• The Journal of the Acoustical Society of Korea
• /
• v.23 no.7
• /
• pp.540-547
• /
• 2004

Since speech enhancement algorithms based on Kalman/Wiener filter require a priori knowledge of the noise and have focused on the minimization of the variance of the estimation error between clean and estimated speech signal, small estimation error on the noise statistics may lead to large estimation error. However, H/sub ∞/ filter does not require any assumptions and a priori knowledge of the noise statistics, but searches the best estimated signal among the entire estimated signal by applying least upper bound, consequently it is more robust to the variation of noise statistics than Kalman/Wiener filter. In this paper, we Propose a speech enhancement method using HMM and multi H/sub ∞/ filters. First, HMM parameters are estimated with the training data. Secondly, speech is filtered with multiple number of H/sub ∞/ filters. Finally, the estimation of clean speech is obtained from the sum of the weighted filtered outputs. Experimental results shows about 1dB∼2dB SNR improvement with a slight increment of computation compared with the Kalman filter method.

• Journal of the Korean Geographical Society
• /
• v.46 no.2
• /
• pp.197-211
• /
• 2011

Due to the complexity of spatial interaction and the necessity of spatial representation and modeling, aggregation of spatial interaction data is indispensible. Given this, the purpose of this paper is to evaluate the effects of modifiable areal unit problem (MAUP) on a spatial interaction model. Four aggregation schemes are utilized at eight different scales: 1) randomly select seeds of district and then allocate basic spatial units to them, 2) minimize the sum of population weighted distance within a district, 3) maximize the proportion of flow within a district, and 4) minimize the proportion of flow within a district. A simple Poisson regression model with origin and destination constraints is utilized. Analysis results demonstrate that spatial characteristics of residuals, parameter values, and goodness-of-fit of the model were influenced by aggregation scale and schemes. Overall, the model responded more sensitively to aggregation scale than aggregation schemes and the scale effect on the model was varied according to aggregation schemes.

• Science of Emotion and Sensibility
• /
• v.7 no.2
• /
• pp.141-148
• /
• 2004

This paper proposed a new time adapted threshold using the standard deviations of Wavelet coefficients after Wavelet transform by frame scale. The time adapted threshold is set up using the sum of standard deviations of Wavelet coefficient in cA3 and weighted cDl. cA3 coefficients represent the voiced sound with low frequency and cDl coefficients represent the unvoiced sound with high frequency. From simulation results, it is demonstrated that the proposed algorithm improves SNR and MSE performance more than Wavelet transform and Wavelet packet transform does. Moreover, the reconstructed signals by the proposed algorithm resemble the original signal in terms of plosive sound, fricative sound and affricate sound but Wavelet transform and Wavelet packet transform reduce those sounds seriously.

• Nuclear Engineering and Technology
• /
• v.27 no.4
• /
• pp.503-517
• /
• 1995

A mechanistic model for forced convective transition boiling has been developed to predict transition boiling heat flux realistically. This model is based on a postulated multi­stage boiling process occurring during the passage time of an elongated vapor blanket specified at a critical heat flux condition. Between the departure from nucleate boiling (DNB) and the departure from film boiling (DFB) points, the boiling heat transfer is established through three boiling stages, namely, the macrolayer evaporation and dryout governed by nucleate boiling in a thin liquid film and the unstable film boiling. The total heat transfer rate during the transition boiling is the sum of the heat transfer rates after the DNB weighted by the time fractions of each stage, which are defined as the ratio of each stage duration to the vapor blanket passage time. The model predictions are compared with some available experimental transition boiling data. From these comparisons, it can be seen that the transition boiling heat fluxes including the maximum heat flux and the minimum film boiling heat flux are nil predicted at low qualities/high pressures near 10 bar.

• Bulletin of the Korean Chemical Society
• /
• v.5 no.3
• /
• pp.117-121
• /
• 1984

The structure of $CS_{7.3}Ag_{4.7}Si_{12}Al_{12}O_{48}$, vacuum dehydrated zeolite A with all Na+ ions replaced by $Cs^+$ and $Ag^+$ as indicated, has been determined by single-crystal x-ray diffraction techniques in the cubic space group, Pm3m (a = 12.282 (1) ${\AA}$). The structure was refined to the final error indices $R_1$$R_2$ (weighted) = 0.099 using 347 independent reflections for whind intlch $I_0\;>\;3{\sigma}(I_0)$. Although deydration occurred at $360^{\circ}C$, no silver atoms or clusters have been observed. The 8-ring sites are occupied only by $Cs^+$ ion, and the 4-ring sites only by a single $Ag^+$ ion. The 6-ring sites contain $Ag^+$ and $Cs^+$ ions with $Ag^+$ nearly in 6-ring planes and $Cs^+$ well off them, one on the sodalite unit side. With regard to the 6-rings, the structure can be represented as a superposition of two types of unit cells: about 70 % have $4Ag^+$ and $4Cs^+$ ions, and the remaining 30 % have $3Ag^+$ and $5Cs^+$. In all unit cells, $3Cs^+$ ions lie at the centers of the 8-rings at sites of D4h symmetry; these ions are approximately 0.3 ${\AA}$ further from their nearest framework-oxygen neighbors than the sum of the appropriate ionic radii would indicate. To minimize electrostatic repulsions, the $Cs^+$ ions at Cs(1) are not likely to occupy adjacent 6-rings in the large cavity; they are likely to be tetrahedrally arranged when there are 4.

• Korean Journal of Remote Sensing
• /
• v.18 no.2
• /
• pp.117-125
• /
• 2002

Multitemporal analysis with remotely sensed data is complicated by numerous intervening factors, including atmospheric attenuation and occurrence of clouds that obscure the relationship between ground and satellite observed spectral measurements. Using an adaptive reconstruction system, dynamic compositing approach was developed to recover missing/bad observations. The reconstruction method incorporates temporal variation in physical properties of targets and anisotropic spatial optical properties into image processing. The adaptive system performs the dynamic compositing by obtaining a composite image as a weighted sum of the observed value and the value predicted according to local temporal trend. The proposed system was applied to the sequence of NDVI images of AVHRR observed on the Korean Peninsula from 1999 year to 2000 year. The experiment shows that the reconstructed series can be used as an estimated series with complete data for the observations including bad/missing values. Additionally, the gradient image, which represents the amount of temporal change at the corresponding time, was generated by the proposed system. It shows more clearly temporal variation than the data image series.

• Korean Journal of Remote Sensing
• /
• v.26 no.5
• /
• pp.581-591
• /
• 2010

In this paper, we proposed a new wavelet-based image fusion algorithm, which has advantages in both frequency and spatial domains for signal analysis. The developed algorithm compares the ratio of SAR image signal to optical image signal and assigns the SAR image signal to the fused image if the ratio is larger than a predefined threshold value. If the ratio is smaller than the threshold value, the fused image signal is determined by a weighted sum of optical and SAR image signal. The fusion rules consider the ratio of SAR image signal to optical image signal, image gradient and local variance of each image signal. We evaluated the proposed algorithm using Ikonos and TerraSAR-X satellite images. The proposed method showed better performance than the conventional methods which take only relatively strong SAR image signals in the fused image, in terms of entropy, image clarity, spatial frequency and speckle index.

• Journal of the Korea Society for Simulation
• /
• v.28 no.3
• /
• pp.25-40
• /
• 2019

Today, container terminals are fiercely competing to attract an increasing number of containers. As a way to improve terminal productivity, this study proposes two dispatching rules for yard truck allocation priorities. First, Multi-Attribute Dispatching Rule(MADR) is an allocation method to calculate the weighted sum of multiple factors affecting container terminal productivity and priority them. Especially, the workload of the quay crane was considered one of the factors to reduce the residence time of the ship. Second, Cycling Dispatching Rule(CDR) is the effective way to increase the number of double cycles that directly affect terminal productivity. To identify the effects of combinations of pooling and dispatching, a comparative experiments was performed on 8 scenarios that combined them. A simulation environment has been developed for experiments and the results have demonstrated that the combination of terminal level pooling and Multi-attribute Dispatching could be an excellent combination in KPIs consisting of GCR and delayed departure of ships, etc.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.63 no.6
• /
• pp.141-150
• /
• 2021

In order to assess the impact of climate change on irrigation reservoirs, climate exposure (EI), sensitivity (SI), and potential impact (PI) were evaluated for 1,651 reservoirs nationwide. Climate exposure and sensitivity by each reservoir were calculated using data collected from 2011 to 2020 for seven proxy variables (e.g. annual rainfall) and six proxy variables (e.g. irrigation days), respectively. The potential impact was calculated as the weighted sum of climate exposure and sensitivity, and was classified into four levels: 'Low (PI<0.4)', 'Medium (PI<0.6)', 'High (PI<0.8)', and 'Critical (PI≥0.8)'. The result showed that both the climate exposure index and the sensitivity index were on average high in Daegu and Gyeongbuk with high temperature and low rainfall. About 79.8% of irrigation reservoirs in Daegu, Gyeongbuk, and Ulsan with high climate exposure and sensitivity resulted in a 'High' level of potential impact. On the contrary, 64.5% of the study reservoirs in Gyeongnam and Gangwon showed 'Low' in potential impact. In further studies, it is required to reorganize the proxy variables and the weights in accordance with practical alternatives for improving adaptive capacity to drought, and it is expected to contribute to establishing a framework for vulnerability assessment of an irrigation reservoir.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.5
• /
• pp.496-502
• /
• 2021

Recently, deep speaker embedding approach has been widely used in text-independent speaker verification, which shows better performance than the traditional i-vector approach. In this work, to improve the deep speaker embedding approach, we propose a novel method called group-based speaker embedding which incorporates group information. We cluster all speakers of the training data into a predefined number of groups in an unsupervised manner, so that a fixed-length group embedding represents the corresponding group. A Group Decision Network (GDN) produces a group weight, and an aggregated group embedding is generated from the weighted sum of the group embeddings and the group weights. Finally, we generate a group-based embedding by adding the aggregated group embedding to the deep speaker embedding. In this way, a speaker embedding can reduce the search space of the speaker identity by incorporating group information, and thereby can flexibly represent a significant number of speakers. We conducted experiments using the VoxCeleb1 database to show that our proposed approach can improve the previous approaches.