• Proceedings of the Korean Statistical Society Conference
• /
• 2000.11a
• /
• pp.135-140
• /
• 2000

베이지안 신경망 모형(Bayesian Neural Networks Models)에서 주어진 입력값(input)은 블랙 박스(Black-Box)와 같은 신경망 구조의 각 층(layer)을 거쳐서 출력값(output)으로 계산된다. 새로운 입력 데이터에 대한 예측값은 사후분포(posterior distribution)의 기대값(mean)에 의해 계산된다. 주어진 사전분포(prior distribution)와 학습데이터에 의한 가능도함수(likelihood functions)를 통해 계산되어진 사후분포는 매우 복잡한 구조를 갖게 됨으로서 기대값의 적분계산에 대한 어려움이 발생한다. 이때 확률적 추정에 의한 근사 방법인 몬테칼로 적분을 이용한다. 이러한 방법으로서 Hybrid Monte Carlo 알고리즘은 우수한 결과를 제공하여준다(Neal 1996). 본 논문에서는 Hybrid Monte Carlo 알고리즘과 기존에 많이 사용되고 있는 Gibbs sampling, Metropolis algorithm, 그리고 Slice Sampling등의 몬테칼로 방법들을 비교한다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38A no.3
• /
• pp.224-239
• /
• 2013

An exact performance analysis of an ML detector for a 2-dimensional rotation-transform aided QPSK system operating over an impulsive noise environment is presented using Rieman integrals of a two-dimensional Gaussian Q-function over Voronoi cells. A set of interesting features of the Voronoi cells is also characterised systematically. An optimum rotation angle yielding the minimum BER is also studied. The differences between the proposed exact method and the previous approximate analysis method are investigated in terms of the corresponding BERs and the derived optimum angles.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.22 no.5
• /
• pp.83-86
• /
• 1985

An alternative technique (or the numerical solution of Fredholm integral equations of second kind is presented. The approximate solution is obtained by fitting the data in mixed form at knots in the region of the problem. To decrease the error in the numerical solution, cubic B-spline functions which are twice continuously differentiable at knots are employed as basis function. For a given example, the results of this technique are compared with those of Moment method employing pulse functions for basis function and delta functions for test function and found to br in good agreement.

• Proceedings of the KIEE Conference
• /
• 1994.11a
• /
• pp.42-44
• /
• 1994

This paper presents the method of estimating integral coefficients of new distance relaying algorithm for transmission line protection. The proposed method is based on the differential equation calculates impedance value by approximation of integral term of integro-differential equation which relate voltage with current. As a result, we can determine the integral coefficients in least square error sense in frequency domain and we take into consideration the analog filter characteristics and frequency domain characteristics of the system to be protected. The simulation results showed that these coefficients can be successfully used to obtain impedance value in distance relay.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.43 no.5
• /
• pp.711-720
• /
• 1994

A distance relaying algorithm for detecting faults at power transmission line is presented in this paper. The algorithm is based on differential equation from relaton between voltage and current, which is composed of lumped resistance and inductance. During the fault transient state,the voltage and current signals are severely distorted due to the exponentially decaying DC offset and high frequency components, In spite of using small data, the presented integral method to evaluate R and L from voltage and current has high performance against these harmonics including DC offset. Therefore, the presented algorithm can be implemented with only a low order anti-aliasing analog filter and dosen't need any digital filter to remove specific components.

• Proceedings of the KIEE Conference
• /
• 1993.07a
• /
• pp.180-182
• /
• 1993

A distance protection algorithm for detecting faults at power transmission lines is presented in this paper. The algorithm is based on the differential equation related to the voltage and the current at an equivalent circuit of a transmission line which is composed of the lumped resistance and inductance. The presented integration method has high performance at though the fault voltage and the current are heavily distorted with the DC offset and harmonics which occurred at transient states after faults.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.8 no.1
• /
• pp.72-80
• /
• 1996

Two analytical solutions for oscillations in a rectangular harbour are presented. In this paper, the correct solution is obtained by use of matched asymptotic expansion method, which was first derived by Mei(1989). The other solution derived from eigenfunction expansion method is also presented, in which more accurate numerical integration is employed. In order to check the solutions, amplification factors inside the harbor are calculated and plotted by both analytical methods and numerical boundary integral equation method.

• Economic and Environmental Geology
• /
• v.18 no.2
• /
• pp.159-166
• /
• 1985

The computation of gravity anomalies by multi pole expansion is derived and compared with exact calculation for right rectangular prisms and right circular cylinders. For sources near field points, the multipole expansion results in a better approximation in volume integrals than in surface integrals. Nonetheless two approximate methods are coincident in the far-field of the general geophyical prospecting.

• The KIPS Transactions:PartA
• /
• v.11A no.5
• /
• pp.401-406
• /
• 2004

Among many algorithms for the integration problems in which one wants to compute the approximation to the definite integral in the average case setting, we study the average case errors of numerical integration rules using interpolation. In particular, we choose the composite Newton-Cotes quadratures and the function values at equally spaced sample points on the given interval as information. We compute the average case error of composite Newton-Cotes quadratures and show that it is minimal(modulo a multiplicative constant).

• Journal of Advanced Marine Engineering and Technology
• /
• v.29 no.5
• /
• pp.509-518
• /
• 2005

Liquid film thickness in laminar film condensation for flow over a flat plate generally is so thin that both fluid acceleration and thermal convection within the liquid film can be neglected. An integral solution method is proposed to solve the conjugate problems of laminar film condensation and heat conduction in a solid wall. It is found that approximate solutions of the governing equations involve four physical parameters to describe the conjugate heat transfer problem for laminar film condensation. It is shown that the effects of interfacial shear. mass transfer and local heat transfer are strongly dependent on the thermo-physical properties of the working fluids and the Jacob number.