• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.1C
• /
• pp.65-71
• /
• 2004

We improved the MLLR speaker adaptation algorithm with reduction of the order of HMM parameters using PCA(Principle Component Analysis) or ICA(Independent Component Analysis). To find a smaller set of variables with less redundancy, we adapt PCA(principal component analysis) and ICA(independent component analysis) that would give as good a representation as possible, minimize the correlations between data elements, and remove the axis with less covariance or higher-order statistical independencies. Ordinary MLLR algorithm needs more than 30 seconds adaptation data to represent higher word recognition rate of SD(Speaker Dependent) models than of SI(Speaker Independent) models, whereas proposed algorithm needs just more than 10 seconds adaptation data. 10 components for ICA and PCA represent similar performance with 36 components for ordinary MLLR framework. So, compared with ordinary MLLR algorithm, the amount of total computation requested in speaker adaptation is reduced by about 1/167 in proposed MLLR algorithm.

• Journal of Hydrogen and New Energy
• /
• v.10 no.4
• /
• pp.191-196
• /
• 1999

4-probe resistivity measurement technique was used to study hydrogen A-D(Absorption-Desorption)kinetics on Pd films(18 to 67nm thick) from 25 to $50^{\circ}C$, from 0 to 5 torr hydrogen pressure. Pd films were made on sapphire substrate by thermal evaporation technique under high vacuum at room temperature. Upto about 100 hydrogen A-D cyclings, no pulverization was observed, but film was detached partially from substrate. Forward reaction and backward reaction rate were analyzed separately. The activation energies of hydrogen A-D processes were obtained from the Arrhenius plot of the reaction rates. The activation energies of Pd films are not strongly dependent on the thickness of the film. But the activation energy of very thin film( l8nm thick) was smaller than the others.

• Journal of Hydrogen and New Energy
• /
• v.29 no.2
• /
• pp.219-225
• /
• 2018

This study was carried out to investigate the injection characteristics of 800 kPa compressed natural gas compressed natural gas (CNG) injector developed in Korea. The CNG injector with multi-holes, employed in this experiment, was designed to inject CNG in the manifold at high pressure of 800 kPa. The spray macroscopic visualization test was carried out via Schlieren photography to study fuel-air mixing process. The fundamental spray characteristics, such as spray penetration, spray cone angle and spray velocity, were evaluated in the constant volume combustion chamber (CVCC) with varying the constant back pressure in CVCC from 0 to 1.8 bar. For the safety reason, nitrogen ($N_2$) and an acetone tracer were utilized as a surrogate gas fuel instead of CNG. The surrogate gas fuel pressures were controlled at 3, 5.5, and 8 bar, respectively. Injection durations were set at 5 ms throughout the experiment. The simulating events of the low engine speed were arranged at 1,000 rpm. The spray images were recorded by using a high-speed camera with a frame rate of 10,000 f/s at $512{\times}256pixels$. The spray characteristics were analyzed by using the image processing (Matlab). The results showed the significant difference that higher injection pressure had more effect on the spray shape than the lower injection pressure. When the injection pressure was increased, the longer spray penetration occurred. Moreover, the linear relation between speed and time are dependent on the injection pressure as well.

• Biotechnology and Bioprocess Engineering:BBE
• /
• v.6 no.5
• /
• pp.326-331
• /
• 2001

With the aim of developing of pH-sensitive controlled drug release system, a poly(Llysine) (PLL) based cationic semi-interpenetrating polymer network (semi-IPN) has been synthesized. This cationic hydrogel was designed to swell at lower pH and de-swell at higher pH and therefore be applicable for achieving regulated drug release at a specific pH range. In addition to the pH sensitivity, this hydrogel was anticipated to interact with an ionic drug, providing another means to regulate the release rate of ionic drugs. This semi-IPN hydrogel was prepared using a free-radical polymerization method and by crosslinking of the polyethylene glycol (PEG)-methacrylate polymer through the PLL network. The two polymers were penetrated with each other via interpolymer complexation to yield the semi-IPN structures. The PLL hydrogel thus prepared showed dynamic swelling/de-swelling behavior in response to pH change, and such a behavior was influenced by both the concentrations of PLL and PEG-methacrylate. Drug release from this semi-IPN hydrogel was also investigated using a model protein drug, streptokinase. Streptokinase release was found to be dependent on its ionic interaction with the PLL backbones as well as on the swelling of the semi-IPN hydrogel. These results suggest that a PLL semi-IPN hydrogel could potentially be used as a drug delivery platform to modulate drug release by pH-sensitivity and ionic interaction.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.42 no.2
• /
• pp.393-402
• /
• 2017

Just like terrestrial cellular networks, underwater cellular communication networks, which can manage the overall network resource by adaptively allocating backhaul resource for each base station according to its ingress traffic, are necessary. In this paper, a new resource allocation protocol is proposed for the underwater cellular communication network, allocating backhaul resource of a base station proportional to its ingress traffic to the base station. This protocol is classified into two types dependent upon allocation period: the resource allocation protocol with adaptive period and that with fixed period. In order to determine a proper resource allocation protocol, the performance of the two protocols, in terms of reception rate, message overhead, and latency is compared and investigated via simulation. As a result, the resource protocol with adaptive period outperforms that with fixed period; the resource allocation protocol with fixed period results in a maximum of $10^2$ order longer queueing delay as well as $10^2$ order greater message overhead than that with adaptive period.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2010.06a
• /
• pp.215-215
• /
• 2010

The feasibility of InSbTe (IST) chalcogenide materials prepared by metalorganic chemical vapor deposition (MOCVD) for phase-change memory (PRAM) applications was demonstrated. Films grown below $225^{\circ}C$ exhibited an amorphous structure, and the films grown at $300^{\circ}C$ Cincluded various crystalline phases such as In-Sb-Te, In-Sb, In-Te, and Sb-Te. The composition of the amorphous films grown at $225^{\circ}C$ was dependent on the working pressure. Films grown at $225^{\circ}C$ exhibited a smooth morphology with a root mean square(rms) roughness of less than 1nm, and the step-coverage of the films grown on a trench structure with an aspect ratio of 5:1 was greater than 90%. An increase in deposition time increased the filling rate, while retaining the conformal step-coverage. Films grown at $225^{\circ}C$ for 3h in a working pressure of $13{\times}10^2$ Pa exhibited a reproducible and complete filling in a trench structure.

• Journal of the Korea Convergence Society
• /
• v.8 no.5
• /
• pp.13-17
• /
• 2017

Existing Speech feature extracting method in speech Signal, there are incorrect recognition rates due to incorrect speech which is not clear threshold value. In this article, the modeling method for improving speech recognition performance that combines the feature extraction for speech and silence characteristics normalized to the non-speech. The proposed method is minimized the noise affect, and speech recognition model are convergence of speech signal feature extraction to each speech frame and the silence feature normalization. Also, this method create the original speech signal with energy spectrum similar to entropy, therefore speech noise effects are to receive less of the noise. the performance values are improved in signal to noise ration by the silence feature normalization. We fixed speech and non speech classification standard value in cepstrum For th Performance analysis of the method presented in this paper is showed by comparing the results with CHMM HMM, the recognition rate was improved 2.7%p in the speech dependent and advanced 0.7%p in the speech independent.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.47 no.2
• /
• pp.25-33
• /
• 2010

Introducing the quantized channel state information (CSI), space division multiple access (SDMA) can extract the multiplexing gain with the limited feedback burden. However, huge signaling burden of feedback can still suffer SDMA system because the total feedback data of SDMA is linearly dependent on the number of users. Hence, we propose a new feedback scheme to control the feedback load decided by the number of users. In this scheme, the cut-off level, which restricts the feedbacks of poor conditioned users, is suggested for the reduction of the feedback burden without the performance loss. From simulation results, then, we show that the proposed feedback scheme can achieve not only the sum-rate gain but also the reasonable feedback reduction.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.49 no.1
• /
• pp.23-28
• /
• 2012

The performance improvement of the secondary storage is very slow compared to the main memory and processor. The data is loaded from secondary storage to memory for the execution of an application. At this time, there is a bottleneck. In this paper, we propose an Executable Compression Method to speed up the initial loading time of application. and we examined the performance. So we implemented the two applications. The one is a compressor for Execution Binary File. and The other is a decoder of Executable Compressed application file on the Embedded System. Using the test binary files, we performed the speed test in the six files. At the result, one result showed that the performance was decreased. but others had a increased performance. the average increasing rate was almost 29% at the initial loading time. The level of compression had different characteristics of the file. And the performance level was dependent on the file compressed size and uncompress time. so the optimized compression algorithm will be needed to apply the execution binary file.

• Journal of Institute of Control, Robotics and Systems
• /
• v.18 no.5
• /
• pp.487-495
• /
• 2012

In this paper, a Smart Home Service Robot, McBot II, which performs mess-cleanup function etc. in house, is designed much more optimally than other service robots. It is newly developed in much more practical system than McBot I which we had developed two years ago. One characteristic attribute of mobile platforms equipped with a set of dependent wheels is their omni- directionality and the ability to realize complex translational and rotational trajectories for agile navigation in door. An accurate coordination of steering angle and spinning rate of each wheel is necessary for a consistent motion. This paper develops trajectory controller of 3-wheels omni-directional mobile robot using fuzzy azimuth estimator. A specialized anthropomorphic robot manipulator which can be attached to the housemaid robot McBot II, is developed in this paper. This built-in type manipulator consists of both arms with 4 DOF (Degree of Freedom) each and both hands with 3 DOF each. The robotic arm is optimally designed to satisfy both the minimum mechanical size and the maximum workspace. Minimum mass and length are required for the built-in cooperated-arms system. But that makes the workspace so small. This paper proposes optimal design method to overcome the problem by using neck joint to move the arms horizontally forward/backward and waist joint to move them vertically up/down. The robotic hand, which has two fingers and a thumb, is also optimally designed in task-based concept. Finally, the good performance of the developed McBot II is confirmed through live tests of the mess-cleanup task.