• Bulletin of the Korean Chemical Society
• /
• v.20 no.1
• /
• pp.35-41
• /
• 1999

Stability characteristics of hyperbolic reaction-diffusion equations with a reversible Brusselator model are investigated as an extension of the previous work. Intensive stability analysis is performed for three important parameters, Nrd, β and Dx, where Nrd is the reaction-diffusion number which is a measure of hyperbolicity, β is a measure of reversibility of autocatalytic reaction and Dx is a diffusion coefficient of intermediate X. Especially, the dependence on Nrd of stability exhibits some interesting features, such as hyperbolicity in the small Nrd region and parabolicity in the large Nrd region. The hyperbolic reaction-diffusion equations are solved numerically by a spectral method which is modified and adjusted to hyperbolic partial differential equations. The numerical method gives good accuracy and efficiency even in a stiff region in the case of small Nrd, and it can be extended to a two-dimensional system. Four types of solution, spatially homogeneous, spatially oscillatory, spatio-temporally oscillatory and chaotic can be obtained. Entropy productions for reaction are also calculated to get some crucial information related to the bifurcation of the system. At the bifurcation point, entropy production changes discontinuously and it shows that different structures of the system have different modes in the dissipative process required to maintain the structure of the system. But it appears that magnitude of entropy production in each structure give no important information related for states of system itself.

• MALSORI
• /
• no.46
• /
• pp.37-50
• /
• 2003

In this paper, we evaluate and compare the performance of speech/music discrimination based on various feature parameters and classifiers. As for feature parameters, we consider High Zero Crossing Rate Ratio (HZCRR), Low Short Time Energy Ratio (LSTER), Spectral Flux (SF), Line Spectral Pair (LSP) distance, entropy and dynamism. We also examine three classifiers: k Nearest Neighbor (k-NN), Gaussian Mixure Model (GMM), and Hidden Markov Model (HMM). According to our experiments, LSP distance and phoneme-recognizer-based feature set (entropy and dunamism) show good performance, while performance differences due to different classifiers are not significant. When all the six feature parameters are employed, average speech/music discrimination accuracy up to 96.6% is achieved.

• Journal of Magnetics
• /
• v.21 no.3
• /
• pp.468-475
• /
• 2016

In this article, entropy generation on MHD Williamson nanofluid over a porous shrinking sheet has been analyzed. Nonlinear thermal radiation and chemical reaction effects are also taken into account with the help of energy and concentration equation. The fluid is electrically conducting by an external applied magnetic field while the induced magnetic field is assumed to be negligible due to small magnetic Reynolds number. The governing equations are first converted into the dimensionless expression with the help of similarity transformation variables. The solution of the highly nonlinear coupled ordinary differential equation has been obtained with the combination of Successive linearization method (SLM) and Chebyshev spectral collocation method. Influence of all the emerging parameters on entropy profile, temperature profile and concentration profile are plotted and discussed. Nusselt number and Sherwood number are also computed and analyzed. It is observed that entropy profile increases for all the physical parameters. Moreover, it is found that when the fluid depicts non-Newtonian (Williamson fluid) behavior then it causes reduction in the velocity of fluid, however, non-Newtonian behavior enhances the temperature and nanoparticle concentration profile.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2013.06a
• /
• pp.1-2
• /
• 2013

KLT-AECQ 는 지각적인 성능 향상을 위하여 formant weighting 필터를 사용한다.Code Excited Linear Prediction(CELP) 코더는 사람의 음성신호를 압축하는 대표적인 방식이다. CELP 의 Rate-Distortion 성능을 향상 시키기 위해서 Karhunen-Loeve-Transform (KLT) 기반의 Classified Vector Quantization (KLT-CVQ) 방식이 제안되었으며, 이는 KLT 기반의 Adaptive Entropy-Constrained Quantization (KLT-AECQ) 방식으로 확장되었다. 기존의 KLT-AECQ 에서는 지각적인 성능 향상을 위하여 formant weighting 필터를 사용한다. 본 논문에서는 이 필터 대신에 Spectral Amplitude Warping (SAW) 필터를 적용함으로써, KLT-AECQ 코더의 지각적인 성능을 향상하였다.

• The Bulletin of The Korean Astronomical Society
• /
• v.28 no.2
• /
• pp.39-39
• /
• 2003

• The Bulletin of The Korean Astronomical Society
• /
• v.27 no.1
• /
• pp.38-38
• /
• 2002

• 전기의세계
• /
• v.34 no.2
• /
• pp.94-98
• /
• 1985

Autocorrelation에 관계되는 Fourier변환의 기본적인 성질을 이용하여 speckle잡음제거가 power spectrum estimation 문제와 같이 해석될 수 있다는 것을 보였고 spectral estimation 방법으로서 최대 entrppy방법을 사용하여 딴 방법들과 비교하여 볼때 좋은 결과를 얻었다. 앞으로 2차원 test object까지의 확장, 이 알고리즘의 각 파라메타들에 대한 sensitivity, optimal한 Hanning window크기 판단 기준으로서 normalized mean squared error(NMSE)를 사용하였다.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.48 no.1
• /
• pp.101-107
• /
• 2011

The background noises and distortions by channel are major factors that disturb the practical use of speech recognition. Usually, noise reduce the performance of speech recognition system DSR(Distributed Speech Recognition) based speech recognition also bas difficulty of improving performance for this reason. Therefore, to improve DSR-based speech recognition under noisy environment, this paper proposes a method which detects accurate speech region to extract accurate features. The proposed method distinguish speech and noise by using entropy and detection of spectral energy of speech. The speech detection by the spectral energy of speech shows good performance under relatively high SNR(SNR 15dB). But when the noise environment varies, the threshold between speech and noise also varies, and speech detection performance reduces under low SNR(SNR 0dB) environment. The proposed method uses the spectral entropy and harmonics of speech for better speech detection. Also, the performance of AFE is increased by precise speech detections. According to the result of experiment, the proposed method shows better recognition performance under noise environment.

• Journal of Korea Multimedia Society
• /
• v.17 no.9
• /
• pp.1064-1069
• /
• 2014

In this paper, we propose a novel approach to improve the performance of a voice activity detection(VAD) which is based on the adaptive band-partitioning with the likelihood ratio(LR). The previous method based on the adaptive band-partitioning use the weights that are derived from the variance of the spectral. In our VAD algorithm, the weights are derived from LR, and then the weights are incorporated with the entropy. The proposed algorithm discriminates the voice activity by comparing the weighted entropy with the adaptive threshold. Experimental results show that the proposed algorithm yields better results compared to the conventional VAD algorithms. Especially, the proposed algorithm shows superior improvement in non-stationary noise environments.

• Journal of Korean Society for Geospatial Information Science
• /
• v.15 no.3
• /
• pp.33-39
• /
• 2007

Hyperspectral remote sensing data contain plenty of information about objects, which makes object classification more precise. In this paper, we proposed a new spectral similarity measure, called Spectral Mutual Information (SMI) for hyperspectral image classification problem. It is derived from the concept of mutual information arising in information theory and can be used to measure the statistical dependency between spectra. SMI views each pixel spectrum as a random variable and classifies image by measuring the similarity between two spectra form analogy mutual information. The proposed SMI was tested to evaluate its effectiveness. The evaluation was done by comparing the results of preexisting classification method (SAM, SSV). The evaluation results showed the proposed approach has a good potential in the classification of hyperspectral images.