|
|
|
题名
|
作者
|
年代
|
出处
|
被引量
|
| 1 | Accelerating DNN-based 3D point cloud processing for mobile computing显示文摘3D point cloud data,which are produced by various 3D sensors such as LIDAR and stereo cameras,have been widely deployed by industry leaders such as Google,Uber,Tesla,and Mobileye,for mobile robotic applications such as autonomous driving and humanoid robots.Point cloud data,which are composed of reliable depth information,can provide accurate location and shape characteristics for scene understanding,such as object recognition and semantic segmentation.However,deep neural networks(DNNs),which directly consume point cloud data,are particularly computation-intensive because they have to not only perform multiplication-and-accumulation(MAC)operations but also search neighbors from the irregular 3D point cloud data.Such a task goes beyond the capabilities of general-purpose processors in realtime to figure out the solution as the scales of both point cloud data and DNNs increase from application to application.We present the first accelerator architecture that dynamically configures the hardware onthe-fly to match the computation of both neighbor point search and MAC computation for point-based DNNs.To facilitate the process of neighbor point search and reduce the computation costs,a grid-based algorithm is introduced to search neighbor points from a local region of grids.Evaluation results based on the scene recognition and segmentation tasks show that the proposed design harvests 16.4x higher performance and saves 99.95%of energy than an NVIDIA Tesla K40 GPU baseline in point cloud scene understanding applications. | Bosheng LIU Xiaoming CHEN Yinhe HAN Jiajun LI Haobo XU Xiaowei LI | 2019 | Science China(Information Sciences)2019，62，11： | 4 |
| 2 | Breaking the von Neumann bottleneck:architecture-level processing-in-memory technology显示文摘The'memory wall'problem or so-called von Neumann bottleneck limits the efficiency of conventional computer architectures,which move data from memory to CPU for computation;these architectures cannot meet the demands of the emerging memory-intensive applications.Processing-in-memory(PIM)has been proposed as a promising solution to break the von Neumann bottleneck by minimizing data movement between memory hierarchies.This study focuses on prior art of architecture level DRAM PIM technologies and their implementation.The key challenges and mainstream solutions of PIM are summarized and introduced.The relative limitations of PIM simulation are discussed,as well as four conventional PIM simulators.Finally,research directions and perspectives are proposed for future development. | Xingqi ZOU Sheng XU Xiaoming CHEN Liang YAN Yinhe HAN | 2021 | Science China(Information Sciences)2021，64，6： | 4 |
| 3 | Enhancement of memory function in aged mice by a novel derivative of xanomeline显示文摘 | Yihui Cui Dong Wang Wen Si Wen Lv Yan Niu Xiaoping Lei Yinhe Hu Xiaohua Cao | 2008 | Cell Research2008，18，11： | 3 |
| 4 | Control synthesis for a class of nonlinear systems based on partition of unity显示文摘一条 partition-of-unity-based 途径被建议为非线性的系统的一个班导出一个近似模型。近似模型的精确被使用连续功能的连续性的模量分析。近似模型的系统稳定性被使用 Lyapunov 稳定性理论分析。为构造与追踪与追踪错误有关的性能追踪控制器的一个设计算法基于近似模型和统一方法的分区被给。 | Dongfang HAN Yinhe WANG Siying ZHANG | 2007 | 控制理论与应用(英文版)2007，5，2： | 2 |
| 5 | Effect of temperature on Lu’an bituminous char structure evolution in pyrolysis and combustion显示文摘In the process of pyrolysis and combustion of coal particles, coal structure evolution will be affected by the ash behavior, which will further affect the char reactivity, especially in the ash melting temperature zone. Lu’an bituminous char and ash samples were prepared at the N_(2) and air atmospheres respectively across ash melting temperature. A scanning electron microscope (SEM) was used to observe the morphology of char and ash. The specific surface area (SSA) analyzer and thermogravimetric analyzer were respectively adopted to obtain the pore structure characteristics of the coal chars and combustion parameters. Besides, an X-ray diffractometer (XRD) was applied to investigate the graphitization degree of coal chars prepared at different pyrolysis temperatures. The SEM results indicated that the number density and physical dimension of ash spheres exuded from the char particles both gradually increased with the increasing temperature, thus the coalescence of ash spheres could be observed obviously above 1100℃. Some flocculent materials appeared on the surface of the char particles at 1300℃, and it could be speculated that β-Si_(3)N_(4) was generated in the pyrolysis process under N_(2). The SSA of the chars decreased with the increasing pyrolysis temperature. Inside the char particles, the micropore area and its proportion in the SSA also declined as the pyrolysis temperature increased. Furthermore, the constantly increasing pyrolysis temperature also caused the reactivity of char decrease, which is consistent with the results obtained by XRD. The higher combustion temperature resulted in the lower porosity and more fragments of the ash. | Yandi ZHANG Yinhe LIU Xiaoli DUAN Yao ZHOU Xiaoqian LIU Shijin XU | 2021 | Frontiers in Energy2021，15，1： | 2 |
| 6 | Statistical characteristics of the surface ducts over the South China Sea from GPS radiosonde data显示文摘Based on the global position system(GPS) radiosonde data near the sea surface, the surface duct characteristics over the South China Sea(SCS) were statistically analyzed. The annual surface duct occurrence over the SCS was about 64%. Of the observed surface ducts, duct heights mainly distributed between 18 and 42 m, with M slopes in the range of –0.3 to –0.2 M units/m. Those ducts accounted for about 80% of the ducting cases. For the total profiles, the duct occurrences in a day changed slowly and were more than 60% in all times. The surface ducts formed more easily in the daytime than in the nighttime and most of the duct height were at bellow about 32 m.Additionally, The seasonal variation of the SCS ducts appeared to be evident, except that the mean duct thickness was almost constant, about 33 m for all seasons. The highest occurrence was about 71% in the autumn, followed by in the summer, spring and winter. In spring, their top-height existed more often at a height of more than 48 m.Their mean duct strength became stronger trend from spring to winter, with the M-slope in the range between–0.26 and –0.18 M units/m. Those results agreed well with other studies, provided considering the data resolution.The statistical analysis was reliable and gave the duct estimation for the SCS. Such duct climatology not only has important implications for communication systems and the reliability of the radar observation, but also can provide useful information to improve the accuracy of the meteorological radar measurements. | CHENG Yinhe ZHOU Shengqi WANG Dongxiao LU Yuanzheng YAO Jinglong | 2015 | Acta Oceanologica Sinica2015，34，11： | 2 |
| 7 | Analysis of a smoothing method for symmetric conic linear programming显示文摘 | Liu Yongjin Zhang Liwei Wang Yinhe | 2006 | Journal of Applied Mathematics and Computing2006，22，12： | 1 |
| 8 | Release and transformation of sodium during combustion of Zhundong coals显示文摘 | GUANGYU L CHANG'AN W YU Y XI J YINHE L DEFU C | 2015 | J Energy Inst2015，，： | 1 |
| 9 | ISO10303 compatible data model and its applications for PC configuration management 显示文摘 | YAO Yinhe TRAPPEY A J C | 2000 | Robotics and Computer Integrated Manufacturing2000，16，5： | 1 |
| 10 | TLR2andTLR4inautoimmunedisease:acomprehensivereview显示文摘 | LiuY YinH ZhaoM etal | 2014 | ClinicRevAllergImmunol2014，47，2： | 1 |
| 11 | Two-phase region in the DTAB/SL mixed surfactant system显示文摘 | YINH MAOM HUANGJ etal | 2002 | Langmuir2002，18，： | 1 |
| 12 | The Synchronization of Complex Dynamical Networks with Similar Nodes and Coupling Time-delay显示文摘 | Fan Yongqing Wang Yinhe Zhang Yun | 2013 | Applied Mathematics and Computation2013，219，12： | 1 |
| 13 | Body Waves Revealed by Spatial Stacking on Long-Term Cross-Correlation of Ambient Noise显示文摘Theoretical and experimental studies indicate that complete Green's Function can be retrieved from cross-correlation in a diffuse field. High SNR(signal-to-noise ratio) surface waves have been extracted from cross-correlations of long-duration ambient noise across the globe. Body waves, not extracted in most of ambient noise studies, are thought to be more difficult to retrieve from regular ambient noise data processing. By stacking cross-correlations of ambient noise in 50 km inter-station distance bins in China, western United States and Europe, we observed coherent 20–100 s core phases(Sc S, PKIKPPKIKP, PcP PKPPKP) and crustal-mantle phases(Pn, P, PL, Sn, S, SPL, SnS n, SS, SSPL) at distances ranging from 0 to 4 000 km. Our results show that these crustal-mantle phases show diverse characteristics due to different substructure and sources of body waves beneath different regions while the core phases are relatively robust and can be retrieved as long as stations are available. Further analysis indicates that the SNR of these body-wave phases depends on a compromise between stacking fold in spatial domain and the coherence of pre-stacked cross-correlations. Spatially stacked cross-correlations of seismic noise can provide new virtual seismograms for paths that complement earthquake data and that contain valuable information on the structure of the Earth. The extracted crustal-mantle phases can be used to study lithospheric heterogeneities and the robust core phases are significantly useful to study the deep structure of the Earth, such as detecting fine heterogeneities of the core-mantle boundary and constraining differential rotation of the inner core. | Kai Wang Yinhe Luo Kaifeng Zhao Limeng Zhang | 2014 | Journal of Earth Science2014，25，6： | 1 |
| 14 | EffectofstannumadGditionon microstructureofasGcastandasGextruded MgG5Lialloys显示文摘 | JingB YinH M YangQS etal | 2011 | TransactionsofNonferrousMetalsSocietyofChina2011，21，： | 1 |
| 15 | Fault Tolerance M echanism in Chip M any-core Processors显示文摘 | Zhang Lei Han Yinhe Li Huawei | 2007 | Tsinghua Science&Technology2007，12，1： | 1 |
| 16 | TheileriosisofsheepandgoatsinChina显示文摘 | LUOJX YINH | 1997 | TropicalAnimalHealthandProduction1997，，29： | 1 |
| 17 | ElectrochemicaldeterminationofmicroRNA-21basedongraphene,LNAintegratedmolecularbeacon,AuNPsandbiotinmultifunctionalbiobarcodesandenzymaticassaysystem显示文摘 | YinH ZhouY ZhangH etal | 2012 | BiosensorsBioelectron2012，33，1： | 1 |
| 18 | Crowd mo- nitoring using image processing 显示文摘 | DAVIES A C YIN VELASTIN S A | 1995 | Electronics and Communication Engineering Journal1995，7，1： | 1 |
| 19 | Pyrolysis and Combustion Characteristics of Coals in Oxyfuel Combustion 显示文摘 | WANG Chang'an ZHANG Xiaoming LIU Yinhe | 2012 | Applied Energy2012，97，26： | 1 |
| 20 | Paleo-Environment Reconstruction of the Middle Permian Lucaogou Formation,Southeastern Junggar Basin,NW China:Implications for the Mechanism of Organic Matter Enrichment in Ancient Lake显示文摘The Middle Permian Lucaogou Formation is the source rock and the main oil shale producing formation in the southeastern Junggar Basin.This study focused on the Lucaogou Formation exposed in two outcrop sections on the northern flank of the Bogda Mountain,namely the Jingjingzigou and Dalongkou sections.Here,we present integrated analysis of the sedimentology,major and trace elements,mineral components and total organic carbon contents.The paleo-environment was reconstructed including provenance,redox conditions,paleo-salinity,chemical weathering intensity and primary organic matter productivity.The results showed that the upper and lower units were deposited in distinct depositional environments with different organic matter accumulation mechanisms.The lower unit was characterized by low lake level,dry climate,fresh-brackish and well-oxygenated water.While during the deposition of the upper unit the lake level rose,climate turned wetter and the bottom water became less oxidized and much saltier.The mechanism of the organic matter accumulation is different for these two units.The preserved organic matters were mainly controlled by the primary productivity in the lower unit and by the redox conditions in the upper unit. | Dawei Cheng Chuanmin Zhou Zhijie Zhang Xuanjun Yuan Yinhe Liu Xingyu Chen | 2022 | Journal of Earth Science2022，33，4： | 1 |
您所在的机构暂未开通此部分的功能，请联系管理员