models based on Transformers have become highly competitive architectures in the area. Although variants of Transformer models have achieved increasing accuracy on image classification tasks, the size of the training set and the number of parameters required by the models have increased dramatically. When dealing with small datasets, such models face problems such as overfitting and undergeneralization, leading to poor accuracy on the test set. We propose a new lightweight vision transformer (LVT) to address these issues. We reconstructed the backbone network, which learns the relationship between pixels through local window self-attention and global self-attention computation. We also use the attention pooling approach to fuse the token sequences generated by the backbone network more meticulously. We trained on the CIFAR-10 and CIFAR-100 datasets from scratch and compared them with a modern convolutional neural network. The experimental results show that LVT outperforms the modern convolutional neural network in terms of accuracy and efficiency. On the test set of CIFAR-10, we have obtained an accuracy of 96.83%, which indicates that our model can effectively solve the problems facing the training of small datasets and has a wide range of application prospects.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Seasonal characteristics analysis and uncertainty measurement for wind speed time series

引用

Energy engineering: Journal of the Association of Energy engineering 2020年第5期117卷 289-299页

作者： Deng, Xing Shao, Haijian Wang, Xia School of Computer Science Jiangsu University of Science and Technology Zhenjiang212003 China School of Automation Key Laboratory of Measurement and Control for CSE Ministry of Education Southeast University Nanjing210096 China School of Information Science and Technology Nantong University Nantong226019 China Department of Electrical and Computer Engineering National University of Singapore 117583 Singapore

Wind speed’s distribution nature such as uncertainty and randomness imposes a challenge in high accuracy forecasting. Based on the energy distribution about the extracted amplitude and associated frequency, the uncertainty measurement is processed through Rényi entropy analysis method with time-frequency nat-ure. Nonparametric statistical method is used to test the randomness of wind speed, more precisely, whether or not the wind speed time series is independent and iden-tically distribution (i.i.d) based on the output probability. Seasonal characteristics of wind speed are analyzed based on self-similarity in periodogram under scales range generated by wavelet transformation to reasonably divide the original dataset and effectively reflect the seasonal distribution characteristics. Experimental evaluation based on the dataset from National Renewable Energy Laboratory (NREL) is given to demonstrate the performance of the proposed approach. © 2020, Tech Science Press. All rights reserved.

关键词： Wind speed

来源：评论

学校读者我要写书评

暂无评论

Sensing terminal for big data oil and gas pipeline corrosion monitoring 3

Sensing terminal for big data oil and gas pipeline corrosion...

引用

3rd International Conference on Computer Network, Electronic and Automation, ICCNEA 2020

作者： Li, Lin Wang, Yiding School of Electronic Engineering Xi'An Shiyou University Xi'an China Shaanxi Provincial Key Lab of Oil and Gas Well Measurement and Control Technology Xi'an China

ISBN: (纸本)9781728170831

Facing the construction requirements of big data pipelines, a distributed sensor terminal for multi-parameter monitoring of oil and gas pipeline anti-corrosion was designed. The terminal uses the microprocessor S3C2440A as the core, and several sensors form a multi-parameter sensor array monitoring circuit, and realizes the data transmission of the Internet of Things through the 4G DTU communication module and TCP protocol. The device can sense the temperature value, reference potential value, programmable power supply current and voltage output value of the long-distance pipeline in real time, and realize the data interaction with the cloud server. Tests show that the device is practical in realizing the application of big data pipeline monitoring value. © 2020 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Creation and control of high-dimensional multi-partite classically entangled light

引用

Light(Science & Applications) 2021年第5期10卷 796-805页

作者： Yjie Shen Iaac Nape Xilin Yang Xing Fu Mali Gong Darryl Naidoo Andrew Forbes School of Physics University of the WitwatersrandPrivate Bag 3Wits 2050South Africa State Key Laboratory of Precision Measurement Technology and Instruments Department of Precision InstrumentTsinghua University100084 BeijingChina Electrical and Computer Engineering Department University of CaliforniaLos AngelesCA 90095USA Key Laboratory of Photonic Control Technology(Tsinghua University) Ministry of Education100084 BeijingChina CSIR National Laser Centre PO Box 395Pretoria 0001South Africa Optoelectronics Research Centre University of SouthamptonSouthampton S0171BJUK

Vector beams,non-separable in spatial mode and polarisation,have emerged as enabling tools in many diverse applications,from communication to *** applicability has been achieved by sophisticated laser designs controlling the spin and orbital angular momentum,but so far is restricted to only two-dimensional *** we demonstrate the first vectorially structured light created and fully controlled in eight dimensions,a new *** externally modulate our beam to control,for the frst time,the complete set of classical Greenberger-Horne-Zeilinger(GHZ)states in paraxial structured light beams,in analogy with high-dimensional multi-partite quantum entangled states,and introduce a new tomography method to verify their *** complete theoretical framework reveals a rich parameter space for further extending the dimensionality and degrees of freedom,opening new pathways for vetorilly structured light in the classical and quantum regimes.

关键词： quantum analogy dimensions

来源：评论

学校读者我要写书评

暂无评论

Research on license plate recognition of two-wheel motor vehicle based on image processing 3

Research on license plate recognition of two-wheel motor veh...

引用

3rd International Conference on Computer Network, Electronic and Automation, ICCNEA 2020

作者： Li, Tong Zhang, Qizhi Shaanxi Provincial Key Lab of Oil and Gas Well Measurement and Control Technology Xi'an China Xi an Shiyou University School of Electronic Engineering Xi'an China

ISBN: (纸本)9781728170831

The convenience and economy of non-four-wheel motor vehicles are favored, and the number of non-four-wheel motor vehicles is rapidly increasing. However, it also brings some problems to road traffic safety management. The traffic management department has put forward a series of management measures, and how to recognize its license plate has become the premise. This paper realizes the license plate recognition by preprocessing the original image, license plate positioning, character segmentation, character recognition and other processes, and achieves the expected effect, which has reference significance for the research of license plate recognition. © 2020 IEEE.

关键词： Roads and streets

来源：评论

学校读者我要写书评

暂无评论

Research on an improved single neuron pi control strategy 3

Research on an improved single neuron pi control strategy

引用

3rd International Conference on Computer Network, Electronic and Automation, ICCNEA 2020

作者： Li, Lin Luo, Mingshuai School of Electronic Engineering Xi'An Shiyou University Xi'an China Shaanxi Provincial Key Lab of Oil and Gas Well Measurement and Control Technology Xi'an China

ISBN: (纸本)9781728170831

In the control system of AC variable frequency speed regulation of asynchronous motor, vector control method is adopted in most of them, and PI control strategy is usually used in speed loop of vector control system. For pulse load system, such as sports training treadmill, the parameters of PI controller in vector control system can't be modified timely according to the characteristics of pulse load, so it can't obtain high-efficiency smooth speed regulation characteristics. Therefore, a single neuron PI control strategy is proposed, which can improve the online performance of single neuron PI controller by establishing a learning algorithm of single neuron gain K Real time adjustment capability. The motor control system based on motion training treadmill is studied in this paper. The model of asynchronous motor vector control system is built in Matlab / Simulink environment. The PI algorithm and single neuron PI algorithm are compared, and the influence of single neuron PI algorithm on the control quality of the whole system is discussed. © 2020 IEEE.

关键词： MATLAB

来源：评论

学校读者我要写书评

暂无评论

Performance analysis of linear regression based on python 2nd

Performance analysis of linear regression based on python

引用

2nd International Cognitive Cities Conference, IC3 2019

作者： Shen, Gongxin Liu, Qihe School of Information Engineering Nanjing Polytechnic Institute Nanjing210048 China Jiangsu Province Fluid Sealing and Measurement and Control Engineering Technology Research and Development Center Nanjing210044 China

ISBN: (纸本)9789811561122

Python language has become the most popular machine learning language because of its simplicity, readability and expansibility. By expanding the library NumPy, it can achieve fast array processing. At the same time, Python can directly implement machine learning algorithm by extending TensorFlow framework, but there is a lack of quantitative analysis among different methods. This paper simulates different amounts of data and then realizes linear regression with NumPy library and TensorFlow framework respectively. By comparing the calculation time of the same number of iterations and the analysis of the same number of iterations with the same accuracy, A quantitative data analysis is given, which provides some reference for the practical application of NumPy library and TensorFlow framework. © Springer Nature Singapore Pte Ltd 2020.

关键词： Python

来源：评论

学校读者我要写书评

暂无评论

Design and Research of a Nonlinear Distortion Three-stage Coupling Amplifier

Design and Research of a Nonlinear Distortion Three-stage Co...

引用

Image Processing, Electronics and Computers (IPEC), Asia-Pacific Conference on

作者： Shilong Mu Zhimin Wu Guannan Liu Zhen Chen Wanwan Yu School of Information and Control Engineering China University of Mining and Technology Xuzhou China School of Electrical and Power Engineering China University of Mining and Technology Xuzhou China School of Information and Communication Engineering Harbin Engineering University Harbin China

ISBN: (纸本)9781728190198

In applications, the nonlinear characteristics of transistor, unreasonable setting of static operating point or excessive signal input often lead to the amplifier working in the nonlinear region, resulting in waveform distortion. In this paper, a three-stage coupled amplifier with nonlinear distortion is designed. The original signal is amplified by two-stage emitter bias amplifier circuit and a class B single-source complementary symmetrical power amplifier. The first stage realizes signal amplification. The top distortion, bottom distortion and bidirectional waveform output are realized by adjusting the base divider resistance and emitter resistance of the second stage. In the third stage, the short-circuit diode of relay is used to realize the output of cross distortion waveform. The control end uses stm32f4 as the main control chip, keyboard as the input module, LCD as the display module, the control end outputs the amplitude on the display screen through AD sampling, and uses fast Fourier transform (FFT) to calculate the total harmonic distortion (THD), which is verified to have high accuracy.

关键词： Resistance Total harmonic distortion Nonlinear distortion Power amplifiers Keyboards Approximation algorithms Liquid crystal displays

来源：评论

学校读者我要写书评

暂无评论

An abundance estimation algorithm based on orthogonal bases for hyperspectral image

引用

Optoelectronics Letters 2019年第5期15卷 396-400页

作者： ZHAO Yan ZHOU Zhen WANG Dong-hui School of Measurement and Communication Engineering Harbin University of Science and Technology Harbin 150080 China College of Information and Communication Engineering Harbin Engineering University Harbin 150001 China

An abundance estimation algorithm based on orthogonal bases is proposed to address the problem of high computational complexity faced by most abundance estimation algorithms that are based on a linear spectral mixing model(LSMM) and need to perform determinant operations and matrix inversion operations. The proposed algorithm uses the Gram-Schmidt method to calculate the endmember vector set to obtain the corresponding orthogonal basis set and solve the unmixing equations to obtain the eigenvector of each endmember. The spectral vector to be decomposed is projected onto the eigenvector to obtain projection vector, and the ratio between the length of the projection vector and the length of the orthogonal basis corresponding endmember is calculated to obtain an abundance estimation of the endmember. After a comparative analysis of different algorithms, it is concluded that the proposed algorithm only needs to perform vector inner product operations, thereby significantly reducing the computational complexity. The effectiveness of the algorithm was verified by experiments using simulation data and actual image data.

关键词： linear spectral mixing model(LSMM) perform determinant operations matrix inversion operations

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：