http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
객체-관계형 데이터베이스와 관계형 데이터베이스의 벤치마크
김홍근,홍의경 서울시립대학교정보기술연구소 2001 정보기술연구소 논문집 Vol.3 No.-
The object-relational database(ORDB) is now viewed as target of migration for relational databases(RDB). ORDBs offer several features that RDBs do not have. These include a general type system, user defined functions/methods, and inheritance. Further, ORDBs provide support for set-valued attributes and composition(nested data). Although users can use RDBs to model set-valued attributes and composition, ORDBs make it much more natural and simple to model them. The objective of this paper is to establish performance differences between ORDBs and RDBs in supporting these two important data modeling facilities.
송오성,이상현 서울시립대학교정보기술연구소 2002 정보기술연구소 논문집 Vol.4 No.-
Recent electronic devices are integrated by multi-layered thin film of ㎚ to ㎛ order thickness. Characterization of mechanical bonding strength of each layer became important to avoid failure and damages during process such as silicon-on-insulator wafer pairs(SOI) and hermit packaging. We propose a new characterization method employing four-point-bending(FPB) method instead of conventional tensile test or razor blade crack opening test. We employed the FPB characterization to the Si/SiO₂∥Si₃N₄/Si wafer pairs with varying rapid thermal annealing temperature from 600~1200℃. FPB successfully characterized the bonding strength of SiO₂-Si₃N₄as well as mechanical property of each layer. Our results imply that we may characterized simultaneously the mechanical property and bonding strength of each layers of multi-layered electric devices with FPB method.
다분야통합최적설계를 위한 컴퓨팅기반구조의 연구동향분석 및 개발방향에 관한 연구
이세정 서울시립대학교정보기술연구소 2002 정보기술연구소 논문집 Vol.4 No.-
This paper makes a survey of MDO framework developments occurring worldwide and shows some guidelines for Korean developers to build design frameworks. The design framework is a software system which can produce another specialized software for industries to apply to their domain design problem which usually involves analyses, decision making, iterations, and design strategies. Using a framework for design tasks, becomes more popular than ever before due to the development of computing infrastructures such as distributed computing, efficient scripting engines, and so forth. This paper will show a snapshot of current technology and also suggest a way to go in the future.
MetaComputing을 위한 Two Computer System Paradigm의 성능 비교에 관한 연구
이용우 서울시립대학교정보기술연구소 2000 정보기술연구소 논문집 Vol.2 No.-
This paper studies the file access performance of two computer system paradigms for metacomputing : clustering computer systems (the distributed memory message passing computer systems) and the shared memory shared variable computer systems First, this paper explores file access performance of various design choices in the two paradigms. Second, this paper evaluates the effect computer communication speed growth. computing power growth and transaction size growth ell the file access performance of the two paradigms.
Distributed Monitoring and Control on LonWorks/IP Virtual Device Network
Song, Ki-Won,Kim, Jong-Hwi,Choi, Gi Sang 서울시립대학교정보기술연구소 2002 정보기술연구소 논문집 Vol.4 No.-
Requirements for Distributed Monitoring and Control Networks (DMCN) differ greatly from those of typical data networks. Specifically, any DMCN technology which employs a fieldbus protocol is different from IP network protocol TCP/IP. In general, one needs to integrate fieldbus protocol and TCP/IP to realize DMCN over IP network or internet, which can be viewed as Virtual Device Network (VDN). Interoperability between devices and equipments is essential to enhance the quality and the performance of predictive maintenance (PM). This paper suggests a basic framework for VDN using DMCN over IP network and a method to guarantee interoperability between devices and equipments.
박상규,김용철 서울시립대학교정보기술연구소 2000 정보기술연구소 논문집 Vol.2 No.-
We present a method of tracking vehicles with reconstruction of structure by template matching. This method separates vehicles and the background out of an Image sequence. The overall algorithm is as follows: First, the frame difference between the current frame and the previous frame is computed. Then, blobs are extracted where the absolute value of the difference is above a threshold.. Similar and closely positioned blobs are connected into a graph Motion is computed by graph matching of blobs. Approximate shapes of vehicles are obtained from the graph of blobs. Edges and vertexes extracted from the vehicle are matched with templates. Identification of the vehicle is based on the coherence with template. The proposed method was tested with real image sequences. The motion was detected with high accuracy and the overall performance was good
The Time Dependency Analysis of PDP Transparent Dielectric Layer With a new Firing system
Choi, Chang Hoon,Han, Sang Ho,Yun, Hae Sang,Kim, Bum Su,Park, Sun Woo 서울시립대학교정보기술연구소 2000 정보기술연구소 논문집 Vol.2 No.-
The transparent dielectric layer characteristics have been influenced b the firing process conditions. As firing process method, we reported tile rapid thermal sinterring(RTS) method to overcome the disadvantages such as low through put. In case of RTS system, the annealing tune was reduced about 1/5 without the loss of the optic transmittance rate compare with the conventional heating circulation method.[1] In addition, we confirmed the optic transmittance rate was changed acrorduig to annealing time of firing process with RTS system. As a result of experiment, the surface morphology of the dielectric layer was flattened and clean without pin holes, the optic transmittance rate of the transparent dielectric layer was increased with finng time. We verified it has to be needed at least 30min to acquire the the transmittance rate over 80%
토러스 및 하이퍼큐브 다중컴퓨터를 위한 고장감내형 링 임베딩 알고리즘에 관한 연구
박주영,김진석 서울시립대학교정보기술연구소 2000 정보기술연구소 논문집 Vol.2 No.-
Torus and Hypercube are attractive structures for parallel processing due to their symmetry and regularity To increase the reliability of torus/hypercube based systems and to allow their use in the presence of faulty nodes, efficient fault-tolerant schemes in torus/hypercubes are necessary. In this paper we present an algorithm for embedding rings in torus based multiprocessor network in e event of node failures Given any f≤6 faulty nodes, a ring which is optimal can be found in a 3-D torus. Furthermore, we obtain the following result in hypercube. The algorithm can tolerate up to θ(2^(n/2)) faults, and guarantee that given any f<(n-2k) 2 exp (k) faulty nodes, it can find a ring of size at least 2 exp (n)-2f for k=0 and 2 exp (n)-2 exp (2k) for k≥1 in an n-dimensional hypercube, where the variable k is a squeezed step number of hypercube. It improves over existing algorithms in the size of ring.
김희식,남택주 서울시립대학교정보기술연구소 1999 정보기술연구소 논문집 Vol.1 No.-
A magnetic ballast is developed and achieved high efficiency of luminousness by fluorescent lamps. The magnetic ballast adopted G-9 as core material. The shape of core is modified. The diameter of the coil is upgraded to 0.5 ㎜, and a new power-saving circuit is designed for the semiconductor ignition starter. The experimental results efficiency is increased 6.2 lm/Watt (7.6 % and the ballast efficiency factor(BEF) of 1.09(7.6 %). The prototype was tested through governmental testing procedure. A high efficient energy-using equipment (the second grade in the efficiency of energy consumption) was certified. The saving power of 17[W] is shown by lightning apparatus for fluorescent lamps