• Proceedings of the IEEK Conference
• /
• 2002.06d
• /
• pp.195-198
• /
• 2002

We propose an efficient mixture Gaussian synthesis method for decision tree based state tying that produces better context-dependent models in a short period of training time. This method makes it possible to handle mixture Gaussian HMMs in decision tree based state tying algorithm, and provides higher recognition performance compared to the conventional HMM training procedure using decision tree based state tying on single Gaussian GMMs. This method also reduces the steps of HMM training procedure. We applied this method to training of PBS, and we expect to achieve a little point improvement in phoneme accuarcy and reduction in training time.

• Structural Engineering and Mechanics
• /
• v.29 no.2
• /
• pp.203-221
• /
• 2008

In this paper, we present an idea of the geometry-dependent MITC method. The simple concept is exemplified to improve a 2-node iso-beam (isoparametric beam) finite element of varying section. We first study the behavior of a standard 2-node iso-beam finite element of prismatic section, which has been widely used with reduced integration (or the equivalent MITC method) in order to avoid shear locking. Based on analytical studies on cantilever beams of varying section, we propose the axial strain correction (ASC) scheme and the geometry-dependent tying (GDT) scheme for the 2-node iso-beam element. We numerically analyze varying section beam problems and present the improved performance by using both ASC and GDT schemes.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.6
• /
• pp.167-176
• /
• 2005

This paper describes how the state tying model based on the decision tree which is one of Acoustic models used for speech recognition optimizes the model by reducing the number of mixture Gaussians of the output probability distribution. The state tying modeling uses a finite set of questions which is possible to include the phonological knowledge and the likelihood based decision criteria. And the recognition rate can be improved by increasing the number of mixture Gaussians of the output probability distribution. In this paper, we'll reduce the number of mixture Gaussians at the highest point of recognition rate by clustering the Gaussians. Bhattacharyya and Euclidean method will be used for the distance measure needed when clustering. And after calculating the mean and variance between the pair of lowest distance, the new Gaussians are created. The parameters for the new Gaussians are derived from the parameters of the Gaussians from which it is born. Experiments have been performed using the STOCKNAME (1,680) databases. And the test results show that the proposed method using Bhattacharyya distance measure maintains their recognition rate at $97.2\%$ and reduces the ratio of the number of mixture Gaussians by $1.0\%$. And the method using Euclidean distance measure shows that it maintains the recognition rate at $96.9\%$ and reduces the ratio of the number of mixture Gaussians by $1.0\%$. Then the methods can optimize the state tying model.

• Proceedings of the IEEK Conference
• /
• 2002.07a
• /
• pp.38-41
• /
• 2002

Many scientifc applications include manipulation of data points tying in a space. We describe a method, based on sector labeling to generate a space-filling curve for partitioning such given data points. Our method is straightforward and flexible, equipping a one-one correspondence between point-values on the curve and data points in space in more efficient than designated methods found in the literature. It is widely believed that the Hilbert curve achieves the desired properties on linear mappings due to the locality between data points. Therefore we focus on the Hilbert curve since, later on, we identify it as the most suitable for our application. We demonstrate on using our method for the data particles of an n-body simulation that based on Barnes-Hut algorithm.

• Proceedings of the KSME Conference
• /
• 2000.04a
• /
• pp.176-181
• /
• 2000

Stress intensity factors for a rigid line inclusion tying along a bimaterial interface are calculated by the boundary element method with the multiregion and double-Point techniques. The formula between the stress intensity factors and the inclusion surface stresses are derived. The numerical values of the stress intensity factors for the bimaterial interface rigid line inclusion in the infinite body are proved to be in good agreement within 3% when compared with the previous exact solutions. In the finite bimaterial systems, the stress intensity factors for the center and edge rigid line inclusions at interface are computed with the variation of the rigid line inclusion length and the shear modulus ratio under the biaxial and uniaxial loading conditions.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.42 no.11
• /
• pp.911-918
• /
• 2014

In this paper, a selective activation strategy of cohesive elements using user subroutine UMPC was studied as an efficient solution for the added compliance problem in cohesive zone model crack propagation analyses. The cohesive elements were inserted between every bulk elements in region where cracks were expected to initiate and propagate, but initially not activated by tying the cohesive nodes using multi-point constraints. During analyses, the cohesive elements for which specified criterion was met were selectively activated by releasing the constraints. The effect of initial cohesive stiffness and the release criterion on the crack propagation behavior was carefully investigated.

• Journal of the Korean Geotechnical Society
• /
• v.15 no.6
• /
• pp.297-306
• /
• 1999

In this research, the frost heave mechanism of the weathered tuff soil sampled from the area tying between Ulanbator and Beijing was studied. The frost heave tests were carried maintaining the constant temperature at both upper$(+5^{\circ}C)\; and\; lower(-5^{\circ}C)$ ends of the sample. Here, main emphasis is given on variation of the freezing point depression with the variation of applied overburden pressure. The expansion of ice lens and migration of the pore water towards freezing front were observed in the test. It was found that with the increase in overburden pressure there is decrease in heave rate and increase in the absolute value of a segregation-freezing temperature. Hence the equation between segregation-freezing temperature and overburden pressure could be suggested. Also the water content of the samples at the frozen side was shown to be higher than those at the unfrozen side. Similarly, the water at warmer part of the frozen tuff was found much higher than that of the cooler part.

• Journal of Biosystems Engineering
• /
• v.29 no.6 s.107
• /
• pp.515-520
• /
• 2004

A device far estimating the optimal artificial insemination time consisted of a computer, RS module, and six direct-reflex photo-sensors. Program was written in LabWindows CVI. In order to establish references for estimating optimal artificial insemination time, lying rate of the thirty three $Berkshire{\times}Hampshire crossbred sows (parity 5 to 6) was recorded from 2 days after moving into stalls to artificial insemination using a time lapse VCR and was measured every hour with one minute interval using the scan point sampling method. Twenty low hours in a day were divided into three comparing periods : 21:00 to 06:59, 07:00 to 13:59, and 14:00 to 20:59. If sum of the percentages of tying during a comparing period was less than the reference value, the starting hour of the comparing period was considered as onset of estrus and the optimal artificial insemination time was estimated at 20 to 24 hours after onset of estrus. The experimental device was evaluated with twenty five $Berkshire{\times}Hampshire$ crossbred sows (parity 2 to 7) and 23 of 25 sows $(92\%)$ were pregnant. This result suggests that the experimental device is excellent in performance for estimating optimal artificial insemination time.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38A no.12
• /
• pp.1135-1144
• /
• 2013

Deploying small cells is a reliable and influential solution to handle the skyrocketing traffic increase in the cellular network, and the small cell technology is evolving to ultra-dense network (UDN). In this paper we propose a small cell on/off algorithm with a simple but essential framework composed of access point (AP), user equipment (UE), and small cell controller (SCC). We propose Device-Assisted Networking for Cellular grEening (DANCE) algorithms that save the energy consumption by tying to minimize the number of turned-on APs while maintaining the network throughput. In doing so, SCC firstly gathers the feedback messages from UEs and then makes a decision including a set of turned-on APs and user association. DANCE algorithm has several variations depending on the number of bits of the UE's feedback message (1 bit vs. N bit), and is divided into AP-first, UE-first, or Proximity ON according to the criteria of selecting the turned-on APs. We perform extensive simulations under the realistic UDN environment, and the results confirm that the proposed algorithms, compared to the baseline, can significantly enhance the energy efficiency, e.g., more than a factor of 10.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.50-50
• /
• 2000

In this paper, a new measuring system is :proposed which can measure the fine 6-DOF displacement of rigid bodies. Its measurement principle is based on detection of laser beam reflected from a specially fabricated mirror that looks like a triangular pyramid having an equilateral cross-sectional shape. The mirror has three lateral reflective surfaces inclined 45$^{\circ}$ to its bottom surface. We call this mirror 3-facet mirror. The 3-facet mirror is mounted on the object whose 6-DOF displacement is to be measured. The measurement is operated by a laser-based optical system composed of a 3-facet mirror, a laser source, three position-sensitive detectors(PSD). In the sensor system, three PSDs are located at three corner points of a triangular formation, which is an equilateral triangular formation tying parallel to the reference plane. The sensitive areas of three PSDs are oriented toward the center point of the triangular formation. The object whose 6-DOF displacement is to be measured is situated at the center with the 3-facet mirror on its top surface. A laser beam is emitted from the laser source located at the upright position and vertically incident on the top of the 3-fatcet mirror. Since each reflective facet faces toward each PSD, the laser beam is reflected at the 3-facet mirror and splits into three sub-beams, each of which is reflected from the three facets and finally arrives at three PSDs, respectively. Since each PSD is a 2-dimensional sensor, we can acquire the information on the 6-DOF displacement of the 3-facet mirror. From this principle, we can get 6-DOF displacement of any object simply by mounting the 3-facet mirror on the object. In this paper, we model the relationship between the 6-DOF displacement of the object and the outputs of three PSDs. And, a series of simulations are performed to demonstrate the effectiveness of the proposed method. The simulation results show that the proposed sensing system can be an effective means of obtaining 3-dimensional position and orientation of arbitrary objects.