• 한국구조물진단유지관리공학회지
• /
• 제19권3호
• /
• pp.27-32
• /
• 2015

• 한국주거학회논문집
• /
• 제15권6호
• /
• pp.107-113
• /
• 2004

Fire was and remains a great threat to historic buildings of wood. This threat is omnipresent and often results to irreparable property losses especially cultural heritages. Fire of historic buildings continues to increase every year. The purpose of this study is to preserve a life safety, historical heritage and architectural & cultural values by pursuing for fire safety in the historical buildings. This study shall employ two research methods: a literature review and field survey methods. The literature review shall focus on the observations of characteristics noted in fire codes. The field survey shall cover a heat analysis and fire protection equipment. A method is proposed to investigate fire codes related historical building, reported fire history Fire Codes and Regulations for modern buildings are not always applicable on the historic buildings. Old buildings, and some new buildings with architectural, historical or cultural importance may have values which cannot be measured on a monetary scale. Their protection against the damage caused by fire is considered to be one of the basic moral duties of society. It is strongly recommended, therefore, legislation for fire safety in historical buildings must be extended to a complementary relationship with 'active' fire protection measures, to provide safety in real historical building in real fire situations. The development and application of performance based Codes and the Fire Safety Engineering approach seems to be a rather reasonable and effective way to deal with this historical buildings.

• Journal of Communications and Networks
• /
• 제17권3호
• /
• pp.213-220
• /
• 2015

In this paper, we present a reliability-based iterative proportionality-logic decoding algorithm for two classes of structured low-density parity-check (LDPC) codes. The main contributions of this paper include: 1) Syndrome messages instead of extrinsic messages are processed and exchanged between variable nodes and check nodes, which can reduce the decoding complexity; 2) a more flexible decision mechanism is developed in which the decision threshold can be self-adjusted during the iterative process. Such decision mechanism is particularly effective for decoding the majority-logic decodable codes; 3) only part of the variable nodes satisfying the pre-designed criterion are involved for the presented algorithm, which is in the proportionality-logic sense and can further reduce the computational complexity. Simulation results show that, when combined with factor correction techniques and appropriate proportionality parameter, the presented algorithm performs well and can achieve fast decoding convergence rate while maintaining relative low decoding complexity, especially for small quantized levels (3-4 bits). The presented algorithm provides a candidate for those application scenarios where the memory load and the energy consumption are extremely constrained.

• Nuclear Engineering and Technology
• /
• 제50권1호
• /
• pp.151-158
• /
• 2018

The neutron powder diffractometer (NPD) is used to study a variety of technologically important and scientifically driven materials such as superconductors, multiferroics, catalysts, alloys, ceramics, cements, colossal magnetoresistance perovskites, magnets, thermoelectrics, zeolites, pharmaceuticals, etc. Monte Carlo-based codes are powerful tools to evaluate the neutronic behavior of the NPD. In the present study, MCNPX 2.6.0 and Vitess 3.3a codes were applied to simulate NPD facilities, which could be equipped with different optic devices such as pyrolytic graphite or neutron chopper. So, the Monte Carlo-based codes were used to simulate the NPD facility of the 5 MW Tehran Research Reactor. The simulation results were compared to the experimental data. The theoretical results showed good conformity to experimental data, which indicates acceptable performance of the Vitess 3.3a code in the neutron optic section of calculations. Another extracted result of this work shows that application of neutron chopper instead of monochromator could be efficient to keep neutron flux intensity higher than $10^6n/s/cm^2$ at sample position.

• Computers and Concrete
• /
• 제2권1호
• /
• pp.31-53
• /
• 2005

In the present structural codes the safety verification is based on a linear analysis of the structure and the satisfaction of ultimate and serviceability limit states, using a semi-probabilistic security format through the consideration of partial safety factors, which affect the action values and the characteristic values of the material properties. In this context, if a non-linear structural analysis is wanted a difficulty arises, because the global safety coefficient, which could be obtained in a straightforward way from the non-linear analysis, is not directly relatable to the different safety coefficient values usually used for the different materials, as is the case for reinforced concrete structures. The work here presented aims to overcome this difficulty by proposing a methodology that generalises the format of safety verification based on partial safety factors, well established in structural codes within the scope of linear analysis, for cases where non-linear analysis is needed. The methodology preserves the principal assumptions made in the codes as well as a reasonable simplicity in its use, including a realistic definition of the material properties and the structural behaviour, and it is based on the evaluation of a global safety coefficient. Some examples are presented aiming to clarify and synthesise all the options that were taken in the application of the proposed methodology, namely how to transpose the force distributions obtained with a non-linear analysis into design force distributions. One of the most important features of the proposed methodology, the ability for comparing the simplified procedures for second order effects evaluation prescribed in the structural codes, is also presented in a simple and systematic way. The potential of the methodology for the development and assessment of alternative and more accurate procedures to those already established in codes of practice, where non-linear effects must be considered, is also indicated.

• 한국통신학회논문지
• /
• 제22권3호
• /
• pp.444-453
• /
• 1997

This paper presents the prototype design of hardware/software cosynthesis system for pipelined application-specific instruction processors. Taking application programs in VHDL as inputs, the proposed system generates a pipelined instruction-set processor and the instruction sequences running on the generated machine. The design space of datapath and controller is defined by the architectural templates embedded in the system. Generating the intyermediate code adequate for parallelism analysis and extraction, the system converts it into assembly codes. Experimental results show the effectiveness of the proposed system.

• 한국멀티미디어학회논문지
• /
• 제8권12호
• /
• pp.1658-1668
• /
• 2005

This paper presents the bytecode-to-SIL translator which enables the execution of the java program in EVM(Embedded Virtual Machine) environment without JVM(java Virtual Machine), translating bytecodes produced by compiling java programs into SIL(Standard Intermediate Language) codes. EVM, what we are now developing, is a virtual machine solution that can download and execute dynamic application programs written in sequential languages like C language as well as object oriented languages such as C#, Java, etc. EVM is a virtual machine mounted on embedded systems such as mobile device, set-top box, or digital TV, and converts the application program into SIL, an assembly language symbolic form, and execute it. SIL is a virtual machine code for embedded systems, based on the analysis of existing virtual machine codes such as bytecode, MSIL, etc. SIL has such features as to accommodate various programming languages, and in particularly has an operation code set to accept both object-oriented languages and sequential languages. After compiling, a program written in java language is converted to bytecode, and also executed by JVM platform but not in other platform such as .NET, EVM platform. For this reason, we designed and implemented the bytecode-to-SIL translator system for programs written in java language to be executed in the EVM platform without JVM. This work improves the execution speed of programs, enhances the productivity, and provides an environment for programmers to execute application programs at various platforms.

• 대한전자공학회논문지SD
• /
• 제42권9호
• /
• pp.41-50
• /
• 2005

회로가 커짐에 따라 테스트 데이터양이 증가하고, 테스트 적용시간이 길어지고 있다. 따라서 테스트 데이터양과 테스트 적용시간을 줄이기 위해서, 테스트 데이터의 압축/복원을 위한 새로운 방법을 제안하고자 한다. 제안하는 방법은 시드 벡터를 생성할 때, 압축률을 높이기 위해 무상관비트를 사용하는 XOR 트리에 기반을 두고 있다. 시드 벡터가 생성이 되면, 2비트 길이를 가진 코드를 사용하여 시드를 병합한다. 이렇게 병합된 시드는 1 클럭 시간동안에 재사용될 수가 있어, 테스트 데이터 적용시간을 크게 감소시킬 수 있다 제안하는 방법의 효율성은 ISCAS '89 벤치 회로에 대한 실험 결과로 알 수 있다.

• Journal of Advanced Marine Engineering and Technology
• /
• 제40권1호
• /
• pp.45-49
• /
• 2016

수중에서의 음향 통신의 성능은 신호의 다중경로 전달과정에 의해 발생하는 지역 확산 현상으로 인하여 인접 심볼간 간섭의 영향을 받는다. 이러한 다중경로의 영향으로 신호는 왜곡되고 원활한 수신을 방해하게 된다. 이러한 다중경로 환경에서 본 논문에서는 수신 신호의 성능을 향상시키고자 수중통신에 적합한 반복부호를 설정하였다. 적용 가능한 반복부호로는 터보 부호와 BCJR 기반의 컨볼루션 부호가 있으며, 동일한 부호화율 및 비슷한 부호어 길이에서 터보부호의 성능을 비교 분석하였다. 본 논문에서는 경북 문경시 경천호에서의 실제 수중 실험을 통하여 두 가지 방식의 성능을 분석하였다. 송수신 거리가 400m 그리고 데이터 속도를 1Kbps에서 측정된 실제 데이터를 이용하였다. 실험 결과 컨볼루션 부호 기반의 BCJR 복호기는 81%의 성공률, 터보 부호 기반의 터보 등화기는 93%의 성공률을 나타내는 것으로 보아 터보 부호 기반의 터보 등화기가 성능이 약간 우수함을 알 수 있다.

• 산업공학
• /
• 제8권3호
• /
• pp.11-23
• /
• 1995

During the past few years, the productivity has been considered as one of the most important aims in developing softwares. Many practitioners and theorists are trying reuse ready made codes to increase the productivity of software development and the quality of their products. This paper introduces our experiences in developing a highly reusable class library and its reusing environment for building a database interface application. And the concepts, Structured Multi-Library and Dual Views of Class Inheritance, are also suggested, which consist our main conceptual framework in our work. Details in implementation are discussed in brief at the end.