检索结果-内蒙古大学图书馆

2023 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting, AP-S/URSI 2023

作者： Yuan, Yuan He, Ruisi Ai, Bo Yang, Mi Chen, Ruifeng Zhong, Zhangdui Beijing Jiaotong University State Key Laboratory of Rail Traffic Control and Safety Beijing100044 China Institute of Computing Technology China Academy of Railway Sciences Corporation Limited Beijing100081 China

ISBN: (纸本)9781665442282

In this paper, a geometry-based millimeter-wave channel model is proposed for on-board reconfigurable intelligent surface (RIS)-assisted high-speed train communications, where three propagation cases are considered to describe real channels. In addition, RIS phase optimization method is presented to enhance received power and mitigate channel fading. Based on the proposed model, impacts of RIS phase optimization and channel parameters on channel impulse responses are analyzed. Numerical results indicate that received power at a specific area can be enhanced by adjusting the phase of RIS elements, and a larger number of RIS elements can effectively improve performance of high-speed train systems. These observations and conclusions are insightful for designing and realizing RIS-assisted high-speed train communications. © 2023 IEEE.

关键词： Fading channels

来源：评论

学校读者我要写书评

暂无评论

Improved Kalman Filter-Based Weighted Center-of-Mass Localization Algorithm 15

Improved Kalman Filter-Based Weighted Center-of-Mass Localiz...

引用

15th International Conference on Communication Software and Networks, ICCSN 2023

作者： Chang, Yanzhao Wang, Qin Yin, Zhenyu Ju, Chengen Bi, Zhiying Wang, Jie Shenyang Institute of Computing Technology Chinese Academy of Sciences University of Chinese Academy of Sciences Liaoning Key Lab. of Domestic Industrial Control Platform Technology on Basic Hardware and Software Shenyang China

ISBN: (纸本)9798350329360

In wireless sensor networks, for the problem that the received signal strength indication (RSSI) is affected by signal noise interference in the mining environment and has a large deviation in distance measurement accuracy, this paper proposes an improved weighted center-of-mass localization algorithm based on Kalman filtering. The collected RSSI signal values are corrected using the Kalman filter to mitigate the effects caused by the environment to a certain extent. The environmental parameters in the ranging equation were optimized using linear regression, and the weighted location of the point to be measured was calculated by calculating the sum of the squares of the ratio of the radius distance between the wireless access nodes and the RSSI value to improve the weighted center-of-mass positioning algorithm. The experiments show that the improved algorithm has significantly reduced the positioning error compared with the original positioning algorithm, and can basically meet the positioning requirements of underground coal mines. © 2023 IEEE.

关键词： Kalman filters

来源：评论

学校读者我要写书评

暂无评论

Vision Transformers(ViT) Pretraining on 3D ABUS Image and Dual-CapsViT: Enhancing ViT Decoding via Dual-Channel Dynamic Routing

Vision Transformers(ViT) Pretraining on 3D ABUS Image and Du...

引用

2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

作者： Xu, Mingwang Wang, Wei Wang, Kuanquan Dong, Suyu Sun, Pengzhong Sun, Jinwei Luo, Gongning Harbin Institute of Technology Faculty of Computing Harbin China Harbin Institute of Technology School of Computer Science and Technology Shenzhen China Northeast Forestry University College of Computer and Control Engineering Harbin China Harbin Institute of Technology School of Instrumentation Science and Engineering Harbin China

ISBN: (纸本)9798350337488

Breast cancer continues to be a pressing global health concern, emphasizing the essential need for effective diagnostic techniques. Automated Breast Ultrasound Systems (ABUS) provide a promising advance in breast tumor detection, yet they require significant expertise in interpreting 3D ABUS images, a task fraught with distinctive challenges. Although Vision Transformers (ViT) display remarkable potential for image processing, their low inductive bias and significant data requirements pose obstacles, particularly in the data-constrained medical field. To mitigate these issues, we introduce a Mask-Recover strategy for pretraining Transformer models on 3D ABUS images, enhancing model adaptability and reducing the data demands of the ViT model. Moreover, recognizing the risk that ViTs' average pooling approach may unintentionally mask small but vital features, we propose Dual-CapsViT, an inventive model combining Transformers and Capsule Networks. This integration affords efficient token routing while preserving fine-grained details. To reconcile potential inconsistencies between capsules and tokens, we engineer a novel dual-channel routing algorithm, strengthening the decoder's performance. We benchmarked our models against well-known standards such as ResNet and ViT for classifying breast tumors in ABUS images. Our models exhibited superior performance, as evidenced by improved accuracy, specificity, and Area Under the Receiver Operating Characteristic Curve (AUC) metrics, thereby affirming Dual-CapsViT's potential to enhance breast cancer diagnostics. © 2023 IEEE.

关键词： abus breast cancer capsule network classification vit

来源：评论

学校读者我要写书评

暂无评论

RadarCam-Depth: Radar-Camera Fusion for Depth Estimation with Learned Metric Scale

arXiv

引用

arXiv 2024年

作者： Li, Han Ma, Yukai Gu, Yaqing Hu, Kewei Liu, Yong Zuo, Xingxing The Institute of Cyber-Systems and Control Zhejiang University Hangzhou China The Department of Computing and Mathematical Sciences California Institute of Technology United States

We present a novel approach for metric dense depth estimation based on the fusion of a single-view image and a sparse, noisy Radar point cloud. The direct fusion of heterogeneous Radar and image data, or their encodings, tends to yield dense depth maps with significant artifacts, blurred boundaries, and suboptimal accuracy. To circumvent this issue, we learn to augment versatile and robust monocular depth prediction with the dense metric scale induced from sparse and noisy Radar data. We propose a Radar-Camera framework for highly accurate and fine-detailed dense depth estimation with four stages, including monocular depth prediction, global scale alignment of monocular depth with sparse Radar points, quasi-dense scale estimation through learning the association between Radar points and image patches, and local scale refinement of dense depth using a scale map learner. Our proposed method significantly outperforms the state-of-the-art Radar-Camera depth estimation methods by reducing the mean absolute error (MAE) of depth estimation by 25.6% and 40.2% on the challenging nuScenes dataset and our self-collected ZJU-4DRadarCam dataset, respectively. Our code and dataset will be released at https://***/MMOCKING/RadarCam-Depth. Copyright © 2024, The Authors. All rights reserved.

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

Observer Based Fast Fixed-Time Non-Singular Terminal Sliding Mode control of Teleoperation Systems

Observer Based Fast Fixed-Time Non-Singular Terminal Sliding...

引用

IEEE SSD International Multi-Conference on Systems, Signals and Devices

作者： Neila Mezghani Ben Romdhane Moncef Triki Sofiene Hajji Tarak Damak Laboratory of Sciences and Techniques of Automatic control and computing (Lab-STA) Higher Institute of Industriels Systems of Gabes Gabes Tunisia Laboratory of Sciences and Techniques of Automatic control and computing (Lab-STA) National School of Electronics and Telecommunications of Sfax Sfax Tunisia Laboratory of Sciences and Techniques of Automatic control and computing (Lab STA) National School of Engineering of Sfax Sfax Tunisia

ISBN: (数字)9798331542726

ISBN: (纸本)9798331542733

This paper introduces a fast fixed-time non-singular terminal sliding mode control approach, utilizing a sliding mode observer, to tackle the tracking challenge in teleoperation systems. A robust fixed-time control law is formulated to achieve accurate tracking performance, effectively countering parametric uncertainties, external disturbances, and time-varying delays. Furthermore, the proposed fast fixed-time non-singular terminal sliding surface ensures convergence within a fixed time frame once the sliding condition is satisfied. To improve practical implementation, a sliding mode observer is integrated, removing the necessity for a velocity sensor. The system's stability is demonstrated through Lyapunov theory.

关键词： Uncertainty Simulation Observers Robot sensing systems Main-secondary Stability analysis Delays Sliding mode control Time-varying systems Convergence

来源：评论

学校读者我要写书评

暂无评论

An Optimization Method for Accelerating Convolutional Neural Networks Based on FPGA 13

An Optimization Method for Accelerating Convolutional Neural...

引用

13th International Conference of Information and Communication Technology, ICTech 2024

作者： Li, Dongji Yin, Zhenyu Gao, Wei Hu, MingZhi Du, Jianglin Shenyang University of Chemical Technology Liaoning Key Laboratory of Domestic Industrial Control Platform Technology on Basic Hardware & Software Shenyang China Shenyang Institute of Computing Technology Chinese Academy of Sciences Liaoning Key Laboratory of Domestic Industrial Control Platform Technology on Basic Hardware & Software Shenyang China Shenyang University of Chemical Technology Shenyang China

ISBN: (纸本)9798350376258

Aiming at the problems of slow deployment of convolutional neural networks in the central processing unit (CPU) and high power consumption in the graphics processing unit (GPU), the convolutional neural network (CNN) developed based on the field programmable gate array (FPGA) platform is adopted.) recognition system, which can achieve the effect of improving the operation speed of the algorithm. FPGA is a reconfigurable hardware device that can be programmed to implement various logic functions and can be optimized according to specific applications. Through the design optimization of LeNet5, this paper designs a LeNet_Min network model with fewer parameters. Considering that the calculation amount of the algorithm and the consumption of logic resources are mainly concentrated in the convolution and pooling layers, a pipeline-based dual-channel data model is proposed. Comparison methods. In the pooling layer and activation function of convolutional neural network, flow pooling is designed, and the design method of activation function and pooling layer isomorphism is proposed. The test results show that the recognition speed of Letnet_Min is about 13 times that of the same network model realized by the CPU on the computer side, and higher than that of the GPU, and the system power consumption is only 1.96w. In addition, the system has high versatility and low resource occupation, and can be applied to other neural network models. ©2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

MRI Brain Tumor Classification Using Optimized APSO-LLRBFNN 2

MRI Brain Tumor Classification Using Optimized APSO-LLRBFNN

引用

2nd International Conference on Machine Learning and Autonomous Systems, ICMLAS 2025

作者： Krishna, Tiruveedula Gopi Sreenivas, Nune Ajay Kumar, M. Reddy, Chandra Sekhar Rathee, Davinder Sing Urgessa, Teklu College of Electrical Engineering and Computing Adama Science and Technology University Department of Computer Science and Engineering Adama Ethiopia Andhra Loyala institute of Engineering and Technology Department of Electrical and Electronics Engineering Vijayawada India College of Electrical Engineering and Computing Adama Science and Technology University Department of Electrical Power and Control Engineering Adama Ethiopia College of Electrical Engineering and Computing Adama Science and Technology University Department of ECE Adama Ethiopia

ISBN: (纸本)9798331505745

The research presents a new efficient machine learning method to classify brain tumors because this task remains vital in fighting the high incidence of brain cancers. The proposed approach unites all its operations including image segmentation with feature extraction and classification into a single processing system. This study uses Particle Swarm Optimization (PSO) and Adaptive PSO (APSO) methodologies to optimize Locally Linear Radial Basis Function Neural Network (LLRBFNN) resulting in PSO-LLRBFNN and APSO-LLRBFNN model implementations. The research examined the optimized APSO-LLRBFNN models against Support Vector Machines (SVM) and Least Mean Squares (LMS) based LLRBFNN classifier methods. Results from experiments show that APSO-LLRBFNN gives superior outcomes with 98.93% accuracy and takes 12.23 seconds for execution, which surpasses currently available methods. © 2025 IEEE.

关键词： Particle swarm optimization (PSO)

来源：评论

学校读者我要写书评

暂无评论

A Low Latency Processing System Architecture Strategy for Edge Service Nodes in Fog computing 16

A Low Latency Processing System Architecture Strategy for Ed...

引用

16th International Conference on Communication Software and Networks, ICCSN 2024

作者： Ju, Chengen Ma, Yue Yin, Zhenyu Bi, Zhiying Shenyang Institute of Computing Technology Chinese Academy of Sciences Shenyang110168 China University of Chinese Academy of Sciences Shenyang110168 China Liaoning Key Laboratory of Domestic Industrial Control Platform Technology on Basic Hardware Software Shenyang110168 China

ISBN: (纸本)9798331506322

This paper proposes a low latency processing system architecture strategy for edge service nodes in fog computing to address the network latency issue in the cloud based monitoring system in shipyards. By utilizing the characteristics of fog computing that are closer to terminal devices, local image acquisition devices directly send the collected image data to the fog server for processing, overcoming the problems of network latency and excessive network usage under the cloud monitoring system architecture. The main motivation for using fog computing in the proposed method is to minimize the delay and network usage generated by the entire monitoring system during image transmission and processing. To demonstrate the effectiveness of the proposed method in reducing latency and network usage, this paper conducted simulation validation on the fog computing monitoring system architecture in iFogSim and compared the results with cloud based monitoring systems. The experimental results show that the fog based monitoring system proposed in this article significantly reduces the delay and network usage generated during image transmission and processing. © 2024 IEEE.

关键词： Fog computing

来源：评论

学校读者我要写书评

暂无评论

Self-Supervised Monocular Indoor Depth Estimation via Log-SSIM and Two-Step Pose Network 13

Self-Supervised Monocular Indoor Depth Estimation via Log-SS...

引用

13th International Conference of Information and Communication Technology, ICTech 2024

作者： Fan, Chao Yin, ZhenYu Zhang, FeiQing Li, Yue Xu, FuLong Feng, Dan University of Chinese Academy of Sciences Beijing100049 China Shenyang Institute of Computing Technology Chinese Academy of Sciences Shenyang110168 China Liaoning Key Laboratory of Domestic Industrial Control Platform Technology on Basic Hardware & Software Shenyang110168 China

ISBN: (纸本)9798350376258

—Self-supervised monocular depth estimation based on image sequences has now achieved remarkable results in outdoor scenes. However, self-supervised learning approches on indoor datasets are extremely challenging. First, it is easy to ignore the influence of regions with low absolute pixel values on training by using the structural similarity (SSIM)-based photometric loss function. Another difficulty is that indoor datasets usually have more complex camera poses. In this paper, we propose LTdepth, a novel approach that can address the above issues. In particular, we first design a Log-SSIM function by combining the log-transform and the original SSIM function, which can effectively balance the effect of different pixel values on training. We then integrated a two-step pose estimation network into the basic model PLNet in order to handle the problem of complex camera poses. Extensive experiments on the NYUv2 dataset not only show the validity of the above innovations, but also demonstrate that our model outperforms other state-of-the-art methods. ©2024 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Trajectory Tracking of AUV under Current Disturbance Based on Adaptive LADRC 2

Trajectory Tracking of AUV under Current Disturbance Based o...

引用

2nd International Conference on Automation, Robotics and Computer Engineering, ICARCE 2023

作者： Ji, Daxiong Wang, Xinwei Xu, Minghui Zhang, Qiang Sharma, Sanjay Sutton, Robert Institute of Marine Electronics and Intelligent Systems Ocean Collage Zhejiang University Donghai Lab Zhoushan China School of Engineering Computing and Mathematics University of Plymouth Intelligent Autonomous Control Systems Plymouth United Kingdom

ISBN: (纸本)9798350308341

In consideration of the trajectory tracking problem of autonomous underwater vehicle (AUV) under external current disturbance, an adaptive linear active disturbance rejection control (LADRC) method is proposed. Unlike the conditional LADRC using fixed parameters for extended state observer, the proposed controller makes use of the error of estimation and automatically adjusts the parameter online. As shown in the simulation, the adaptive LADRC estimates the external current disturbance effectively, and tracks the 3D trajectory better compared with traditional LADRC. © 2023 IEEE.

关键词： Disturbance rejection

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：