• The KIPS Transactions:PartB
• /
• v.8B no.6
• /
• pp.641-650
• /
• 2001

Jini architecture's Network Plug and Work provides simple and flexible network environment for distributed applications. Through the Jini technology, facilities for dynamically registering and locating mobile agent services can be easily supported, as well as the services useful for activities of mobile agents can be dynamically supported. In this paper, we describe a Jini-based mobile agent system named JMoblet, which provides the basic functions of a mobile agent system such as creation, control, transfer, location and communication among agents. To increase the reliability of the system, it also provides exception handling and persistence of the mobile agent systems for reliability.

• The KIPS Transactions:PartC
• /
• v.18C no.6
• /
• pp.423-432
• /
• 2011

Recently, OpenFlow has been paid attention to as a fundamental technology which provides a function of virtualization and programmability in network. In Korea, deployment of OpenFlow networks in campuses and the interconnection between them through tunneling in layer 3 has been performed. However, the performance of the interconnected networks is decreased due to delay in IP layer. In this paper, we design and deploy nation-wide, not local, OpenFlow networks in a pure layer 2 environment over KREONET. After that, we do end-to-end Round-trip Time measurements and TCP/UDP performance tests in OpenFlow and normal networks, and do comparison and analysis on the test results. The results show that the nation-wide OpenFlow networks provide equal performance to normal networks except for the initial packet loss for UDP streaming. In regards to the performance decrease due to early UDP packet loss, we can mitigate it by implementing exceptional procedures in a controller which deal with the same continuous "Packet_in" events.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11a
• /
• pp.772-774
• /
• 2005

우리는 내장형 시스템에서 자바의 성능 문제를 해결하기 위해 수행시간 이전에(ahead-of-time) 자바 바이트코드를 기계어 코드로 변환하는 AOTC를 개발하였다. 우리가 개발한 AOTC는 자바 클래스 파일을 읽어서 C코드로 변환한 후에 이를 C컴파일러로 컴파일하는 방식으로 동작한다. 이러한 방법으로 짧은 기간안에 하드웨어 아키텍처에 종속적이지 않으면서도 안정적인 고성능의 자바 실행 환경을 구축할 수 있었다. 본 논문에서는 AOTC의 전체 구조에서부터 C코드를 생성하는 방법 및 예외 처리와 쓰레기 수집기(GC)를 지원하기 위한 방법들에 대해서 다룬다. 또한 C컴파일러에 의한 최적화의 한계를 극복하기 위해 자바에 특화된 최적화를 AOTC에 포함하였는데 이에 대해서도 설명한다. 우리의 AOTC를 Sun Microsystems의 J2ME CDC VM인 CVM위에 적용해본 결과 벤치마크에 따라서 평균 $5\~14$배의 성능 향상을 관찰 할 수 있었다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.340-342
• /
• 2005

임베디드 소프트웨어의 적용영역이 확장됨에 따라 학계와 업계에서 임베디드 소프트웨어 개발 기술에 대한 관심이 고조되고 있다. UML 2.0은 산업체에서 많이 사용되는 모델링 언어로, 그 동안 현장에서는 주로 상태머신 다이어그램(State Machine Diagrams)을 사용하여 임베디드 소프트웨어의 동적 행위를 모델링하여 왔다. 그러나 모델러는 상태머신 다이어그램보다 시퀀스 다이어그램(Sequence Diagrams)을 선호하는데, 이는 시퀀스 다이어그램을 사용하는 것이 직관적이고 정확한 행위 모델을 개발할 수 있기 때문이다. UML 2.0이 최근 다양한 모델링 요소를 반영하도록 확장되었음에도 불구하고, 시퀀스 다이어그램을 사용하여 임베디드 소프트웨어를 모델링을 하기에는 아직 부족한 점이 있다. 이를 보완하고, 임베디드 소프트웨어를 더욱 잘 디자인하기 위하여 예외 상황과 인터럽트를 처리할 수 있는 방법을 제안한다.

• Annual Conference on Human and Language Technology
• /
• 1999.10e
• /
• pp.167-172
• /
• 1999

구어는 프로그래밍 언어와는 달리 주어진 문장 내에서의 해당 어휘의 뜻(semantic information)을 알고 다른 어휘들과의 연관성 (grammatical information)을 알아야만 적절한 형태소분석이 가능하다. 또한 구어는 방대한 양의 어휘들로 구성되어 있으며 사용하는 사람마다의 다양한 응용과 공식화되기 어려운 수많은 예외들로 운용되기 때문에 단순히 찾아보기표와 오토마타만으로는 형태소분석에 한계가 있다. 이에 본 논문에서는 주어진 어휘집과 그 어휘들로 만들어진 다양한 문장들로부터 구어운용의 근본기제를 스스로 학습해나가는 강화학습중심의 언어모델을 제안하고 실제로 한국어 형태소분석에 적용하여 그 성능과 특성을 파악해보았다. 구어파서의 입력은 음절단위의 발음이며 인간이 문장을 듣거나 보는 것과 동일하게 시간에 따라 순차적으로 입력된다. 파서의 출력 또한 시간에 따라 변화되면서 나타나며 입력된 연속음절을 형태소단위로 분리(segmentation)하고 분류(labeling)한 결과를 나타낸다. 생성인식 언어모델이 기존의 언어모델과 다른 점은 구어 파싱에 있어서 필수적인 미등륵어에 대한 유연성과 앞단의 음성인식기 오류에 적절한 반응(fault tolerance)을 나타내는 것이다.

• Tunnel and Underground Space
• /
• v.23 no.3
• /
• pp.212-218
• /
• 2013

Tuff specimens were thermally treated with predetermined temperatures of 200, 400 and $600^{\circ}C$ to construct specimens simulating weathered tuff rocks. Specific gravity, absorption ratio, elastic wave velocity, uniaxial compressive strength, Brazilian tensile strength, Young's modulus, Poisson's ratio and slake-durability index were measured for pre-heated specimens. Heating of rock specimens entailed the degradation of material properties except for slake-durability index. It was found that correlations among P-wave velocity, uniaxial compressive strength, Brazilian tensile strength, Young's modulus and absorption ratio are high. Regression equations which use the P-wave velocity as an independent variable were presented to evaluate uniaxial compressive strength, Brazilian tensile strength, Young's modulus and absorption ratio.

• KSCI Review
• /
• v.2 no.1
• /
• pp.45-54
• /
• 1995

This study is, by assigning ASCII codes hardly used to Bandaljum(ˇ) and making the fonts of Korean-English character mode, to design the way of converting automatically a word, a sentence or a document of korean into phonetic letters by applying the algorismized phonological principles inputted as a letter string, even though a user do not konw the basic principles of the usage of Korean-to-Romanization notation rule. This is designed so that it may be possible to turn into a mechanical code with reference to the corresponding character in the table of Korean-to-Romanization notation rule that is the currently used standard proposition of the government. Consequently this program makes it user more convenient in the manipulations of special case words, the assistance of colorful-screen or pull-down, pop-up menu and the adoptation of utilizable mouse works for a user convienency. This program could be installed in a single diskette of 5.25"(2HD) and be made in C programming language to mplement various font, expansion or condense of font, alternative printing.ting.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.5
• /
• pp.653-662
• /
• 2022

Recently, various researches have been made on a fingerprint classification method using Convolutional Neural Networks (CNN), which is widely used for multidimensional and complex pattern recognition such as images. The CNN-based fingerprint classification method can be executed by integrating the two-step process, which is generally divided into feature extraction and classification steps. Therefore, since the CNN-based methods can automatically extract features of fingerprint images, they have an advantage of shortening the process. In addition, since they can learn various features of incomplete or low-quality fingerprints, they have flexibility for feature extraction in exceptional situations. In this paper, we intend to identify the research trends of CNN-based fingerprint classification and discuss future direction of research through the analysis of experimental methods and results.

• Journal of Korea Spatial Information System Society
• /
• v.4 no.2 s.8
• /
• pp.55-64
• /
• 2002

VEfficient generation of road data is one of the most important issues in GIS (Geographic Information System). In this paper, we propose a hybrid approach for automatic generation of road data by combining mobile mapping and image processing techniques. Mobile mapping systems have a form of vehicle equipped with CCD camera, GPS, and INS. They can calculate absolute position of objects that appear in acquired image by photogrammetry, but it is labor-intensive and time-consuming. Automatic road detection methods have been studied also by image processing technology. However, the methods are likely to fail because of obstacles and exceptive conditions in the real world. To overcome the problems, we suggest a hybrid method for automatic road generation, by exploiting both GPS/INS data acquired by mobile mapping system and image processing algorithms. We design an estimator to estimate 3-D coordinates of road line and corresponding location in an image. The estimation process reduces complicated image processing operations that find road line. The missing coordinates of road line due to failure of estimation are obtained by cubic spline interpolation. The interpolation is done piecewise, separated by rapid change such as road intersection. We present experimental results of the suggested estimation and interpolation methods with image sequences acquired by mobile mapping system, and show that the methods are effective in generation of road data.

• The KIPS Transactions:PartB
• /
• v.9B no.3
• /
• pp.263-270
• /
• 2002

Most conventional systems of fire fighting control in a ship have been based on rule-based system in which expert knowledges are expressed with production rules. Renewing and adding of rules is needed continuously for the improvement of the system capability in an already build-up system and such adding and renewing procedures could hinder users from fluent utilization of a system. The author proposes an advanced fire fighting control intelligent system (A-FFIS) using rule-based and carte-based reasoning by clustering to implement conventional hybrid system (H-FFIS). Compared with H-FFIS, new approach with A-FFIS shows that the system proposed here improves fire detection rate and reduces fire detection time.