• Communications for Statistical Applications and Methods
• /
• 제19권6호
• /
• pp.885-898
• /
• 2012

In this paper, we consider Bayesian approaches to partially linear models, in which a regression function is represented by a semiparametric additive form of a parametric linear regression function and a nonparametric regression function. We make a comparative study on the performance of widely used Bayesian partially linear models in terms of empirical analysis. Specifically, we deal with three Bayesian methods to estimate the nonparametric regression function, one method using Fourier series representation, the other method based on Gaussian process regression approach, and the third method based on the smoothness of the function and differencing. We compare the numerical performance of three methods by the root mean squared error(RMSE). For empirical analysis, we consider synthetic data with simulation studies and real data application by fitting each of them with three Bayesian methods and comparing the RMSEs.

• Journal of the Optical Society of Korea
• /
• 제18권2호
• /
• pp.180-183
• /
• 2014

The surface of semiconductor solar cells, such as a-Si or CIGS (CuInGaSe) solar cells is not flat but textured in the microscopic domain. With textured surfaces, the optical reflectivity of a solar cell is different from that of flat surfaces in the wavelength region. In this paper, the effects of a textured surface on a CIGS solar cell are presented by RCWA (Rigorous Coupled Wave Analysis) method. The effect of incoherent light is also considered by RCWA with a Fourier analysis while conventional optical simulation uses the input light on the solar cell as coherent light. Using experimental results, the author showed that the RCWA method with a Fourier analysis is a proper method to simulate the optical properties of CIGS solar cells.

• Computational Structural Engineering : An International Journal
• /
• 제2권1호
• /
• pp.43-50
• /
• 2002

This paper presents a method to predict dynamic responses of floating bodies in the time domain. Because of the frequency-dependence of the radiation wave forces, the memory effect must be taken into account when the responses are evaluated in the time domain. Although the formulations firstly developed by Cummins (1962) have been well-known for this purpose, the effective numerical procedure has not been established yet. This study employs FFT (Fast Fourier Transform) algorithm to evaluate the memory effect function, and the equations of motion of an integro-differential type are solved by Newmark-β method. Numerical examples for a truncated circular cylinder have indicated the effectiveness of the proposed numerical procedure.

• Journal of the Optical Society of Korea
• /
• 제9권1호
• /
• pp.6-12
• /
• 2005

The perturbation approach is applied to solve the nonlinear Schrodinger equation, and its valid range has been determined by comparing with the results of the split-step Fourier method over a wide range of parameter values. With γ= 2㎞/sup -1/mW/sup -1/, the critical distance for the first order perturbation approach is estimated to be(equation omitted). The critical distance, Z/sub c/, is defined as the distance at which the normalized square deviation compared to the split-step Fourier method reaches 10/sup -3/. Including the second order perturbation will increase Z/sub c/ more than a factor of two, but the increased computation load makes the perturbation approach less attractive. In addition, it is shown mathematically that the perturbation approach is equivalent to the Volterra series approach, which can be used to design a nonlinear equalizer (or compensator). Finally, the perturbation approach is applied to obtain the sinusoidal response of the fiber, and its range of validity has been studied.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• 제12권2호
• /
• pp.150-160
• /
• 2002

The free vibration characteristics of the triple cylindrical shells filled with fluid are investigated. The triple cylindrical shells are filled with compressible fluid. The boundary condition is clamped at both ends. Analytical method is developed to evaluate natural frequencies of triple cylindrical shells using Sanders' shell theory and courier series expansion by Stokes' transformation. Their results are compared with those of finite element method to verify the validation of the method developed. The modal characteristics of shells filled with fluid at region 1, 2 and 3 are evaluated.

• ETRI Journal
• /
• 제34권2호
• /
• pp.268-271
• /
• 2012

The high peak-to-average power is one of the main drawbacks in OFDM systems. This letter proposes a partial shift mapping (PSM) method for peak power reduction in OFDM systems. By utilizing the properties of the discrete Fourier transform, the proposed method generates a set of candidate signals without additional complex multiplication and selects the one with minimum peak power for transmission. Analyses and simulations confirm that the PSM method achieves satisfactory peak power reduction performance and low complexity compared with other kindred methods, for example, selected mapping and partial transmit sequences.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 한국전기전자재료학회 1999년도 춘계학술대회 논문집
• /
• pp.169-172
• /
• 1999

Recently, the wavelet transform has been a new and powerful tool for signal processing. It is more suitable specially for the feature extraction and detection of non-stationary signals than traditional methods such as, the Fourier Transform(FT), the Fast Fourier Transform(FFT) and the Least Square Method etc. because of the characteristic of the multi-scale analysis and time-frequency domain localization. The wavelet transform has been developed for the analysis of PD pulse signal to raise in the progress of insulation degradation. In this paper, the wavelet transform was applied to one foundational method for feature extraction. For the obtain experimental data, a computer-aided partial discharge measurement system with a single acoustic sensor was used. If we are applying to the neural network method the accumulated data through the extracted feature, it is expected that we can detect the PD pulse signal in the insulation materials on the on-line.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 한국전산구조공학회 1996년도 봄 학술발표회 논문집
• /
• pp.79-86
• /
• 1996

A boundary element method which utilizes the fundamental solution in the half plane is developed to analyze the multi-layered elastic media. The objective of this study is to derive numerically the fundamental solutions and to apply those to the exterior multi-layered domain problems. To obtain numerical fundamental solutions of multi-layered structural system, the same number of solutions as that of layers in Fourier transform domain are employed. The numerical integration technique is used in order to inverse the Fourier transform solution to real domain. To verify the proposed boundary element method, two examples are treated: (1) a circular hole near the surface of a half plane; and (2) a circular cavity within one layer of four layered half plane.

• Journal of the Optical Society of Korea
• /
• 제19권2호
• /
• pp.165-168
• /
• 2015

A simple method to determine the focal length of a lens using the Talbot image is presented. This method uses only one grating, requiring neither Moir$\acute{e}$ fringe analysis nor the angle between the gratings. The original Fourier transform was used to access the spectrum beyond the limitation set of the usual fast Fourier transform to determine the (de)magnification accurately enough to be used for the focal length. A set of Talbot images simulated numerically with the Fresnel diffraction integral was used to demonstrate the method. For focal lengths between 5550 mm and 5650 mm, the mean difference between the focal lengths determined from the Talbot images and the true values was 3.3 mm with the standard deviation of the difference being 3.8 mm. The true focal lengths can be recovered with an accuracy of 0.06%.

• Journal of Industrial Technology
• /
• 제18권
• /
• pp.295-302
• /
• 1998

Classification of fingerprints is one of the major subjects on which many researchers have been studying for efficient identification. But fingerprints should be preprocessed in various ways prior to being classified. Factors such as the accuracy and the processing time should be considered in classification of fingerprints. In this paper, we propose a method for classifying fingerprints into several frequent patterns. This method consists of two stages. A fingerprint image is first converted to a skeleton form to find out the center. Then it is identified as a member of one of preclassified pattern by the frequency domain feature. Experiments show that the proposed method is quite useful in classifying fingerprints into typical patterns.