• International journal of advanced smart convergence
• /
• v.8 no.1
• /
• pp.75-81
• /
• 2019

Data imbalance problem is common and causes serious problem in machine learning process. Sampling is one of the effective methods for solving data imbalance problem. Over-sampling increases the number of instances, so when over-sampling is applied in imbalanced data, it is applied to minority instances. Under-sampling reduces instances, which usually is performed on majority data. We apply under-sampling and over-sampling to imbalanced data and generate sampled data sets. From the generated data sets from sampling and original data set, we construct a heterogeneous ensemble of classifiers. We apply five different algorithms to the heterogeneous ensemble. Experimental results on an intrusion detection dataset as an imbalanced datasets show that our approach shows effective results.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.14 no.1
• /
• pp.77-83
• /
• 2014

Over-sampled discrete wavelet transformation is one way to overcome the disadvantages of the standard wavelet transform of shift invariance even though it increases the number of subband signals. Non-separable based discrete wavelet transform is efficient that it satisfies shift invariance and directional selectivity. In this paper, since efficient over-sampled wavelet transform is possible in a two-dimensional image processing, we show that the proposed method is well applied with performance improvement of digital image and noise removal.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.65 no.8
• /
• pp.1418-1423
• /
• 2016

The purpose of this research is to investigate the effect of sampling frequency on the photoplethysmography (PPG) and to evaluate the performance of interpolation methods for under-sampled PPG. We generated down-sampled PPG using 10 kHz-sampled PPG then evaluated waveshape changes with correlation coefficient. Correlation coefficient was significantly decreased at 50 Hz or below sampling frequency. We interpolated the down-sampled PPG using four interpolation method-linear, nearest, cubic spline and piecewise cubic Hermitt interpolation polynomial - then evaluated interpolation performance. As a result, it was shown that PPG waveform that was sampled over 20 Hz could be reconstructed by interpolation. Among interpolation methods, cubic spline interpolation showed the highest performance. However, every interpolation method has no or less effect on 5 Hz sampled PPG.

• Korean Journal of Optics and Photonics
• /
• v.15 no.4
• /
• pp.369-374
• /
• 2004

In this paper, we propose a new tunable laser diode structure. The laser diode consists of a sampled-grating distributed feedback laser diode monolithically integrated with a sampled-grating distributed-Brags-Reflector. For a specific design, the possibility of continuous/discrete wavelength tuning over 27nm is confirmed by a numerical analysis using a split-step time domain model. Because the laser diode can be directly coupled with optical fiber without the intervention of the passive section, the laser diode exhibits higher output power than the conventional laser diode.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.5 no.1
• /
• pp.13-20
• /
• 2005

This paper addresses a new blind channel equalization method using fourth-order cumulants of channel inputs and a three-layer neural network equalizer. The proposed algorithm is robust with respect to the existence of heavy Gaussian noise in a channel and does not require the minimum-phase characteristic of the channel. The transmitted signals at the receiver are over-sampled to ensure the channel described by a full-column rank matrix. It changes a single-input/single-output (SISO) finite-impulse response (FIR) channel to a single-input/multi-output (SIMO) channel. Based on the properties of the fourth-order cumulants of the over-sampled channel inputs, the iterative algorithm is derived to estimate the deconvolution matrix which makes the overall transfer matrix transparent, i.e., it can be reduced to the identity matrix by simple recordering and scaling. By using this estimated deconvolution matrix, which is the inverse of the over-sampled unknown channel, a three-layer neural network equalizer is implemented at the receiver. In simulation studies, the stochastic version of the proposed algorithm is tested with three-ray multi-path channels for on-line operation, and its performance is compared with a method based on conventional second-order statistics. Relatively good results, withe fast convergence speed, are achieved, even when the transmitted symbols are significantly corrupted with Gaussian noise.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.1
• /
• pp.69-75
• /
• 2019

The standard application area of over-sampled discrete wavelet transform is noise removal technology for digital images. Comparing dual density discrete wavelet transform with dual tree discrete wavelet transform, we have almost similar characteristics. In this paper, several discrete wavelet transforms are accomplished on digital image existing with noise, noises are removed with threshold processing algorithm on subband, performance evaluation experiments of the reconstructed images are accomplished. If we decide appropriate threshold value, the effect noise removal is possible. In this paper, we can certified that the suggested algorithm of 3-direction separable processing with 2 dimension dual density discrete wavelet transform is superior to several experiment results.

• Proceedings of the IEEK Conference
• /
• 1998.10a
• /
• pp.981-982
• /
• 1998

We fabricated sampled fiber grating by double-exposure method. First, a short-period grating was written into the hydrogen-loaded single mode fiber and then the refractive index was modulated over it by an amplitude mask. It was observed that several transmission dips appear due to the index modulation. The thermal and strain responses were measured over $40-180^{\circ}C$ and $0-1800\mu\varepsilon,$ respectively. The dips have the same and linear sensitivity to both physical quantities over the range of measurement.

• The Korean Journal of Applied Statistics
• /
• v.24 no.3
• /
• pp.477-484
• /
• 2011

When an event of target variable is rare, a widespread strategy is to build a model on the sample that disproportionally over-represents the events, that is over-sampled. Using the data over-sampled from the original data set, the predicted values would be biased; however, it can be easily corrected to represent the population. In this study, we investigate into the relationship between the proportion of rare event on a data-mart and the model performance using real world data of a Korean credit card company. Also, we use the methods for adjusting of posterior probability for over-sampled data of the offset method and the weighted method. Finally, we compare the performance of the methods using real data sets.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.1
• /
• pp.61-66
• /
• 2009

This paper proposes an efficient sampled-data controller design technique for formation flying. To deal with the nonlinearity in the formation flying dynamics and to obtain a linear, rather than affine, model, we utilize the optimal linearization technique. The digital redesign technique is then developed based on the optimal linear model and formulated in terms of linear matrix inequalities. Simulation results show the advantage of the proposed methodology over the conventional controller emulation technique.

• Journal of Korea Multimedia Society
• /
• v.7 no.6
• /
• pp.781-790
• /
• 2004

This paper presents a fourth-order cumulants based iterative algorithm for blind channel equalization. It is robust with respect to the existence of heavy Gaussian noise in a channel and does not require the minimum phase characteristic of the channel. In this approach, the transmitted signals at the receiver are over-sampled to ensure the channel described by a full-column rank matrix. It changes a single-input/single-output (SISO) finite-impulse response (FIR) channel to a single-input/multi-output (SIMO) channel. Based on the properties of the fourth-order cumulants of the over-sampled channel outputs, the iterative algorithm is derived to estimate the deconvolution matrix which makes the overall transfer matrix transparent, i.e., it can be reduced to the identity matrix by simple reordering and scaling. Both a closed-form and a stochastic version of the proposed algorithm are tested with three-ray multi-path channels in simulation studies, and their performances are compared with a method based on conventional second-order cumulants. Relatively good results are achieved, even when the transmitted symbols are significantly corrupted with Gaussian noise.