• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.10a
• /
• pp.369-372
• /
• 2018

This paper suggests non-parallel-voice-conversion network conversing voice between unmapped voice pair as source voice and target voice. Conventional voice conversion researches used learning methods that minimize spectrogram's distance error. Not only these researches have some problem that is lost spectrogram resolution by methods averaging pixels. But also have used parallel data that is hard to collect. This research uses PPGs that is input voice's phonetic data and a GAN learning method to generate more clear voices. To evaluate the suggested method, we conduct MOS test with GMM based Model. We found that the performance is improved compared to the conventional methods.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.38 no.4
• /
• pp.18-26
• /
• 2001

In this paper, we propose an ATC(Automatic Transfer mode Controller) algorithm and an average-mode-change method for use in Gear shift PLL which can automatically change loop gain. The proposed ATC algorithm accurately detects proper timing or the mode change and has a very simpler structure - than the conventional lock detector algorithm often used in QPSK. And the proposed average mode change method can obtain low errors of estimated frequency offset by averaging the loop filter output of frequency component in shift register. These algorithms are also useful in designing ASIC, since these algorithms occupy small circuit area and are adaptable for high speed digital processing. We also present phase tracking performance of proposed Gear-shift PLL, which is composed of polarity decision PD, ATC and average mode change circuit, and analyze the results by examining constellation at each mode.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.10C
• /
• pp.951-962
• /
• 2002

A DMT transceiver recovers the sampling time from reserved sub-carriers, the pilots. Since the pilots are available after the FFT, the symbol synchronization must be done before sample synchronization. In DMT VDSL system, symbol synchronization is handled separately from sample synchronization, although the two processes are intimately related. The DMT symbol itself contains sufficient information, the cyclic extension, for symbol synchronization. Using only the sign bit of received signal, the Maximum Likelihood Estimation solution is derived. The Tx windowing in the transmitter of DMT VDSL system results in the blurring of MLE peaks. We propose the weighted summing MLE method using the sign bit which produces the clearly sharp top of MLE peaks. The stability of symbol synchronization is improved significantly by averaging over a few symbols. This paper presents the study of the original MLE and the weighted summing MLE using sign bit. A clock difference between transmitter and receiver destroys the oahogonality of the carriers. Therefore, a receiver using asynchronous sampling must perform timing correction in the discrete-time domain. We introduce an efficient digital sample synchronization method which is based on temporal and frequency domain digital signal processing.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.16 no.1
• /
• pp.13-26
• /
• 1988

^x This study was executed to analyze the vegetational landscape structure of Donggu Royal Tomb forest by four kinds of ordination techniques(polar, principal component analysis, reciprocal averaging, and detrended correspondence analysis) and comparison of the couple photographs between 1920s and 1987. Thirteen sites were sampled with clumped sampling method in June of 1987 and five quadrats were examined in each site. The result of this study can be summarized as follows. 1. Environmental impact grade 3,4 and 5 area covered 16.11ha and these area should be restored by the input of human energy because self - refair seemed to be impossible. 2. Pinus densiflora community of actual vegetation covered 8.2%, Alnus japonica community 5.7% and quercus aliena community 15.8% of the surveyed area(73.3ha). And afforested vegetation including Pinus koraiensis, Pinus rigida, Alnus hirsuta and Robinia pseudoacacia covered 27.0%. 3. Donggu Royal Tomb forest was divided by Pinus densiflora, P. densiflora-Quercus aliena, Alnus japonia - Q. aliena and A. japonica community. DCA was the most effective method of this study. 4. DCA ordination were showed that succesional trends of tree species seem to be from Pinus densiflora through Prunus sargentii, Quercus serrata, Quercus aliena to Carpinus laxiflora in P. densiflora community and from Alnus japonica through Acer ginnala to Q. aliena in A. japonica community of the upper layer and from Rosa multiflora, Symplocos chinensis for. pilosa, Zanthoxylum schiniflolium through Rhus trichocarpa, Styrax japonica, Rhododendron scilippenbachii to Viburum erosum, Lindera obtusiloba in the middle layer. 5. By the comparsion of the couple photographs beween 1920s and 1987, we can recognize the change of historical landscape composed by P. densiflora and A. japonica community and those community is succeed to Q. aliena.

• Journal of KIISE:Computing Practices and Letters
• /
• v.5 no.5
• /
• pp.628-638
• /
• 1999

In this paper, we presents a method of retrieving 24 bpp RGB images based on color-spatial features. For each image, it is subdivided into regions by using similarity of color after converting RGB color space to CIE L*u*v* color space that is perceptually uniform. Our segmentation algorithm constrains the size of region because a small region is discardable and a large region is difficult to extract spatial feature. For each region, averaging color and center of region are extracted to construct color-spatial features. During the image retrieval process, the color and spatial features of query are compared with those of the database images using our similarity measure to determine the set of candidate images to be retrieved. We implement a content-based color image retrieval system using the proposed method. The system is able to retrieve images by user graphic or example image query. Experimental results show that Recall/Precision is 0.80/0.84.

• Water for future
• /
• v.8 no.1
• /
• pp.61-79
• /
• 1975

An attempt is made to develope a scheme for synthesizing unit hydrograph for any arbitrary small watershed in the Han or Geum River basin, which can be applied in determining various sizes of design flood for flood control prijects. Stage gauging stations, seven in the Han and five in the Geun River basin with rating curves, were selected as subbasins for the analysis. Unit hydrographs of 2-hour duration were derived for several heavy storm events using the storm and the corresponding flood runoff data for each subbasin. The Clark method programmed by the Hydrologic Engineering Center, U.S. Corps of Engineers, was utilized for derivation of instantaneous unit hydrographs which were, in turn, converted into 2-hour unit hydrograph. By averaging the 2-hour unit hydrographs from several storm events a representative 2-hour unit hydrograph was determined for each subbasin and hence a separate derivation of dimensionless unit hedrograph was also possible for the Han and Geum River basins. The physiographic characteristics such as stream length, distance to the centroid of each watershed were correlated with the characteristic parameters of the derived unit hydrograph for the subbasins within two large basins. correlation analyses between the characteristic parameters were also made. These correlation analyses resulted a series of four equations and a dimensionless unit hydrograph for the two large basins, which made it possible to draw a synthetic 2-hour unitgraph for any small watershed within the Han or Geum River basin. A detailed procedure for aplying the derived method for an arbitrary basin is summarized with one sample computation for each of the two basins. A comparison of the actually derived 2-hour unit hydrogrograph and the synthesized one showed a fair agreement. A recommendation is made for the further study.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.12 no.5
• /
• pp.959-967
• /
• 1988

In this paper, 2-DOF vehicle suspension system with a semiactive on-off damper was studied for improving the ride comfort. It is known that a nonlinear hydraulic damper, which generates force proportional to the square of the relative velocity, can describe the actual fluid resisting type damper more properly than the traditional viscous damping model. On the other hand, hydraulic damper adoption in analysis makes the system nonlinear and causes difficulties to get the system response. In this work, time domain direct integration method was used to calculate system displacement and acceleration. first of all, the response of the suspension system experiencing a given road profile was optimized by Lagrangian multiplier method within the range of given damping coefficients. The appropriate on-loaf damping values were obtained by averaging the already calculated optimum damping coefficients from Lagrangian techniques. The criterion to control the on-off mechanism was determined by examining the suspension efficiency. It was found that the best out of practically applicable criteria is following the sign (positive and negative) of the multiplication of relative displacement and velocity. Judging from the theoretical calculations, it was proved that the semiactive on-off damper can increase suspension efficiency as much as 8-11% in object function.

• The KIPS Transactions:PartB
• /
• v.17B no.3
• /
• pp.233-238
• /
• 2010

In this paper, we propose a multiple classifier combination method based on image degradation modeling to improve recognition performance on low-quality images. Using an image degradation model, it generates a set of classifiers each of which is specialized for a specific image quality. In recognition, it combines the results of the recognizers by weighted averaging to decide the final result. At this time, the weight of each recognizer is dynamically decided from the estimated quality of the input image. It assigns large weight to the recognizer specialized to the estimated quality of the input image, but small weight to other recognizers. As the result, it can effectively adapt to image quality variation. Moreover, being a multiple-classifier system, it shows more reliable performance then the single-classifier system on low-quality images. In the experiment, the proposed multiple-classifier combination method achieved higher recognition rate than multiple-classifier combination systems not considering the image quality or single classifier systems considering the image quality.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.3
• /
• pp.388-393
• /
• 2010

Cultural Content Technology(CT, Culture Technology) for the development of cultural industry and the commercialization of technology, cultural contents, media, mount, pass the value chain process and increase the added value of cultural products that are good for all forms of intangible technology. In the field of Culture Technology, Music by analyzing the characteristics of the development of a variety of applications has been studied. Associated with EEG measures and the results of their research in response to musical stimuli are used to detect and study is getting attention. In this paper, the musical stimuli in EEG signals by amplifying the corresponding reaction to the averaging method, ERP (Event-Related Potentials) experiments based on the process of extracting sound methods for removing noise from the ICA algorithm to extract the tone and noise removal according to the results are applied to analyze the characteristics of EEG.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.4
• /
• pp.840-845
• /
• 2007

In a wireless sensor network, a middleware used to support efficient processing and fast delivering of sensing data should handle the data loss problem at an intermediate sensor node caused by unexpected sudden data burst. In his paper, it proposes a Delta-Average method for enhancing the efficiency of data aggregation and correctness where the sensed data should be delivered only with the limited computing power and energy resource. With the proposed method, unnecessary data transfer of the duplicate data is eliminated and data correctness is enhanced by using the proposed averaging and data differentiating scheme when an instantaneous data burst is occurred. Finally, with the TOSSTM simulation results on TinyDB, we verify that the correctness of the transferred data is enhanced.