检索结果-内蒙古大学图书馆

Sparse signal reconstruction via recurrent neural networks with hyperbolic tangent function

NEURAL NETWORKS 2022年 153卷 1-12页

作者： Wen, Hongsong He, Xing Huang, Tingwen Southwest Univ Sch Elect & Informat Engn Chongqing Key Lab Nonlinear Circuits & Intelligent Chongqing 400715 Peoples R China Texas A&M Univ Qatar Dept Math Doha Qatar

In this paper, several recurrent neural networks (RNNs) for solving the L-1-minimization problem are proposed. First, a one-layer RNN based on the hyperbolic tangent function and the projection matrix is designed. In addition, the stability and global convergence of the previously presented RNN are proved by the Lyapunov method. Then, the sliding mode control technique is introduced into the former RNN to design finite-time RNN (FTRNN). Under the condition that the projection matrix satisfies the Restricted Isometry Property (RIP), a suitable Lyapunov function is constructed to prove that the FTRNN is stable in the Lyapunov sense and has the finite-time convergence property. Finally, we make a comparison of the proposed RNN and FTRNN with the existing RNNs. To achieve this, we implement experiments for sparse signal reconstruction and image reconstruction. The results further demonstrate the effectiveness and superior performance of the proposed RNN and FTRNN. (c) 2022 Elsevier Ltd. All rights reserved.

关键词： L-1-minimization Hyperbolic tangent function Sliding mode control technique Finite-time RNN (FTRNN) Finite-time convergence

来源：评论

学校读者我要写书评

暂无评论

Fractional-order complex correntropy algorithm for adaptive filtering in α-stable environment

引用

ELECTRONICS LETTERS 2021年第21期57卷 813-815页

作者： Qiu, Chen Dong, Zhenyuan Yan, Wenxing Qian, Guobing Southwest Univ Chongqing Key Lab Nonlinear Circuits & Intelligen Coll Elect & Informat Engn Chongqing 400715 Peoples R China

In adaptive filtering applications, the Gaussian distribution cannot be used to model the signal/noise with frequent spikes accurately. In fact, the rational model to simulate the behaviour of such signal/noise is the alpha-stable distribution process. In this letter, a fractional-order complex correntropy algorithm is proposed to deal with the case that both signal and noise processes are modelled as complex-valued alpha-stable signals. Compared with the classical approaches, the proposed fractional-order complex correntropy extends the Gaussian assumption of signal/noise in the complex domain to the assumption of alpha-stable distributions without second-order and higher order statistical moments. Benefitting from the fractional-order calculus and correntropy criterion, fractional-order complex correntropy shows great robustness to the jittery behaviour of complex-valued alpha-stable signal/noise. In addition, a convergence analysis for fractional-order complex correntropy has been carried out. Simulations on system identification revealed that the filtering performance is significantly improved by using fractional-order complex correntropy.

关键词： Other topics in statistics Filtering methods in signal processing Other topics in statistics Signal processing theory

来源：评论

学校读者我要写书评

暂无评论

A Two-Layer Distributed Algorithm Using Neurodynamic System for Solving L1-Minimization

引用

IEEE TRANSACTIONS ON circuits AND SYSTEMS II-EXPRESS BRIEFS 2022年第8期69卷 3490-3494页

作者： Xu, Junpeng He, Xing Han, Xin Wen, Hongsong Southwest Univ Coll Elect & Informat Engn Chongqing Key Lab Nonlinear Circuits & Intelligen Chongqing 400715 Peoples R China

This brief considers a distributed algorithm for solving L-1-minimization problem based on nonlinear neurodynamic system. Compared with centralized algorithms, distributed algorithms have great potential in data privacy protection, distributed storage and processing of data. In this brief, L-1-minimization problem is transformed into a distributed problem by using multiagent consensus theory. For the distributed optimization problem, a two-layer distributed algorithm is designed by utilizing neurodynamic system, projection matrix and derivative feedback technique. Compared with the existing distributed neurodynamic algorithm, the proposed algorithm has a simpler structure and has fewer neurons on the premise that the calculation error does not increase. Besides, the proposed algorithm converges to a minimal point of L-1-minimization problem and is Lyapunov stable. Finally, the comparative examples of sparse signal reconstruction show that the proposed distributed algorithm is effective and superior.

关键词： L-1-minimization problem sparse signal reconstruction distributed algorithm neurodynamic system

来源：评论

学校读者我要写书评

暂无评论

Coexisting attractors in a memcapacitor-based chaotic oscillator

引用

nonlinear DYNAMICS 2016年第1期86卷 37-50页

作者： Yuan, Fang Wang, Guangyi Shen, Yiran Wang, Xiaoyuan Hangzhou Dianzi Univ Key Lab RF Circuits & Syst Minist Educ China Inst Modern Circuits & Intelligent Informat Hangzhou 310018 Peoples R China

In this paper, a smooth curve model of memcapacitor and its equivalent circuit are designed. Based on this memcapacitor, a novel memcapacitive chaotic circuit is presented. Dynamical behaviors of the circuit with various parameters are investigated both theoretically and experimentally. The numerical results indicate that the circuit displays complex nonlinear properties including coexisting and symmetrical bifurcations. The main characteristic of this memcapacitive chaotic circuit is the various coexisting attractors. Different kinds of coexisting attractors and their corresponding conditions are given. The equilibrium set, Lyapunov exponent spectrum and the basin of attraction are also analyzed. Besides, experimental results are given to confirm the correction of the numerical simulations.

关键词： Chaos Chaotic oscillation Memcapacitor Coexisting attractors

来源：评论

学校读者我要写书评

暂无评论

Nystrom Kernel Algorithm Under Generalized Maximum Correntropy Criterion

引用

IEEE SIGNAL PROCESSING LETTERS 2020年 27卷 1535-1539页

作者： Zhang, Tao Wang, Shiyuan Southwest Univ Coll Elect & Informat Engn Chongqing Key Lab Nonlinear Circuits & Intelligen Chongqing 400715 Peoples R China

The kernel adaptive filters (KAFs) based on the minimum mean square error (MMSE) criterion in reproducing kernel Hilbert space (RKHS) improve the performance of linear adaptive filters but result in instability issues and large burdens of computation and memory in impulsive noises. To this end, a novel Nystrom kernel recursive generalized maximum correntropy (NKRGMC) with probability density rank-based quantization (PRQ) sampling (NKRGMC-PRQ) algorithm is proposed to improve filtering performance, robustness, and computational efficiency of the traditional KAFs in this letter. In a fixed dimensional network structure, the proposed NKRGMC-PRQ algorithm can achieve a comparable performance to KAFs with low computational complexity. Monte Carlo simulations are conducted to validate the superiorities of NKRGMC-PRQ in terms of filtering accuracy, computational complexity, and robustness.

关键词： Kernel Signal processing algorithms Sampling methods Approximation algorithms Computational complexity Training data Mathematical model Kernel adaptive filter Nystrom method kernel recursive generalized maximum correntropy PRQ sampling

来源：评论

学校读者我要写书评

暂无评论

Neurodynamic Network for Absolute Value Equations: A Fixed-Time Convergence Technique

引用

IEEE TRANSACTIONS ON circuits AND SYSTEMS II-EXPRESS BRIEFS 2022年第3期69卷 1807-1811页

作者： Ju, Xingxing Li, Chuandong Han, Xin He, Xing Southwest Univ Coll Elect & Informat Engn Chongqing Key Lab Nonlinear Circuits & Intelligen Chongqing 400715 Peoples R China

In this brief, a novel projection neurodynamic network with fixed-time convergence is proposed for solving absolute value equations. In contrast to most existing projection neurodynamic networks, a conservative settling time of the proposed projection neurodynamic network is presented. It is shown that the solution of the proposed approach converges to the solution of the corresponding absolute value equation in fixed-time under some mild conditions. Finally, a numerical example is presented to validate the main results.

关键词： Neurodynamics Convergence Mathematical models circuits and systems Optimization Circuit stability Numerical stability Fixed-time convergence projection neurodynamic network absolute value equations

来源：评论

学校读者我要写书评

暂无评论

Recursive Minimum Kernel Mixture Mean p-Power Error Algorithm Based on the Nystrx00F6;m Method

引用

IEEE TRANSACTIONS ON circuits AND SYSTEMS II-EXPRESS BRIEFS 2020年第11期67卷 2772-2776页

作者： Zhang, Tao He, Fuliang Zheng, Zhi Wang, Shiyuan Southwest Univ Chongqing Key Lab Nonlinear Circuits & Intelligen Coll Elect & Informat Engn Chongqing 400715 Peoples R China

In this brief, a novel kernel mixture mean $p$ -power error (KMP) criterion is proposed by combining the mixture of two Gaussian functions into the kernel function of kernel mean $p$ -power error. The mixture correntropy (MC) measure can be viewed as a special case of the KMP with $p = 2$ . And the KMP with an appropriate $p$ can provide better accuracy than MC for robust learning. Some properties of KMP are presented for discussion. The Nystrx00F6;m method is an efficient method for curbing the growth of network size of kernel adaptive filters (KAFs), and the recursive update form can improve the tracking ability of KAFs. To this end, we apply the Nystrx00F6;m method and recursive update form to the KMP criterion, generating a novel recursive minimum kernel mixture mean $p$ -power error algorithm based on the Nystrx00F6;m method (NysRMKMP). Monte Carlo simulations on chaotic time-series prediction illustrate the desirable accuracy and robustness of NysRMKMP.

关键词： Kernel Eigenvalues and eigenfunctions Cost function circuits and systems Adaptive systems Gaussian noise Bandwidth Kernel adaptive filter Nystrom method kernel mixture mean p -power error recursive update

来源：评论

学校读者我要写书评

暂无评论

Recursive Constrained Maximum Correntropy Criterion Algorithm for Adaptive Filtering

引用

IEEE TRANSACTIONS ON circuits AND SYSTEMS II-EXPRESS BRIEFS 2020年第10期67卷 2229-2233页

作者： Qian, Guobing Ning, Xiaohan Wang, Shiyuan Southwest Univ Chongqing Key Lab Nonlinear Circuits & Intelligen Coll Elect & Informat Engn Chongqing 400715 Peoples R China

Recently, the gradient based constrained maximum correntropy criterion (GCMCC) algorithm has received considerable attention since it provides superior performance to the traditional methods and is robust to the non-Gaussian noise. However, the convergence of GCMCC algorithm depends on the learning rate. With a large learning rate, GCMCC converges fast but achieves a high steady state mean square deviation (MSD), and vice versa. To balance the convergence rate and the MSD, we propose a novel recursive CMCC (RCMCC) algorithm in this brief by utilizing the matrix inversion lemma. Furthermore, we provide a computationally efficient version of RCMCC (ERCMCC) algorithm by using some approximations. More importantly, we provide the convergence analysis of the RCMCC algorithm, and derive the stability condition and theoretical MSD. The theoretical analysis and superiorities of RCMCC and ERCMCC are validated by simulations.

关键词： Signal processing algorithms Convergence Optimized production technology Approximation algorithms Computational complexity Filtering Constrained robust adaptive filter recursive convergence analysis

来源：评论

学校读者我要写书评

暂无评论

Quantised kernel least mean square algorithm with a learning vector strategy

引用

ELECTRONICS LETTERS 2020年第21期56卷 1146-1147页

作者： Zhang, Qiangqiang Wang, Shiyuan Southwest Univ Chongqing Key Lab Nonlinear Circuits & Intelligen Coll Elect & Informat Engn Chongqing 400715 Peoples R China

To decrease the network size of quantised kernel least mean square (QKLMS) dramatically, the QKLMS algorithm with an online learning vector strategy, which is named LV-QKLMS, is proposed in this Letter. The centres of the dictionary in LV-QKLMS are updated dynamically by the online learning vectors. Unlike QKLMS only updating the coefficient of the nearest centre to current input, LV-QKLMS updates the centres of the dictionary using the redundant input data accordingly. Thanks to the iterative learning vector, LV-QKLMS has a constantly changing quantisation area on the arrival of a new sample, and thus generates a superimposed quantisation area containing more redundant input data than QKLMS with a fixed quantisation area. In addition, the learning vector updated by a linear combination of input data and itself in the superimposed quantisation area guarantees the filtering accuracy of LV-QKLMS. Simulation results on the prediction of Mackey-Glass chaotic time series are presented to validate the effectiveness of LV-QKLMS.

关键词： time series least mean squares methods chaos vector quantisation learning (artificial intelligence) quantised kernel least mean square algorithm network size QKLMS algorithm online learning vector strategy centres dictionary online learning vectors nearest centre current input LV-QKLMS updates redundant input data iterative learning vector constantly changing quantisation area superimposed quantisation area fixed quantisation area

来源：评论

学校读者我要写书评

暂无评论

Crossing the Dimensional Divide with Optoelectronic Tweezers: Multicomponent Light-Driven Micromachines with Motion Transfer in Three Dimensions

引用

ADVANCED MATERIALS 2025年第17期37卷 e2417742页

作者： Li, Gong Xu, Bingrui Wang, Xiaopu Yu, Jiangfan Zhang, Yifan Fu, Rongxin Yang, Fan Gu, Hongcheng Huang, Yuchen Chen, Yujie Zhang, Yanfeng Wang, Zhuoran Shen, Guozhen Wang, Yeliang Xie, Huikai Wheeler, Aaron R. Li, Jiafang Zhang, Shuailong Beijing Inst Technol Sch Integrated Circuits & Elect Beijing 100081 Peoples R China Beijing Inst Technol Beijing Adv Innovat Ctr Intelligent Robots & Syst Sch Mechatron Engn Beijing 100081 Peoples R China Shenzhen Inst Artificial Intelligence & Robot Soc Shenzhen 518129 Guangdong Peoples R China Chinese Univ Hong Kong Sch Sci & Engn Shenzhen 518172 Peoples R China Chinese Acad Sci CAS Res Educ Ctr Excellence Mol Sci Key Lab Photochem Inst Chem Beijing 100190 Peoples R China Beijing Inst Technol Sch Med Technol Beijing 100081 Peoples R China Beijing Inst Technol Chongqing Inst Microelect & Microsyst Chongqing 400000 Peoples R China Southeast Univ Sch Biol Sci & Med Engn State Key Lab Digital Med Engn Nanjing 210096 Peoples R China Sun Yat Sen Univ Sch Elect & Informat Technol State Key Lab Optoelect Mat & Technol Guangzhou 510275 Peoples R China Univ Toronto Inst Biomed Engn Toronto ON M5S 3E1 Canada Univ Toronto Dept Chem Toronto ON M5S 3H6 Canada Univ Toronto Donnelly Ctr Cellular & Biomol Res Toronto ON M5S 3E1 Canada Beijing Inst Technol Sch Phys Beijing 100081 Peoples R China

Micromachines capable of performing diverse mechanical tasks in complex and constrained microenvironments are of great interest. Despite important milestones in this pursuit, until now, micromachines are confined to actuation within a single 2D plane due to the challenges of transferring motion across different planes in limited space. Here, a breakthrough method is presented to overcome this limitation: multi-component micromachines that facilitate 3D motion transfer across different planes. These light-driven 3D micromachines, fabricated using standard photolithography combined with direct laser writing, are assembled and actuated via programmable light patterns within an optoelectronic tweezers system. Utilizing charge-induced repulsion and dielectrophoretic levitation effects, the micromachines enable highly efficient mechanical rotation and effective inter-component motion transfer. Through this work, fascinating patterns of similarities are unveiled for the new microscale 3D systems when compared with the macro-scale world in which they live, paving the way for the development of micromechanical devices and microsystems with ever increasing functionality and versatility.

关键词： 3D motion transfer micromachine micromanipulation micromotor optoelectronic tweezers optically-induced dielectrophoresis

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：