检索结果-南通市图书馆

Complementary memtransistors for neuromorphic computing: How, what and why

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Semiconductors 2024年第6期45卷 64-80页

作者： Qi Chen Yue Zhou Weiwei Xiong Zirui Chen Yasai Wang xiangshui miao Yuhui He School of Integrated Circuits Huazhong University of Science and TechnologyWuhan 430074China Department of Electrical and Computer Engineering University of CaliforniaSan DiegoUSA Dipartimento di Elettronica e Informazione and *** Politecnico di MilanoMilanoItaly Department of Applied Physics Hong Kong Polytechnic UniversityHong KongChina

Memtransistors in which the source-drain channel conductance can be nonvolatilely manipulated through the gate signals have emerged as promising components for implementing neuromorphic computing.On the other side,it is known that the complementary metal-oxide-semiconductor(CMOS)field effect transistors have played the fundamental role in the modern integrated circuit technology.Therefore,will complementary memtransistors(CMT)also play such a role in the future neuromorphic circuits and chips?In this review,various types of materials and physical mechanisms for constructing CMT(how)are inspected with their merits and need-to-address challenges discussed.Then the unique properties(what)and poten-tial applications of CMT in different learning algorithms/scenarios of spiking neural networks(why)are reviewed,including super-vised rule,reinforcement one,dynamic vision with in-sensor computing,etc.Through exploiting the complementary structure-related novel functions,significant reduction of hardware consuming,enhancement of energy/efficiency ratio and other advan-tages have been gained,illustrating the alluring prospect of design technology co-optimization(DTCO)of CMT towards neuro-morphic computing.

关键词： complementary memtransistor neuromorphic computing reward-modulated spike timing-dependent plasticity remote supervise method in-sensor computing

维普期刊数据库

A self-selecting memory element based on a method of interconnected ovonic threshold switching device

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第3期67卷 319-320页

作者： Jinyu WEN Lun WANG Jiangxi CHEN Hao TONG xiangshui miao School of Integrated Circuits Huazhong University of Science & Technology Institute of Artificial Intelligence Huazhong University of Science & Technology Hubei Yangtze Memory Laboratories

Currently, the 3-D cross-point structure has been successfully applied in phase change memory(PCM) to achieve scalable and stackable memory arrays [1]. The ovonic threshold switching(OTS) selector consists of a memory element that is used to suppress the parasitic current in the 3-D cross-point arrays due to its highly nonlinear currentvoltage(I-V)characteristics which are significant for largescale memory arrays [2].

关键词：

Efficient implementation of majority-inverter graph logic and arithmetic functions with memristor arrays

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第7期67卷 345-346页

作者： Zhouchao GAN Dongdong ZHANG Chenyu ZHANG Yinghao MA xiangshui miao Xingsheng WANG School of Integrated Circuits & Wuhan National Lab. for Optoelectronics Huazhong University of Science and Technology Hubei Yangtze Memory Laboratories

Memristors are characterized by several outstanding features that permit them to function both as memory and logic units, enabling logic-in-memory (LIM) computing that potentially breaks the von Neumann bottleneck. Typically,in-memory logic is categorized into stateful logic and nonstateful logic based on the physical quantities of inputs and outputs.

关键词：

Reconfigurable and polarization-dependent optical filtering for transflective full-color generation utilizing low-loss phase-change materials

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Semiconductors 2024年第7期45卷 46-53页

作者： Shuo Deng Mengxi Cui Jingru Jiang Chuang Wang Zengguang Cheng Huajun Sun Ming Xu Hao Tong Qiang He xiangshui miao School of Integrated Circuits Huazhong University of Science and TechnologyWuhan 430074China Hubei Yangtze Memory Laboratories Wuhan 430205China School of Microelectronics Fudan UniversityShanghai 200433China

All-dielectric metasurface, which features low optical absorptance and high resolution, is becoming a promising candidate for full-color generation. However, the optical response of current metamaterials is fixed and lacks active tuning. In this work, we demonstrate a reconfigurable and polarization-dependent active color generation technique by incorporating low-loss phase change materials(PCMs) and CaF_2 all-dielectric substrate. Based on the strong Mie resonance effect and low optical absorption structure, a transflective, full-color with high color purity and gamut value is achieved. The spectrum can be dynamically manipulated by changing either the polarization of incident light or the PCM state. High transmittance and reflectance can be simultaneously achieved by using low-loss PCMs and substrate. The novel active metasurfaces can bring new inspiration in the areas of optical encryption, anti-counterfeiting, and display technologies.

关键词： structural color reconfigurable all-dielectric metasurfaces phase change material

维普期刊数据库

Low-time-complexity document clustering using memristive dot product engine

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2022年第2期65卷 225-234页

作者： Houji ZHOU Yi LI xiangshui miao Wuhan National Laboratory for Optoelectronics School of Optical and Electronic InformationHuazhong University of Science and Technology

Document clustering has been commonly accepted in the field of data analysis. Nevertheless, the challenging issues for the clustering are the massive similarity measurement operations in the von Neumann architecture which result in huge time consumption. Memristive in-memory computing provides a brand-new path to solve this problem. In this article, utilizing the memristive dot product engine, we demonstrate a cosine similarity accelerated document clustering method for the first time. The memristor-based clustering method lowers the time complexity from O(N· d) of the conventional algorithm to O(N) by executing similarity measurement in one step. Focused on the unit-length vectors, an in-situ normalization scheme for the stored vectors in the crossbar array is proposed to provide an efficient hardware training scheme and reduce the normalization steps during the clustering. Utilizing the BBCSport dataset as a benchmark, we further discussed the impact of the non-ideal factors in the memristors, including the available quantized states, the inevitable programming noise, and the device failure. Simulation results indicate that the 6-bit quantized states and 5% programming noise are acceptable for the document clustering tasks. Besides, high resistance states of the failure cells are recommended for higher performance clustering results.

关键词： linear-time clustering cosine similarity spherical $K$-means memristor in-memory computing

A memristive neural network based matrix equation solver with high versatility and high energy efficiency

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2023年第2期66卷 198-208页

作者： Jiancong LI Houji ZHOU Yi LI xiangshui miao School of Optical and Electronic Information Huazhong University of Science and Technology School of Integrated Circuits Huazhong University of Science and Technology Wuhan National Laboratory for Optoelectronics

As the main topic in modern scientific computing and machine learning tasks, matrix equation solving is suffering high computational latency and tremendous power consumption due to the frequent data movement in traditional von Neumann computers. Although the in-memory computing paradigms have shown the potential to accelerate the execution of solving matrix equations, the existing memristive matrix equation solvers are still limited by the low system versatility and low computation precision of the memristor arrays. In this work, we demonstrate a hybrid architecture for accurate, as well as efficient, matrix equation solving problems, where the memristive crossbar arrays are used for the parallel vector-matrix multiplication and the digital computer for accuracy. The linear neural-network solving(NNS) method is adopted here and its versatility for various types of matrix equations is proved. The weight-slice computation method is developed to perform the analog matrix multiplication with high efficiency and high robustness in the array. The solution results confirmed that typical matrix equations can be solved by this memristive matrix equation solver with high accuracy. Further performance benchmarking demonstrates that the generalized memristive matrix equation solver has low solving time-complexity while outperforming the state-of-the-art CMOS and in-memory processors.

关键词： matrix equation solving memristor linear neural network matrix-multiplication analog computing

Deep machine learning unravels the structural origin of mid-gap states in chalcogenide glass for high-density memory integration

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

InfoMat 2022年第6期4卷 109-120页

作者： Meng Xu Ming Xu xiangshui miao Wuhan National Laboratory for Optoelectronics School of Integrated Circuits&School of Optical and Electronic InformationHuazhong University of Science and TechnologyWuhanChina Hubei Yangtze Memory Laboratories WuhanChina

The recent development of three-dimensional semiconductor integration technology demands a key component-the ovonic threshold switching(OTS)selector to suppress the current leakage in the high-density memory chips.Yet,the unsatisfactory performance of existing OTS materials becomes the bottleneck of the industrial advancement.The sluggish development of OTS materials,which are usually made from chalcogenide glass,should be largely attributed to the insufficient understanding of the electronic structure in these materials,despite of intensive research in the past decade.Due to the heavy first-principles computation on disordered systems,a universal theory to explain the origin of mid-gap states(MGS),which are the key feature leading to the OTS behavior,is still lacking.To avoid the formidable computational tasks,we adopt machine learning method to understand and predict MGS in typical OTS materials.We build hundreds of chalcogenide glass models and collect major structural features from both short-range order(SRO)and medium-range order(MRO)of the amorphous cells.After training the artificial neural network using these features,the accuracy has reached~95%when it recognizes MGS in new glass.By analyzing the synaptic weights of the input structural features,we discover that the bonding and coordination environments from SRO and particularly MRO are closely related to MGS.The trained model could be used in many other OTS chalcogenides after minor modification.The intelligent machine learning allows us to understand the OTS mechanism from vast amount of structural data without heavy computational tasks,providing a new strategy to design functional amorphous materials from first principles.

关键词： chalcogenide glass machine learning mid-gap states ovonic threshold switching phasechange memory selector

All-van der Waals stacking ferroelectric field-effect transistor based on In2Se3 for high-density memory

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2023年第8期66卷 203-210页

作者： Xiaojie WANG Zeyang FENG Jingwei CAI Hao TONG xiangshui miao Wuhan National Laboratory for Optoelectronics School of Optical and Electronic InformationHuazhong University of Science and Technology Hubei Yangtze Memory Laboratories

High-density integration of ferroelectric field-effect transistors(Fe FETs) is hindered by factors such as interfacial states, short-channel effects, and ferroelectricity degradation in ultrathin films. Accordingly, the introduction of two-dimensional(2D) materials could effectively solve these problems. However,most current studies focus on the replacement of Si-based channels with 2D channels. Little progress has been made in addressing issues caused by bulk-phase ferroelectric gate layers, such as the unavoidable rough interfaces and the fading of ferroelectricity in ultrathin films. Herein, the 2D ferroelectric material In2Se3is introduced as the gate dielectric. Combined with 2D insulating h-BN and 2D channel Mo S2, an all-van der Waals(vd W) stacking Fe FET is fabricated to provide a straight solution for the abovementioned issues.First, the robust ferroelectric phase of In2Se3is verified in an ultrathin film case and a high-temperature case, which is outstanding among recently reported 2D ferroelectrics. Second, device-level out-of-plane ferroelectric polarization switching is achieved in the cross-structure device. Based on these results, In2Se3is adopted as the ferroelectric gate dielectric to fabricate all-vd W stacking Fe FETs. The subsequent transistor performance measurement on the fabricated Fe FETs indicates that the ferroelectric polarization of the In2Se3layer plays a dominating role in forming a counterclockwise hysteresis loop. Further pulse response measurements manifest the feasibility of nonvolatile channel conductance tuning of these devices with a proper pulse design. Our findings suggest that In2Se3is a suitable 2D ferroelectric gate material and that all-vd W stacking Fe FETs based on 2D ferroelectrics are promising in the application of high-density memory.

关键词： high-density memory ferroelectric field-effecttransistors two-dimensional ferroelectrics van der Waals In$_{2}$Se$_{3}$

A comprehensive study of device variability of sub-5 nm nanosheet transistors and interplay with quantum confinement variation

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2023年第2期66卷 287-288页

作者： Haowen LUO Ruihan LI xiangshui miao Xingsheng WANG School of Optical and Electronic Information Huazhong University of Science and Technology Hubei Yangtze Memory Laboratories School of Integrated Circuits Huazhong University of Science and Technology Wuhan National Laboratory for Optoelectronics Huazhong University of Science and Technology

Dear editor,Gate-all-around(GAA) nanosheet transistor(NST) is one of the most promising candidates to replace Fin FET as the next generation because of its best electrostatic integrity and gate control over the channel [1]. However, statistical variability induced by traditional statistical variation sources including random discrete dopants(RDD), line edge roughness(LER), metal gate granularity(MGG), and strong quantum confinement effect has severely limited the performance improvement of NST, which requires a comprehensive study.

关键词： Nanosheets