检索结果-南通市图书馆

CNN Accelerator Using Proposed Diagonal Cyclic Array for Minimizing Memory Accesses

Computers, Materials & Continua 2023年第8期76卷 1665-1687页

作者： Hyun-Wook Son Ali AAl-Hamid Yong-Seok Na Dong-Yeong Lee Hyung-Won Kim Department of Electronics College of Electrical and Computer EngineeringChungbuk National UniversityCheongju28644Korea Department of Electrical Engineering College of EngineeringAl-Azhar UniversityCairo11651Egypt

This paper presents the architecture of a Convolution Neural network(CNN)accelerator based on a newprocessing element(PE)array called a diagonal cyclic array(DCA).As demonstrated,it can significantly reduce the burden of repeated memory accesses for feature data and weight parameters of the CNN models,which maximizes the data reuse rate and improve the computation ***,an integrated computation architecture has been implemented for the activation function,max-pooling,and activation function after convolution calculation,reducing the hardware *** evaluate the effectiveness of the proposed architecture,a CNN accelerator has been implemented for You Only Look Once version 2(YOLOv2)-Tiny consisting of 9 ***,the methodology to optimize the local buffer size with little sacrifice of inference speed is presented in this *** implemented the proposed CNN accelerator using a Xilinx Zynq ZCU102 Ultrascale+Field Programmable Gate Array(FPGA)and ISE Design *** FPGA implementation uses 34,336 Look Up Tables(LUTs),576 Digital Signal Processing(DSP)blocks,and an on-chip memory of only 58 KB,and it could achieve accuracies of 57.92% and 56.42% mean Average Precession@0.5 thresholds for intersection over union(mAP@0.5)using quantized 16-bit and 8-bit full integer data manipulation with only 0.68% as a loss for 8-bit version and computation time of 137.9 and 69 ms for each input image respectively using a clock speed of 200 *** speeds are expected to be doubled five times using a clock speed of 1GHz if implemented in a silicon System on Chip(SoC)using a sub-micron process.

关键词： CNN accelerator systolic array memory optimization YOLOv2-tiny mAP@0.5

来源：

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

A Low-Power 12-Bit SAR ADC for Analog Convolutional Kernel of Mixed-Signal CNN Accelerator

引用

Computers, Materials & Continua 2023年第5期75卷 4357-4375页

作者： Jungyeon Lee Malik Summair Asghar HyungWon Kim Department of Electronics College of Electrical and Computer EngineeringChungbuk National UniversityCheongju28644Korea Department of Electrical and Computer Engineering COMSATS University IslamabadAbbottabad CampusUniversity RoadTobe CampAbbottabad22044Pakistan

As deep learning techniques such as Convolutional Neural networks(CNNs)are widely adopted,the complexity of CNNs is rapidly increasing due to the growing demand for CNN accelerator system-on-chip(SoC).Although conventional CNN accelerators can reduce the computational time of learning and inference tasks,they tend to occupy large chip areas due to many multiply-and-accumulate(MAC)operators when implemented in complex digital circuits,incurring excessive power *** overcome these drawbacks,this work implements an analog convolutional filter consisting of an analog multiply-and-accumulate arithmetic circuit along with an analog-to-digital converter(ADC).This paper introduces the architecture of an analog convolutional kernel comprised of low-power ultra-small circuits for neural network accelerator *** is an essential component of the analog convolutional kernel used to convert the analog convolutional result to digital values to be stored in *** work presents the implementation of a highly low-power and area-efficient 12-bit Successive Approximation Register(SAR)*** most other SAR-ADCs with differential structure;the proposed ADC employs a single-ended capacitor array to support the preceding single-ended max-pooling circuit along with minimal power *** SARADCimplementation also introduces a unique circuit that reduces kick-back noise to increase *** was implemented in a test chip using a 55 nm CMOS *** demonstrates that the proposed ADC reduces Kick-back noise by 40%and consequently improves the ADC’s resolution by about 10%while providing a near rail-to-rail dynamic rangewith significantly lower power consumption than conventional *** ADC test chip shows a chip size of 4600μm^(2)with a power consumption of 6.6μW while providing an signal-to-noise-and-distortion ratio(SNDR)of 68.45 dB,corresponding to an effective number of bits(ENOB)of 11.07 bits.

关键词： Convolution neural networks split-capacitor-based digital-toanalog converter(DAC) SAR analog-to-digital converter artificial intelligence system-on-chip analog convolutional kernel

来源：

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

Quasiparticle interference and impurity resonances on WTe2

引用

Nano Research 2020年第9期13卷 2534-2540页

作者： Hyeokshin Kwon Taehwan Jeong Samudrala Appalakondaiah Youngtek Oh Insu Jeon Hongki Min Seongjun Park Young Jae Song Euyheon Hwang Sungwoo Hwang Samsung Advanced institute of Technology Samsung Electronics Co.Suwon 16678Republic of Korea SKKU Advanced institute of Nano Technology(SAINT) Sungkyunkwan UniversitySuwon 16419Republic of Korea Department of Nano Engineering Sungkyunkwan UniversitySuwon 16419Republic of Korea Department of Physics and Astronomy Seoul National UniversitySeoul 08826Republic of Korea Department of Physics Sungkyunkwan UniversitySuwon 16419Republic of Korea Center for Integrated Nanostructure Physics(CINAP) Institute for Basic Science(IBS)Sungkyunkwan UniversitySuwon 16419Republic of Korea

Using scanning tunneling microscopy/spectroscopy(STM/STS),we examine quasiparticle scattering and interference properties at the surface of ***2,layered transition metal dichalcogenide,is predicted to be a type-ll Weyl *** Weyl fermion states in WTe2 emerge as topologically protected touching points of electron and hole pockets,and Fermi arcs connecting them can be visible in the spectral function on the *** probe the properties of surface states,we have conducted low-temperature STM/STS(at 2.7 K)on the surfaces of WTe2 single *** visualize the surface states of WTe2 with atomic scale *** surface states emerging from the bulk electron pocket have been identified and their connection with the bulk electronic states shows good agreement with *** show the interesting double resonance peaks in the local density of states appearing at localized *** low-energy resonant peak occurs near the Weyl point above the Fermi energy and it may be mixed with the surface state of Weyl points,which makes it difficult to observe the topological nature of the Weyl semimetal WTe2.

关键词： WTe2 Weyl semimetal quasi-particle interference scanning tunneling microscopy/spectroscopy

来源：

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

基于图形化铌酸钾钠纳米棒阵列的自供电压力传感器

引用

science China Materials 2023年第4期66卷 1494-1503页

作者：姜蕾卢梦瑞杨飘云范怡静黄浩熊娟王钊顾豪爽 John Wang Hubei Key Laboratory of Ferro-&Piezo-electric Materials and Devices Faculty of Physics and Electronic ScienceHubei UniversityWuhan 430062China Department of Materials science and Engineering National University of SingaporeSingapore 117574Singapore College of Physics and Electromechanical Engineering Hubei University of EducationWuhan 430205China Key Laboratory of Wireless Sensor network&Communication Shanghai Institute of Microsystem and Information TechnologyChinese Academy of SciencesShanghai 200050China

微/纳尺度压力传感器可以检测来自外部环境的压力,分析所施加力的位置,大小和方向.这种压力传感器在电子屏幕、电子皮肤、运动监测、人工触觉系统等多个领域都有很高的应用需求.本文组装了一种可实现二维映射,基于图案化铌酸钾钠纳米棒... 详细信息

微/纳尺度压力传感器可以检测来自外部环境的压力,分析所施加力的位置,大小和方向.这种压力传感器在电子屏幕、电子皮肤、运动监测、人工触觉系统等多个领域都有很高的应用需求.本文组装了一种可实现二维映射,基于图案化铌酸钾钠纳米棒阵列的压力传感器矩阵.水热合成的正交相铌酸钾钠纳米棒具有优异的柔性和弹性,同时具有较高的压电性能.因此在组装的压力传感器矩阵中,单个单元尺寸低至200μm,灵敏度可达0.20 V N^(-1),检测限低至20 g,且器件的稳定性高.空间分离的传感器单元能有效避免交叉干扰,使器件能准确地实现自驱动压力成像,精确地分析外部压力刺激.

关键词：压力传感器运动监测电子皮肤电子屏幕交叉干扰驱动压力单元尺寸自供电

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Physical analysis on improving the recovery accuracy of the Earth's gravity field by a combination of satellite observations in along-track and cross-track directions

引用

Chinese Physics B 2014年第10期23卷 633-640页

作者：郑伟许厚泽钟敏员美娟 State Key Laboratory of Geodesy and Earth’s Dynamics Institute of Geodesy and Geophysics Chinese Academy of Sciences College of science Wuhan University of Science and Technology

The physical investigations on the accuracy improvement to the measurement of the Earth＇s gravity field recovery are carried out based on the next-generation Pendulum-A/B out-of-plane twin-satellite formation in this paper. Firstly, the Earth＇s gravity field complete up to degree and order 100 is, respectively, recovered by the collinear and pendulum satellite formations using the orbital parameters of the satellite and the matching accuracies of key payloads from the twin GRACE satellites. The research results show that the accuracy of the Earth＇s gravity field model from the Pendulum-A/B satellite formation is about two times higher than from the collinear satellite formation, and the further improvement of the determination accuracy of the Earth＇s gravity field model is feasible by the next-generation Pendulum-A/B out-of-plane twin-satellite formation. Secondly, the Earth＇s gravity field from Pendulum-A/B complete up to degree and order 100 is accurately recovered based on the orbital parameters of the satellite （e.g., an orbital altitude of 400 km, an intersatellite range of 100 km, an orbital inclination of 89° and an orbital eccentricity of 0.001）, the matching accuracies of space- borne instruments （e.g. 10-6 m in the intersatellite range, 10-3 m in the orbital position, 10-6 m/s in orbital velocity, and 10-11 m/s2 in non-conservative force）, an observation time of 30 days and a sampling interval of 10 s. The measurement accuracy of the Earth＇s gravity field from the next-generation Pendulum-A/B out-of-plane twin-satellite formation is full of promise for being improved by about l0 times compared with that from the current GRACE satellite formation. Finally, the physical requirements for the next-generation Pendulum-A/B out-of-plane twin-satellite formation are analyzed, and it is proposed that the satellite orbital altitude be preferably designed to be close to 400±50 km and the matching precision of key sensors from the Pendulum-A/B mission be about one ord

关键词： twin pendulum-MB satellites collinear and pendulum satellite formations along-cross-track ob-servations physical analysis satellite gravity recovery

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Threshold Ring Signature Scheme Based on TPM

引用

China Communications 2012年第1期9卷 80-85页

作者： Gong Bei Jiang Wei Lin Li Li Yu Zhang Xing College of Computer science Beijing University of Technology Beijing 100124 P. R. China institute of Electronic Technology Information Engineering University Zhengzhou 450004 P. R. China State Key Laboratory of information Security Institute of Sottware Chinese Academy of Sciences Beijing Beijing 100049 P. R. China Key Laboratory of information and network Security 3rd Research Institute Ministry of Public Security Shanghai 201204 P. R. China

The conventional ring signature schemes cannot address the scenario where the rank of members of the ring needs to be distinguished, for example, in electronically commerce application. To solve this problem, we presented a Trusted Platform Module （TPM）-based threshold ring signature schen. Employing a reliable secret Share Distribution Center （SDC）, the proposed approach can authenticate the TPM-based identity rank of members of the ring but not track a specific member＇s identity. A subset including t members with the same identity rank is built. With the signing cooperation of t members of the subset, the ring signature based on Chinese remainder theorem is generated. We proved the anonymity and unforgeability of the proposed scheme and compared it with the threshold ring signature based on Lagrange interpolation polynomial. Our scheme is relatively simpler to calculate.

关键词： ring signature TPM threshold Chinese remainder theorem rank

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Measurement of the integrated luminosity of the Phase 2 data of the Belle Ⅱ experiment

引用

Chinese Physics C 2020年第2期44卷 1-12页

作者： F.Abudinén I.Adachi P.Ahlburg H.Aihara N.Akopov A.Aloisio F.Ameli L.Andricek N.Anh Ky D.M.Asner H.Atmacan T.Aushev V.Aushev T.Aziz K.Azmi V.Babu S.Baehr S.Bahinipati A.M.Bakich P.Bambade Sw.Banerjee S.Bansal V.Bansal M.Barrett J.Baudot A.Beaulieu J.Becker P.K.Behera J.V.Bennett E.Bernieri F.U.Bernlochner M.Bertemes M.Bessner S.Bettarini V.Bhardwaj F.Bianchi T.Bilka S.Bilokin D.Biswas G.Bonvicini A.Bozek M.Bračko P.Branchini N.Braun T.E.Browder A.Budano S.Bussino M.Campajola L.Cao G.Casarosa C.Cecchi D.Červenkov M.-C.Chang P.Chang R.Cheaib V.Chekelian Y.Q.Chen Y.-T.Chen B.G.Cheon K.Chilikin H.-E.Cho K.Cho S.Cho S.-K.Choi S.Choudhury D.Cinabro L.Corona L.M.Cremaldi S.Cunliffe T.Czank F.Dattola E.De La Cruz-Burelo G.De Nardo M.De Nuccio G.De Pietro R.de Sangro M.Destefanis S.Dey A.De Yta-Hernandez F.Di Capua S.Di Carlo J.Dingfelder Z.Doležal I.Domínguez Jiménez T.V.Dong K.Dort S.Dubey S.Duell S.Eidelman M.Eliachevitch T.Ferber D.Ferlewicz G.Finocchiaro S.Fiore A.Fodor F.Forti A.Frey B.G.Fulsom M.Gabriel E.Ganiev M.Garcia-Hernandez R.Garg A.Garmash V.Gaur A.Gaz U.Gebauer A.Gellrich J.Gemmler T.Geßler R.Giordano A.Giri B.Gobbo R.Godang P.Goldenzweig B.Golob P.Gomis P.Grace W.Gradl E.Graziani D.Greenwald C.Hadjivasiliou S.Halder K.Hara T.Hara O.Hartbrich K.Hayasaka H.Hayashii C.Hearty M.T.Hedges I.Heredia de la Cruz M.Hernández Villanueva A.Hershenhorn T.Higuchi E.C.Hill H.Hirata M.Hoek S.Hollitt T.Hotta C.-L.Hsu Y.Hu K.Huang T.Iijima K.Inami G.Inguglia J.Irakkathil Jabbar A.Ishikawa R.Itoh M.Iwasaki Y.Iwasaki S.Iwata P.Jackson W.W.Jacobs D.E.Jaffe E.-J.Jang H.B.Jeon S.Jia Y.Jin C.Joo J.Kahn H.Kakuno A.B.Kaliyar G.Karyan Y.Kato T.Kawasaki H.Kichimi C.Kiesling B.H.Kim C.-H.Kim D.Y.Kim S.-H.Kim Y.K.Kim Y.Kim T.D.Kimmel K.Kinoshita C.Kleinwort B.Knysh P.Kodyš T.Koga I.Komarov T.Konno S.Korpar D.Kotchetkov N.Kovalchuk T.M.G.Kraetzschmar P.Križan R.Kroeger J.F.Krohn P.Krokovny W.Kuehn T.Kuhr M.Kumar R.Kumar K.Kumara S.Kurz A.Kuzmin Y.-J.Kwon S.Lacaprara Y.-T.Lai C.La Licata K.Lalwani L.Lanceri J.S.Lange K.Lautenbach I.-S.Le Beihang University Beijing 100191 Brookhaven national Laboratory UptonNew York 11973 Budker institute of Nuclear Physics SB RAS Novosibirsk 630090 Centro de Investigacion y de Estudios Avanzados del Instituto Politecnico Nacional Mexico City 07360 Faculty of Mathematics and Physics Charles University12116 Prague Chiba University Chiba 263-8522 Consejo Nacional de Ciencia y Tecnología Mexico City 03940 Deutsches Elektronen–Synchrotron 22607 Hamburg Duke University DurhamNorth Carolina 27708 Earthquake Research institute University of TokyoTokyo 113-0032 Forschungszentrum Jülich 52425 Jülich Department of Physics Fu Jen Catholic UniversityTaipei 24205 Key Laboratory of Nuclear Physics and Ion-beam Application(MOE)and institute of Modern Physics Fudan UniversityShanghai 200443 II.Physikalisches Institut Georg-August-Universität Göttingen37073 Göttingen Gifu University Gifu 501-1193 The Graduate University for Advanced Studies(SOKENDAI) Hayama 240-0193 Gyeongsang national University Jinju 52828 Department of Physics and institute of Natural sciences Hanyang UniversitySeoul 04763 High Energy Accelerator Research Organization(KEK) Tsukuba 305-0801 J-PARC Branch KEK Theory CenterHigh Energy Accelerator Research Organization(KEK)Tsukuba 305-0801 Indian institute of science Education and Research Mohali SAS Nagar140306 Indian institute of Technology Bhubaneswar Satya Nagar 751007 Indian institute of Technology Hyderabad Telangana 502285 Indian institute of Technology Madras Chennai 600036 Indiana University BloomingtonIndiana 47408 institute for High Energy Physics Protvino 142281 institute of High Energy Physics Vienna 1050 institute of High Energy Physics Chinese Academy of SciencesBeijing 100049 institute of Particle Physics(Canada) VictoriaBritish Columbia V8W 2Y2 institute of Physics Hanoi Instituto de Fisica Corpuscular Paterna 46980 INFN Laboratori Nazionali di Frascati I-00044 Frascati INFN Sezione di Napoli I-80126 Napoli INFN Sezione di Padova I-35131 Padova INFN Sezione di Perugia I-06123 Perugia

From April to July 2018,a data sample at the peak energy of the T(4 S) resonance was collected with the Belle Ⅱ detector at the SuperKEKB electron-positron *** is the first data sample of the Belle Ⅱ *** Bhabha and digamma events,we measure the integrated luminosity of the data sample to be(496.3±0.3±3.0) pb-1,where the first uncertainty is statistical and the second is *** work provides a basis for future luminosity measurements at Belle Ⅱ.

关键词： luminosity Bhabha digamma Belle II

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

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：