检索结果-内蒙古大学图书馆

11th International Conference on intelligent Control and information processing, ICICIP 2021

作者： Chen, keyi Che, Hangjun Yang, Xuanhao Leung, Man-Fai Southwest University Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing College of Electronic and Information Engineering Chongqing400715 China Hong Kong Metropolitan University School of Science and Technology Hong Kong

ISBN: (纸本)9781665425155

Graph nonnegative matrix factorization (GNMF) is superior for mining the intrinsic geometric structure embedded in high-dimensional data. As the sparsity of the factorized matrices is crucial for clustering, the l0 norm is commonly used in the formulated optimization problem to enforce the sparseness which makes the problem NP-hard and discontinuous. In this paper, the sparse graph nonnegative matrix factorization (SGNMF) is formulated as a global optimization problem by using the sum of inverted Gaussian functions to approximate the l0 norm, the multiplicative update rules are developed to solve the problem with guaranteed convergence. The superior performance of the proposed approach is substantiated by clustering tests on four public datasets. © 2021 IEEE.

关键词： Global optimization

来源：评论

学校读者我要写书评

暂无评论

Impact of Crosstalk and Loss on the Training Process of Integrated Photonic Neural Networks 6

Impact of Crosstalk and Loss on the Training Process of Inte...

引用

6th International Conference on Electronic Engineering and Informatics, EEI 2024

作者： Pang, Jing Su, Ye Xie, Yiyuan College of Electronic and Information Engineering Southwest University Chongqing400715 China Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing Chongqing400715 China Key Laboratory of Networks and Cloud Computing Security of Universities in Chongqing Chongqing400715 China

ISBN: (纸本)9798350353594

Integrated photonic neural networks (IPNNs), with their ability to process large-scale data and parallelize multiple computational tasks, have shown impressive promise for applications in the field of artificial intelligence. However, due to some hard-to-ignore non-ideal characteristics of the underlying silicon photonic MZI devices constituting the IPNN, these non-ideal characteristics can cause the input signal to be affected by crosstalk noise and optical loss during transmission, thus affecting the training and inference accuracy of the IPNN. In order to solve these problems, this paper establishes an analytical model of IPNN affected by crosstalk noise and loss during the training process, covering component level, layer level and system level. Simulations using this model show that the inference accuracy of IPNN decreases by about 65% when the crosstalk noise in the training phase of IPNN reaches -20 dB, while the inference accuracy of IPNN decreases by about 55% when the optical loss reaches -0.5 dB. © 2024 IEEE.

关键词： Crosstalk

来源：评论

学校读者我要写书评

暂无评论

High Contrast Ratio 4-to-2 Plasmonic Encoder Utilizing Cascaded Ring Resonators 6

High Contrast Ratio 4-to-2 Plasmonic Encoder Utilizing Casca...

引用

6th International Conference on Electronic Engineering and Informatics, EEI 2024

作者： Chen, Yan Xie, Yiyuan Chai, Junxiong Ye, Yichen College of Electronic and Information Engineering Southwest University Chongqing400715 China Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing Chongqing400715 China Key Laboratory of Networks and Cloud Computing Security of Universities in Chongqing Chongqing400715 China

ISBN: (纸本)9798350353594

In this paper, a novel plasmonic 4-to-2 encoder comprising cascaded ring resonators coated with graphene layers and two metal-insulator-metal (MIM) waveguides. The encoding function operates at 850 nm by modulating the intensity of pump beam focused on the graphene layer to manipulate the state of the plasmonic resonators. The performance of the encoder is analyzed utilizing the finite-difference time-domain method, demonstrating a high contrast ratio of 15.1 dB. Thus, this efficient 4-to-2 optical encoder is well-suited for integration into optical integrated computing devices. © 2024 IEEE.

关键词： Semiconductor insulator boundaries

来源：评论

学校读者我要写书评

暂无评论

A simple motion deblurring method by two continuous images

A simple motion deblurring method by two continuous images

引用

International Conference on Informative and Cybernetics for Computational Social Systems (ICCSS)

作者： Jihai Zhang Jianfeng Li Guangyuan Liu Tong Chen Tengteng Zhu Bo Li The Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing Chongqing China

ISBN: (数字)9781728158570

ISBN: (纸本)9781728158587

Restoring blurred images to clear images is a challenging problem. Most previous methods only analyzed the single image. However, for motion blurring, this method missed the key trajectory description process. Based on the essence of motion blur, our research combined the continuous frame image information in video to estimate the motion track, the SIFT algorithm was proposed for the first time to match the feature points of the target in the continuous frame image. The accurate motion blur angle and motion blur distance were obtained by combining the matching information to obtain an accurate PSF. Finally, the Wiener filtering algorithm combined with PSF was used for deblurring. Experiments showed that the proposed method can improve the image sharpness after restoration compared with the previous algorithms.

关键词： motion blur continuous frame image deblurring PSF SIFT motion blur deblurring Point spread function Sifting SHARPNESS OF IMAGE images Continuous structures continuous image

来源：评论

学校读者我要写书评

暂无评论

Observer-Based Consensus of nonlinear Multi-Agent Systems With Input Delay via Delay-Dependent Event-Triggered Control

IEEE Transactions on Industrial Cyber-Physical Systems

引用

IEEE Transactions on Industrial Cyber-Physical Systems 2023年 1卷 157-166页

作者： Wei Zhang Bingyan Gong Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing College of Electronics and Information Engineering Southwest University Chongqing China

The distributed control for leader-following consensus problem of nonlinear multi-agent systems with input delay is investigated in this paper. Based on observer state information, an improved distributed truncated prediction control strategy is proposed in order to address the issue of input delay. Furthermore, a delay-dependent event-triggered control protocol is designed to alleviate the use of computation resources and communication bandwidth. Unlike the existing event-triggered protocol, the protocol in the article is delay-dependent and specially designed for systems with input delay. Moreover, the Zeno behavior does not occur. It should be pointed out that, our observer-based control is significant for practical applications where the state of agents cannot be directly obtained. Utilizing the Lyapunov-Krasoviskii function theory, the sufficient conditions of the consensus under above two control methods are obtained. Lastly, we illustrate the feasibility of our results by illustrative examples.

关键词： Delays Protocols Observers Multi-agent systems Linear matrix inequalities Sufficient conditions Cyber-physical systems

来源：评论

学校读者我要写书评

暂无评论

Metaheuristics for Index-Tracking with Cardinality Constraints 11

Metaheuristics for Index-Tracking with Cardinality Constrain...

引用

11th International Conference on information Science and Technology, ICIST 2021

作者： Yuen, Man-Chung Ng, Sin-Chun Leung, Man-Fai Che, Hangjun Open University of Hong Kong School of Science and Technology Hong Kong Hong Kong Southwest University College of Electronic and Information Engineering Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing Chongqing China

ISBN: (纸本)9781665412667

As one of the passive investment strategies, index-tracking aims to replicate market indexes to reproduce market performance. It is widely used for long-term investment. Full replication and partial index-tracking are common approaches for index-tracking problems. Although full replication tracks the chosen market index perfectly, the transaction cost is relatively high in practice. Therefore, partial index-tracking is desired that can reduce the transaction cost and avoid illiquid assets. The partial index-tracking approach selects the subset of a benchmark index and applies restrictions for the numbers of stocks with cardinality constraints. The constrained problem is converted into an unconstrained problem by adding the penalty term. This paper is concerned with the sparse index-tracking problem with cardinality constraints by various metaheuristics. Various metaheuristics are used to deal with the sparse index-tracking problem, and their performances are compared. Also, various penalty values are adopted to test the performance of the compared algorithm. © 2021 IEEE.

关键词： Constrained optimization

来源：评论

学校读者我要写书评

暂无评论

Secrecy performance analysis of single-input multiple-output generalized-K fading channels

引用

Frontiers of information Technology and Electronic Engineering 2016年第10期17卷 1074-1084页

作者： Lei, Hong-jiang Ansari, Imran Shafique Gao, Chao Guo, Yong-cai Pan, Gao-feng Qaraqe, Khalid A. Chongqing Key Laboratory of Mobile Communications Technology Chongqing University of Posts and Telecommunications Chongqing400065 China MOE Key Laboratory of Optoelectronic Technology and Systems Chongqing University Chongqing400044 China Department of Electrical and Computer Engineering Texas A&M University at Qatar Doha Qatar Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing Southwest University Chongqing400715 China

In this paper, the transmission of confidential messages through single-input multiple-output (SIMO) independent and identically generalized-K (KG) fading channels is considered, where the eavesdropper overhears the transmission from the transmitter to the receiver. Both the receiver and the eavesdropper are equipped with multiple antennas, and both active and passive eavesdroppings are considered where the channel state information of the eavesdropper’s channel is or is not available at the transmitter. The secrecy performance of SIMO KG systems is investigated. Analytical expressions for secrecy outage probability and average secrecy capacity of SIMO systems are derived via two different methods, in which KG distribution is approximated by the Gamma and mixture Gamma distributions, respectively. Numerical results are presented and verified via the Monte-Carlo simulation. © 2016, Journal of Zhejiang University Science Editorial Office and Springer-Verlag Berlin Heidelberg.

关键词： Fading channels

来源：评论

学校读者我要写书评

暂无评论

Fat-tree-based optical networks-on-chip with WDM: Crosstalk noise exploiting

Fat-tree-based optical networks-on-chip with WDM: Crosstalk ...

引用

作者： Zhang, Jingping Xie, Yiyuan Wang, Shujian Ye, Yichen Liu, Yuzhu Pan, Gaofeng Liu, Yong Zhang, Jiliang Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing National Joint Engineering Laboratory of Intelligent Transmission and Control Technology College of Electronic and Information Engineering Southwest University Chongqing400715 China School of Optoelectronic Information University of Electronic Science and Technology of Chengdu Chengdu611731 China

Wavelength division multiplexing (WDM) technology as an extension of optical networks-on-chip (ONoCs) performance, can be applied to further overcome shortages of low bandwidth and high-power consumption in multiprocessor systems-on-chip. Nevertheless, WDM-based ONoCs (ONoCs_W) can encounter performance degradation and network scalability constraints because of its power loss and intrinsic crosstalk noise, especially the non-linear four-wave mixing crosstalk noise. Here, a formal systematical exploiting for worst-case crosstalk noise and optical signal-to-noise ratio (OSNR) in arbitrary fat-tree-based ONoCs with WDM (FONoCs_W) is proposed. Meanwhile, the worst-case OSNR optical link candidates are identified. To verify the proposed analysis method, FONoCs_W employing an optical turnaround router with WDM (WOTAR) is used for case study. The study results demonstrate that crosstalk noise significantly diminishes OSNR and restricts scalability of FONoCs_W. For instance, when input optical signal power equals 0 dBm in worst case, the average OSNR is −9.47 dB, the mean power of crosstalk noise for eight wavelengths is slightly greater than that of signal when the number of processor cores is 64;when it equals to 128, the average OSNR is −34.37 dB, decreased by 24.90 dB compared with the network with 64 processor cores. © The Institution of Engineering and Technology 2018

关键词： Network-on-chip

来源：评论

学校读者我要写书评

暂无评论

A distributed stochastic optimization algorithm with gradient-tracking and distributed heavy-ball acceleration

引用

Frontiers of information Technology & Electronic Engineering 2021年第11期22卷 1463-1476页

作者： Bihao SUN Jinhui HU Dawen XIA Huaqing LI Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing College of Electronic and Information EngineeringSouthwest UniversityChongqing400715China College of Data Science and Information Engineering Guizhou Minzu UniversityGuiyang550025China

Distributed optimization has been well developed in recent years due to its wide applications in machine learning and signal *** this paper,we focus on investigating distributed optimization to minimize a global *** objective is a sum of smooth and strongly convex local cost functions which are distributed over an undirected network of n *** contrast to existing works,we apply a distributed heavy-ball term to improve the convergence performance of the proposed *** accelerate the convergence of existing distributed stochastic first-order gradient methods,a momentum term is combined with a gradient-tracking *** is shown that the proposed algorithm has better acceleration ability than GT-SAGA without increasing the *** experiments on real-world datasets verify the effectiveness and correctness of the proposed algorithm.

关键词： Distributed optimization High-performance algorithm Multi-agent system Machine-learning problem Stochastic gradient

来源：评论

学校读者我要写书评

暂无评论

Consensus of multi-agent systems with dynamic join characteristics under impulsive control

引用

Frontiers of information Technology & Electronic Engineering 2021年第1期22卷 120-133页

作者： Xiang HU Zufan ZHANG Chuandong LI School of Communication and Information Engineering Chongqing University of Posts and TelecommunicationsChongqing 400065China Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing College of Electronic and Information EngineeringSouthwest UniversityChongqing 400715China

We study how to achieve the state consensus of a whole multi-agent system after adding some new agent groups dynamically in the original multi-agent *** analyze the feasibility of dynamically adding agent groups under differen t forms of net work topologies that are currently common,and obtain four feasible schemes in theory,including one scheme that is the best in actual industrial ***,we carry out dynamic modeling of multi-agent systems for the best *** control theory and Lyapunov stability theory are used to analyze the conditions so that the whole multi-agent system with dynamic join characteristics can achieve state ***,we provide a numerical example to verify the practicality and validity of the theory.

关键词： Multi-agent system Network topology Impulsive input Dynamic join charac teristics State consensus

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：