• Speech Sciences
• /
• v.8 no.4
• /
• pp.59-72
• /
• 2001

In this paper, we experimented with speaker change detection that uses a statistical method for NOD (News On Demand) service. A specified speaker's change can find out content of each data in speech if analysed because it means change of data contents in news data. Speaker change detection acts as preprocessor that divide input speech by speaker. This is an important preprocessor phase for speaker tracking. We detected speaker change using GLR(generalized likelihood ratio) distance base division and BIC (Bayesian information criterion) base division among matrix method. An experiment verified speaker change point using BIC base division after divide by speaker unit using GLR distance base method first. In the experimental result, FAR (False Alarm Rate) was 63.29 in high noise environment and FAR was 54.28 in low noise environment in MDR (Missed Detection Rate) 15% neighborhood.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.22 no.3
• /
• pp.163-170
• /
• 2010

Both scatterer method and transfer matrix method are compared to analyze their characteristics, which are wave propagation models due to topographic change based on plane wave approximation. Results from the scatterer method are closer to the results obtained by the more accurate existing models and it is appraised that the scatterer method gives the clearer explanation about physical process involved in the wave transformation. Since both methods have analytical solutions, in the computational point of view they are very fast and easy to be implemented. Both methods give a good prediction for wave scattering by relatively simple bedform.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.21 no.2
• /
• pp.44-55
• /
• 2018

The importance of establishing a greenhouse gas inventory is emerging for policymaking and its implementation to cope with climate change. Thus, it is needed to establish Approach 3 level Land Use, Land-Use Change and Forestry (LULUCF) matrix that is spatially explicit regarding land use classifications and changes. In this study, four types of spatial information suitable for establishing the LULUCF matrix were analyzed - Cadastral Map, Land Cover Map, Forest Map, and Biotope Map. This research analyzed the classification properties of each type of spatial information and compared the quantitative and qualitative characteristics of the maps in Boryeong city. Drawn from the conclusions of the quantitative comparison, the forest area showed the maximum difference of 50.42% ($303.79km^2$) in the forest map and 46.09%($276.65km^2$) in the cadastral map. The qualitative comparison drew five qualitative characteristics: data construction scope difference, data construction purpose difference, classification standard difference, and classification item difference. As a result of the study, it was evident that the biotope map was the most appropriate spatial information for the establishment of the LULUCF matrix. In addition, if the LULUCF matrix is made by integrating the biotope, the forest map, and the land cover map, the limitations of each spatial information would be improved. The accuracy of the LULUCF matrix is expected to be improved when the map of the level-3 land cover map and the biotope map of 1:5,000 covering the whole country are completed.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2003.05a
• /
• pp.59-64
• /
• 2003

The parameter modification of the initial FEM model to match it with the experimental results needs the modal information and the modal sensitivity matrix to the parameter change. There are two cases this methodology is ill-equip to deal with; the deficiency of the necessary modal information and the ill-conditioning of the sensitivity matrix. In this research, a novel concept of the feedback exciter that uses the summation of the white noise and the signals from the measurement sensors multiplied with feedback gains as the reference signal is proposed. There are 2 advantages using this external feedback excitation. First, we can use the change of the system response such as modal data by the active energy Path from the sensor to the exciter. This change of the system response can be additional clues to the system dynamics that we want to know. Secondly, the external energy Path alternates the offset of the Parameter change to the system response. That means the modal sensitivity of the parameters becomes different from the original sensitivities by the feedback excitation. Through the feedback loop, we can change the similar modal sensitivities of some updating parameters and consequently discriminate the parameters using the closed-loop modal data. To demonstrate the discrimination performance, the parameter estimation of an indeterminate structure by use of the feedback method is introduced.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.22 no.11
• /
• pp.1144-1151
• /
• 2012

A damage in structure alters its dynamic characteristics. The change is characterized by changes in the modal parameter, i.e., modal frequencies, modal damping value and mode shape associated with each modal frequency. Changes also occur in some of the structural parameters; namely, the mass, damping, stiffness matrices of the structure. In this paper, evaluation of changes in stiffness matrix of a structure is presented as a method not only for identifying the presence of the damage but also locating the damage. It is shown that changed stiffness matrix can be accurately estimated a sensitivity coefficient matrix derived from modifying mode shapes, First, with 4 story shear structure models, the effect of presence of damage in a structure on its stiffness matrix is studied. By using these analytical model, the effectiveness of using change of stiffness matrix in detecting and locating damages is demonstrated. To validate the predicted changing stiffness and its location, the obtained results are compared to the reanalysis result which shows good agreement.

• Transactions of the Society of Information Storage Systems
• /
• v.5 no.1
• /
• pp.47-52
• /
• 2009

Most hard disk drives that apply the ramp load/unload technology unload the heads at the outer edge of the disk while the disk is rotating. The load/unload includes the benefits as like an increased areal density, a reduced power consumption and an improved shock resistance. A lot of papers investigating the effects of the various load/unload parameters such as a suspension tab, a limiter, a ramp and air-bearing surface designs have been published. However, in previous researches, an effect of the suspension is not considered at each load/unload step. In this paper, we focus that a variation of the state matrix affects the load/unload performance on based on a state matrix that is a stiffness matrix of the suspension. Because the state matrix is related to the suspension at each load/unload step, to change the state matrix means the structural change of the suspension. Therefore, we investigated a range of a pitch static attitude(PSA) and a roll static attitude(RSA) for load/unload performance. We also analyzed an effect of the variation of the state matrix a range of load/unload velocity occurred a slider-disk contact. We determined the variation of the state matrix to improve the load/unload performance through comparison of each factor of state matrix.

• Journal of Korean Biological Nursing Science
• /
• v.3 no.2
• /
• pp.91-103
• /
• 2001

The bone is composed of the bone matrix of collagen and hydroxyapatite, the mixture of calcium and phosphours. The bone tissue is considered to the special connective tissue that possesses extracellular matrix made by collagen fiber deposited with mineral complex. In order to maintain bone mass measured by the sum of bone matrix and hydroxyapatite, bone resorption by osteoclast during lifetime and bone remodeling to form bone by osteoblast in its resorption region repeat continuously. The osteoblast has a mesodermic fetal origin like fibroblast for the formation of form tissues. Two cells express identical genes and synthesize the identical collagen type I as the major component of the formation of bone matrix and skin. Therefore, it is considered that the decrease of skinfold thickness and the decrease of bone mass related to the age, the change of two tissues composed of collagen type I is caused by the same genetic mechanism. The decrease of bone mass is caused by the change of the amount and structure of bone matrix by several factors and the amount of minerals deposited on bone matrix. Especially, in case of female, the deficiency of estrogen by menopause makes these changes rapidly increased. The decrease of bone mass and skinfold thickness is due to the decrease of the amount of collagen and its structural change the common component of bone tissue and skin tissue. Therefore, the relationship of the amount of cross-linked peptide N-telopeptide, collagen metabolite which excretes as urine. Based upon the proved results about the significant relationship of bone mass, the amount of bone collagen, the amount of skin collagen and skinfold thickness, the bone mass may be expected through a facile determination of skinfold thickness.

• Carbon letters
• /
• v.5 no.1
• /
• pp.18-22
• /
• 2004

The failure behaviours of unidirectional pultruded carbon fiber reinforced polymer (CFRP) composites were monitored by the electrical resistance measurement during tensile loading, three-point-bending, interlaminar shear loading. The tensile failure behaviour of carbon fiber tows was also investigated by the electrical resistance measurement. Infrared thermography non-destructive evaluation was performed in real time during tensile test of CFRP composites to validate the change of microdamage in the materials. Experiment results demonstrated that the CFRP composites and carbon fiber tows were damaged by different damage mechinsms during tensile loading, for the CFRP composites, mainly being in the forms of matrix damage and the debonding between matrix and fibers, while for the carbon fiber tows, mainly being in the forms of fiber fracture. The correlation between the infrared thermographs and the change in the electrical resistance could be regarded as an evidence of the damage mechanisms of the CFRP composites. During three-point-bending loading, the main damage forms were the simultaneity fracture of matrix and fibers firstly, then matrix cracking and the debonding between matrix and fiber were carried out. This results can be shown in Fig. 9(a) and (b). During interlaminar shear loading, the change in the electrical resistance was related to the damage degree of interlaminar structure. Electrical resistance measurement was more sensitive to the damage behaviour of the CFRP composites than the stress/time curve.

• International Journal of Concrete Structures and Materials
• /
• v.18 no.2E
• /
• pp.97-102
• /
• 2006

In this study, the effect of supplementary materials(GGBS, PFA, SF) on sulfuric acid corrosion resistance was assessed by measuring the compressive strength, corroded depth and weight change at 7, 28, 56, 91, 180 and 250 days of immersion in sulfuric acid solution with the pH of 0.5, 1.0, 2.0 and 3.0. Then, it was found that an increase in the duration of immersion and a decrease in the pH, as expected, resulted in a more severe corrosion irrespective of binders: increased corroded depth and weight change, and lowered the compressive strength. 60% GGBS mortar specimen was the most resistant to acid corrosion in terms of the corroded depth, weight change and compressive strength, due to the latent hydraulic characteristics and lower portion of calcium hydroxide. The order of resistance to acid was 60% GGBS>20% PFA>10% SF>OPC. In a microscopic examination, it was found that acid corrosion of cement matrix produced gypsum, as a result of decomposition of hydration products, which may loose the structure of cement matrix, thereby leading to a remarkable decrease of concrete properties.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.14 no.2
• /
• pp.422-429
• /
• 1990

The experimental study investigates two aspects of counterflow sensible heat regenerator operation. First, it examines the regenerator performance in periodic steady state operation with spatially nonuniform inlet temperature in one of the fluid stream. Second, the study examines the transient response of a regenerator to a step change in the inlet temperature of one of the fluid streams. The effect of transient inlet temperatures is analyzed in terms of the response of the outlet fluid temperatures to a step change in temperature of one of the inlet fluid streams. The effect of temperature nonuniformities is analyzed in terms of the change of temperature nonuniformities is analyzed in terms of the change in steady state effectiveness due to a circumferential temperature distribution in one of the inlet fluid streams. an experimental analysis has been conducted using a counterflow, parallel passage, and rotary regenerator made from polyethylene film. Efficiencies follow similar trends with increasing matrix to fluid capacity rate ratio for the balanced and symmetric regenerator with nonuniform inlet temperature.