• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.30 no.3 s.246
• /
• pp.260-268
• /
• 2006

The micromechanical approach was used to investigate the interfacial strain distributions of a unidirectional composite under transverse loading in which fibers were usually found to be randomly packed. Representative volume elements (RVE) for the analysis were composed of both regular fiber arrays such as a square array and a hexagonal array, and a random fiber array. The finite element analysis was performed to analyze the normal, tangential and shear strains at the interface. Due to the periodic characteristics of the strain distributions at the interface, the Fourier series approximation with proper coefficients was utilized to evaluate the strain distributions at the interface for the regular and random fiber arrays with respect to fiber volume fractions. From the analysis, it was found that the random arrangement of fibers had a significant influence on the strain distribution at the interface, and the strain distribution in the regular fiber arrays was one of special cases of that in the random fiber array.

• Economic and Environmental Geology
• /
• v.29 no.3
• /
• pp.345-355
• /
• 1996

The effectiveness of various electrode configurations in horizontal mappings and 1-D inversions of vertical sounding data for delineating 2-D structures was studied. Apparent resistivity values of three point, dipole-dipole, Wenner, and Schlumberger mappings were simulated for such structures as vertical dyke, tabular prism, buried vertical fault, ramp and complex structure by finite difference method (FDM) and they were compared with each other. Also 2-D cross sections for three structures obtained by interpolation of 1-D inverted sounding data in terms of three layers were compared for Schlumberger and Wenner arrays. On these cross sections, horizontal and vertical resistivity interfaces of the 2-D structures are revealed relatively clearly. Apparent resistivity curves of Schlumberger mapping show vertical resistivity discontinuities very well. On the whole, Schlumberger array is superior to the other arrays in electric sounding as well as mapping. This study clearly indicates that interpretations of 2-D structures based on 1-D inversion are possible.

• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.361-366
• /
• 2004

보청기에서 마이크로폰을 여러 개 배치하여 소음 감소 효과를 가져 올 수 있다. 본 논문에서는 양측 귀에 각각 두개씩의 보청기용 마이크로폰을 배치했을 때의 효과를 경계 요소 기법에 의해 해석하였다. 두개의 마이크로폰을 배열하여 마이크로폰 두개 사이의 시간 지연에 의한 특별한 지향성 패턴을 만들게 되고, 이러한 지향성 패턴은 기하학적으로 소음 대 잡음의 비를 증가시킬 수 있다. 두 개의 마이크로폰 배열에 의한 3차원 지향성 패턴 해석에 경계요소기법이 응용되었으며, 이러한 수치 해석은 보청기 DSP칩의 시간 지연 파라미터 계산에 이용될 수 있다. 두 개 마이크로폰사이의 간격은 10mm로 고정시키고, 머리의 전 방향에 폭이 좁은 지향패턴이 생기도록 두 개 마이크로폰의 사이 시간 지연을 바꾸어 주었다. 시간 지연의 변화를 입력 주파수에 따라 관찰하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.10a
• /
• pp.129-132
• /
• 2000

버퍼넘침(buffer overflow)은 특정 프로그램 언어에서 발생하는 배열의 경계파괴 현상을 말한다. 그 대표적인 언어로서 C/C++을 들 수 있는데. 이들 언어는 기본적으로 스트링(문자열)을 정의함에 있어서 크기속성을 배제하고 끝을 의미하는 종료문자(delimiter character. NULL)를 사용함으로써 배열(버퍼)의 경계침범 가능성을 허용하고 있다. 이 때 스택영역에 할당된 버퍼가 넘친다면, 주변에 위치한 지역변수, 레지스터 보관, 복귀주소 둥의 값이 변질되어 원래의 의도된 제어흐름을 보장할 수 없게 된다. 특히 복귀주소 부분을 의도적으로 침범하여 특정 값을 덮어쓸 수 있다면 해당 프로그램의 동작을 인위적으로, 그리고 자유롭게 변경할 수 있게 된다. 본 논문에서는 이와 같은 스택영역에서의 버퍼넘침을 사용한 제어흐름 변경 해킹기법의 과정을 현존하는 UNIX 시스템 및 C/C++ 언어를 이용하여 살펴보고 대응방향을 모색한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10a
• /
• pp.653-655
• /
• 2000

버퍼넘침(buffer overflow)은 특정 프로그램 언어에서 발생하는 배열의 경계파괴 현상을 말한다. 그 대표적인 언어로서 C/C++을 들 수 있는데, 이들 언어는 기본적으로 스트링(문자열)을 정의함에 있어서 크기속성을 배제하고 끝을 의미하는 종료문자(delimiter character, NULL)을 사용함으로써 배열(버퍼)의 경계침범 가능성을 허용하고 있다. 이때 스택영역에 할당된 버퍼가 넘침다면, 주변에 위치한 지역변수, 레지스터 보관, 복귀주소 등의 값이 변질되어 원래의 의도된 제어흐름을 보장할 수 없게 된다. 특히 복귀주소 부분을 의도적으로 침범하여 특정 값을 덮어쓸 수 있다면 해당 프로그램의 동작을 인위적으로, 그리고 자유롭게 변경할 수 있게 된다. 본 논문에서는 이와 같은 스택영역에서의 버퍼넘침을 사용한 제어흐름 변경 해킹기법의 과정을 현존하는 UNIX 시스템 및 C/C++ 언어를 이용하여 살펴보고 대응방향을 모색한다.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.6
• /
• pp.1407-1417
• /
• 1996

In recent researches on thread partitioning algorithms break a thread at the long latency operation and merge threads to get the longer threads under the given constraints. Due to this limitation, even a program with little parallelism is partitioned into small-sized threads and context-swithings occur frequently. In the paper, we propose another method array localization about the array name, dependence distance(the difference of accessed element index from loop index), and the element usage that indicates whether element is used or defined. Using this information we can allocate array elements to the node where the corresponding loop activation is executed. By array localization, remote accesses to array elements can be replaced with local accesses to localized array elements. As a resuit,the boundaries of some threads are removed, programs can be partitioned into the larger threads and the number of context switchings reduced.

• The KIPS Transactions:PartC
• /
• v.9C no.4
• /
• pp.453-458
• /
• 2002

Recently, the number of hacking, that use buffer overflow vulnerabilities, are increasing. Although the buffer overflow Problem has been known for a long time, for the following reasons, it continuos to present a serious security threat. There are three defense method of buffer overflow attack. First, allow overwrite but do not allow unauthorized change of control flow. Second, do not allow overwriting at all. Third, allow change of control flow, but prevents execution of injected code. This paper is for allowing overwrites but do not allow unauthorized change of control flow which is the solution of extending compiler. The previous defense method has two defects. First, a program company with overhead because it do much thing before than applying for the method In execution of process. Second, each time function returns, it store return address in reserved memory created by compiler. This cause waste of memory too much. The new proposed method is to extend compiler, by processing after compiling and linking time. To complement these defects, we can reduce things to do in execution time. By processing additional steps after compile/linking time and before execution time. We can reduce overhead.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.8
• /
• pp.2222-2232
• /
• 1999

This paper proposed a Systolic Arrays architecture for computing edge detection on images. It is very difficult to be processed images to real time because of operations of local operators. Local operators for computing edge detection are to be used in many image processing tasks, involve replacing each pixel in an image with a value computed within a local neighborhood of that pixel. Computing such operators at the video rate requires a computing power which is not provided by conventional computer. Through computationally expensive, it is highly regular. Thus, this paper presents a systolic arrays for tasks such as edge detection and laplacian, which are defined in terms of local operators.

• Journal of Sensor Science and Technology
• /
• v.12 no.2
• /
• pp.79-87
• /
• 2003

In this paper, we propose a method to obtain a linear characteristic using soft computing for systems which have array sensors of nonlinear characteristics. Also a procedure utilizing the pattern information of array sensors without additional sensors is proposed to reduce disturbance effects. For a typical example, even a single CdS cell for CdS array has nonlinear characteristics. Overall linear characteristic for CdS array is obtained using fuzzy logic for each cell and overlapped portion. In addition, further improvement for linearization is obtained applying genetic algorithms for the parameters of membership functions. Also the effect of disturbing external light changes to the CdS array can be reduced without using any additional sensors for calibration. The proposed method based on fuzzy logic shows improvements for position measurements and disturbance reduction to external light changes due to the fuzziness of the shadow boundary as well as the inherent nonlinearity of the CdS array. This improvement is shown by applying the proposed method to the ball position measurements of a magnetic levitation system.

• Korean Journal of Optics and Photonics
• /
• v.9 no.5
• /
• pp.307-314
• /
• 1998

We have studied the effects of interface on polarization dependence of waveguide and designed the polarization independent 2-dimensional waveguides on InP for arrayed waveguide grating. To figure out the effects of interface on polarization dependence, we have solved the 1-dimensional wave equation using a transfer matrix method. It is shown that the birefringence becomes stronger as the number of interfaces increases. In order to design polarization independent 2-dimensional waveguides, we have used effective index method. The structures considered are ridge type, raised strip type, and buried type waveguide. In the cases of ridge and raised strip type, conventional effective index method was used. In the case of buried type corrected effective index method was used. We have determined the height and width of waveguides such that the waveguides become polarization independent.