• The Journal of the Acoustical Society of Korea
• /
• v.14 no.3
• /
• pp.49-60
• /
• 1995

This paper is a study on the design and implementation of the Korean Text-to-Speech system which is used for a small and simple system. In this paper, a parameter synthesis method is chosen for speech syntheiss method, we use PARCOR(PARtial autoCORrelation) coefficient which is one of the LPC analysis. And we use phoneme for synthesis unit which is the basic unit for speech synthesis. We use PARCOR, pitch, amplitude as synthesis parameter of voice, we use residual signal, PARCOR coefficients as synthesis parameter of unvoice. In this paper, we could obtain the 60% intelligibility by using the residual signal as excitation signal of unvoiced sound. The result of synthesis experiment, synthesis of a word unit is available. The controlling of phoneme duration is necessary for synthesizing of a sentence unit. For setting up the synthesis system, PC 486, a 70[Hz]-4.5[KHz] band pass filter for speech input/output, amplifier, and TMS320C30 DSP board was used.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.24 no.2
• /
• pp.95-101
• /
• 2011

In wireless X networks where all transmitters send the independent messages to all receivers, the theoretical bound on the degrees of freedom (DOF) and interference alignment (IA) scheme for its achievability are given by Cadambe and Jafar [1]. However, IA scheme for wireless X network may be infeasible in practice unless the transmitters have the perfect channel information. In addition, if the transmitter with single antenna uses time-varying channel coefficients as a beamforming vector, the infinite channel extension is required to achieve the theoretical bound on the DOF of wireless X networks with perfect IA scheme. In this paper, we consider K-user MIMO X network where K transmitters and K receivers have M antennas each. While the beamforming vectors have been constructed with multiple channel uses over multiple time slot in the earlier work, we consider the beamforming vectors constructed only by a spatial signature over unit time. Accordingly the channel information at the transmitters can be available instantaneously. Then we propose the perfect IA scheme with no channel extension. Based on our sum-rate analysis and the simulation results, we confirm that our proposed scheme can achieve MK/2 DOF which is quite close to the theoretical bound on the DOF region of wireless X networks.

• Journal of Advanced Navigation Technology
• /
• v.22 no.1
• /
• pp.27-30
• /
• 2018

In this study, a 40 GHz band pass filter(BPF) employing a hair-pin structure has been designed, fabricated, and characterized for millimeter-wave wireless communication applications. Using the 3 dimensional(3-D) electromagnetic(EM) tool and design equations of the hairpin BPF, the BPF was desgned on the 5 mil-thick Duroid substrate(RT5880) with a relative dielectric constant (${\varepsilon}_r$) of 2.2. The tapping point (t) of the U-shape resonator in the input and output port has been determined using extracted an external Q-factor ($Q_e$). The coupling coefficients between the other resonators are calculated by adjusting the physical dimensions for the desired response of the BPF. The fabricated BPF was characterized using probing method on a probe station. Its measured center frequency(fc) and fractional BW are 41.6 GHz and 7.43 %, respectively. The measured return loss is below -10 dB at the pass band and the insertion loss is 3.87 dB. The fabricated BPF is as small as $9.1{\times}2.8mm^2$.

• Journal of Korean Society on Water Environment
• /
• v.23 no.5
• /
• pp.607-612
• /
• 2007

In this study, we investigated the variations of the kinetic coefficients and Chemical Oxygen Demand (COD), N and P mass used for assimilation of a sequencing batch reactor (SBR) system with the variation of SRTs; SRTs of 7.5, 10.0, 12.5, 15.0 and 20.0 days were tested in one cycle of SBR operation to determine the optimum conditions for the operation of the SBR and estimate its COD, nitrogen and phosphorus removal efficiencies. The SBR system was operated under the conditions as follows: an operation time of 6 hours per cycle, a hydraulic retention time (HRT) of 12 hours, an influent COD loading of $0.4kg/m^3/day$, and an influent nitrogen loading of $0.068kgT-N/m^3/day$. The yield coefficient (Y) and decay rate coefficient ($k_d$) were estimated to be 0.4198 kgMLVSS/kgCOD and $0.0107day^{-1}$ by calculating the removal rate of substrate according to the variation of SRT. Considering total nitrogen amount removed by sludge waste process, eliminated by denitrification, and in clarified water effluent with reference to 150 mg/cycle of influent nitrogen amount, the percentage of nitrogen mass balance from the ratio of the nitrogen amount in effluent (N output) to that in influent (N input) for Runs 1~5 were 95.5, 97.0, 95.5, 99.5, and 95.5%, respectively, which is well accounted for, with mass balances close to 100%.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.5 no.4
• /
• pp.41-55
• /
• 1995

In this paper, an optimal idenfication method using fuzzy-neural networks is proposed for modeling of nonlinear complex systems. The proposed fuzzy-neural modeling implements system structure and parameter identification using the intelligent schemes together wlth optimization theory, linguistic fuzzy implication rules, and neural networks(NNs) from input and output data of processes. Inference type for this fuzzy-neural modeling is presented as simplified inference. To obtain optimal model, the learning rates and momentum coefficients of fuzzy-neural networks(FNNs) are tuned automatically using improved modified complex method and modified learning algorithm. For the purpose of its application to nonlinear processes, data for route choice of traffic problems and those for activateti sluge process of sewage treatment system are used for the purpose of evaluating the performance of the proposed fuzzy-neural network modeling. The results show that the proposed method can produce the intelligence model with higher accuracy than other works achieved previously.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.16 no.10 s.101
• /
• pp.989-994
• /
• 2005

A MMIC(Monolithic Microwave Integrated Circuit) switch chip using InGaAs/GaAs p-HEMT process has been designed for MSM(Microwave Switch Matrix) of satellite communication system. An absorptive type MMIC switch is adopted for good reflection coefficients performances of input and output ports at both on and off states. And, a quarter wavelength impedance transformer is realized with lumped elements of MIM capacitor and spiral inductor for 3 GHz band to reduce the chip size. This MMIC switch covers the frequency range of $3.2\~3.6\;GHz$. According to the on-wafer measurement, the fabricated MMIC switch with miniature size of $1.6\;mm{\times}1.3\;mm$ demonstrates insertion loss below 2 dB and isolation above 56.8 dB, and the performance coincides with simulation results.

• Computers and Concrete
• /
• v.19 no.3
• /
• pp.275-282
• /
• 2017

The aim of this study is to build Machine Learning models to evaluate the effect of blast furnace slag (BFS) and waste tire rubber powder (WTRP) on the compressive strength of cement mortars. In order to develop these models, 12 different mixes with 288 specimens of the 2, 7, 28, and 90 days compressive strength experimental results of cement mortars containing BFS, WTRP and BFS+WTRP were used in training and testing by Random Forest, Ada Boost, SVM and Bayes classifier machine learning models, which implement standard cement tests. The machine learning models were trained with 288 data that acquired from experimental results. The models had four input parameters that cover the amount of Portland cement, BFS, WTRP and sample ages. Furthermore, it had one output parameter which is compressive strength of cement mortars. Experimental observations from compressive strength tests were compared with predictions of machine learning methods. In order to do predictive experimentation, we exploit R programming language and corresponding packages. During experimentation on the dataset, Random Forest, Ada Boost and SVM models have produced notable good outputs with higher coefficients of determination of R2, RMS and MAPE. Among the machine learning algorithms, Ada Boost presented the best R2, RMS and MAPE values, which are 0.9831, 5.2425 and 0.1105, respectively. As a result, in the model, the testing results indicated that experimental data can be estimated to a notable close extent by the model.

• The Journal of the Acoustical Society of Korea
• /
• v.16 no.5
• /
• pp.43-49
• /
• 1997

Generally, echo cancelers of a TV conference system or a audio conference system are to carry out a real time processing in the case of the closed room having long reverberation time because the system requires much time to modify filter coefficients to environmental changes. Therefore this paper proposes a new subband adaptive filtering method using polyphase filter banks of MPEG(Moving Picture Experts Group) audio system to solve the problems. This method divides signal spectra of input and output into several frequency bands, and each band is adaptively filtered by using ES-NLMS (Exponential Step-Normalized Least Mean Square) algorithm. The optimal number of subband is determined by computational simulations. According to the results of simulation, ERLE of the subband model is 2dB smaller than general full band, calculation rate's of the subband model is decreased about 88%.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.44 no.1
• /
• pp.40-48
• /
• 2007

This paper presents a novel texture segmentation method using multilayer perceptron (MLP) networks and Markov random fields in multiscale Bayesian framework. Multiscale wavelet coefficients are used as input for the neural networks. The output of the neural network is modeled as a posterior probability. Texture classification at each scale is performed by the posterior probabilities from MLP networks and MAP (maximum a posterior) classification. Then, in order to obtain the more improved segmentation result at the finest scale, our proposed method fuses the multiscale MAP classifications sequentially from coarse to fine scales. This process is done by computing the MAP classification given the classification at one scale and a priori knowledge regarding contextual information which is extracted from the adjacent coarser scale classification. In this fusion process, the MRF (Markov random field) prior distribution and Gibbs sampler are used, where the MRF model serves as the smoothness constraint and the Gibbs sampler acts as the MAP classifier. The proposed segmentation method shows better performance than texture segmentation using the HMT (Hidden Markov trees) model and HMTseg.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.18 no.8
• /
• pp.895-903
• /
• 2007

Design equations of DC blocks terminated in arbitrary impedances are newly suggested and a microstrip DC block is tested for the perfect matching. The DC block is a two-port passive component and the power excited at a port is transmitted into another port. However, all the excited power at the input can not be delivered to the output and therefore most of the conventional DC blocks can not be perfectly matched with arbitrary termination impedances. To solve the matching problem, its one-port equivalent resonant circuit model, front which design equations can be derived, is newly suggested. Using the derived design equations, any DC block can be designed, perfectly matched without any restriction of coupling coefficients. To verify the derived design equations, measurements were carried out and the results are in good agreement with prediction, showing insertion and return losses at 4.1 GHz are 0.82dB and -31dB, respectively.