• Journal of information and communication convergence engineering
• /
• v.8 no.6
• /
• pp.609-617
• /
• 2010

This paper is concerned with the use of a selforganizing map (SOM) to estimate the desired channel states of an unknown digital communication channel for blind equalization. The modification of SOM is accomplished by using the Bayesian likelihood fitness function and the relation between the desired channel states and channel output states. At the end of each clustering epoch, a set of estimated clusters for an unknown channel is chosen as a set of pre-defined desired channel states, and used to extract the channel output states. Next, all of the possible desired channel states are constructed by considering the combinations of extracted channel output states, and a set of the desired states characterized by the maximal value of the Bayesian fitness is subsequently selected for the next SOM clustering epoch. This modification of SOM makes it possible to search the optimal desired channel states of an unknown channel. In simulations, binary signals are generated at random with Gaussian noise, and both linear and nonlinear channels are evaluated. The performance of the proposed method is compared with those of the "conventional" SOM and an existing hybrid genetic algorithm. Relatively high accuracy and fast search speed have been achieved by using the proposed method.

• The Bulletin of The Korean Astronomical Society
• /
• v.36 no.2
• /
• pp.100.1-100.1
• /
• 2011

Everybody wants to see the future, but nobody does for sure. Reliably forecasting the solar activity in the near future looks like an easy task, but in fact still remains one of difficult problems in the solar-terrestrial research. We have sought for good univariate methods that can predict future smoothed sunspot numbers reasonably well based on past smoothed sunspot number data only. Here we consider a specific method we call principal component-and-autoregression (PCAR) method. The variation of sunspot number during a period of finite duration (past) before an epoch (present) is modeled by a linear combination of a small number of dominant principal components, and this model is extended to the period (future) beyond the epoch using the autoregressive model of finite order. From the application of this method, we find that the $24^{th}$ solar maximum is likely to occur near the end of the year 2013 (and there is a possibility that it occurs earlier near the start of 2013), and to have a peak sunspot number of about 86, indicating that the activity of the $24^{th}$ cycle will be weaker than the average. We will discuss how much this estimate is reliable.

• Bulletin of the Korean Space Science Society
• /
• 2009.10a
• /
• pp.25.2-25.2
• /
• 2009

한국천문연구원 한국우주전파관측망(Korean VLBI Network, KVN)에서 도입하여 시험운용중인 VLBI 상관서브시스템(VLBI Correlation Subsystem, VCS)은 한일공동 VLBI 상관기(Korea-Japan Joint VLBI Correlator, KJJVC)의 핵심 장비로서, 최대 16 관측국의 관측국 당 최대 8Gbps의 데이터를 처리할 수 있는 상관처리장치이다. VCS의 상관처리 결과는 총 4회선의 10GbE 광케이블을 통하여 UDP 프로토콜로 출력된다. 이 상관처리 결과는 광케이블 하나당 8개씩 총 32개의 상관 처리 블록(correlation block)으로 구성되며, 최대 출력속도는 1.4 GBytes/sec이다. 이 출력은 초고속 대용량 자료저장 시스템(Peta-scale Epoch Data Archive, PEDA)을 이용하여 저장하고 후속 자료처리를 위해 가공된다. PEDA는 총 4대의 고성능 자료 전송 및 저장 서버(writing server) 및 대용량 하드디스크 어레이로 구성된다. 상관처리 과정에 맞추어 PEDA의 writing 서버를 연계하여 제어하는 자료 전송 및 저장 제어 소프트웨어를 개발하였다. 이 소프트웨어는 핵심이 되는 전송 및 저장 프로세서와 이를 제어하는 제어프로세서로 구성된다. 전송 및 저장 프로세서는 개개의 상관 처리 블록에 대한 전송과 저장을 전담한다. 제어 프로세서는 총 32개의 상관 처리 블록을 처리하기 위하여 전송 및 저장 프로세스를 32개를 실행하고 각각의 상관 처리 블록에 해당하는 개별파라미터를 전달하는 전체적인 제어를 담당한다. 이 연구에서는 이 자료전송 및 저장 제어 소프트웨어의 설계 구성과 테스트 내용을 소개한다.

• Journal of The Korean Astronomical Society
• /
• v.36 no.3
• /
• pp.189-212
• /
• 2003

I review the current status of understanding when, how long, and how giant elliptical galaxies formed, focusing on the globular clusters. Several observational evidences show that massive elliptical galaxies formed at z > 2 (> 10 Gyr ago). Giant elliptical galaxies show mostly a bimodal color distribution of globular clusters, indicating a factor of $\approx$ 20 metallicity difference between the two peaks. The red globular clusters (RGCs) are closely related with the stellar halo in color and spatial distribution, while the blue globular clusters (BGCs) are not. The ratio of the number of the RGCs and that of the BGCs varies depending on galaxies. It is concluded that the BGCs might have formed 12-13 Gyr ago, while the RGCs and giant elliptical galaxies might have formed similarly 10-11 Gyr ago. It remains now to explain the existence of a gap between the RGC formation epoch and the BGC formation epoch, and the rapid metallicity increase during the gap (${\Delta}t{\approx}$ 2 Gyr). If hierarchical merging can form a significant number of giant elliptical galaxies > 10 Gyr ago, several observational constraints from stars and globular clusters in elliptical galaxies can be explained.

• Publications of The Korean Astronomical Society
• /
• v.30 no.2
• /
• pp.125-127
• /
• 2015

We present preliminary results from OH ground state phase referenced observations carried out with the Multi Element Radio Linked Interferometer Network (MERLIN) and e-MERLIN towards the massive star forming region W49A. There are three active SFRs within this complex: W49 North (W49 N), W49 South (W49 S) and W49 South West (W49 SW). The first epoch of observations was obtained in 2005 with MERLIN while the second epoch was obtained in 2013 with the e-MERLIN upgraded system. In this paper, we present 1665 and 1720 MHz maser emission towards W49 S and W49 SW. Overall, both epochs show good agreement with the previous observations of Argon et al. (2000) carried out with the Very Large Array (VLA). The better sensitivity and wider velocity coverage of the MERLIN/e-MERLIN observations allowed us to discover a new 1720 MHz OH maser site in W49 S.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.33 no.6
• /
• pp.238-245
• /
• 2021

In this study, we carried out case study to predict dissolved oxygen (DO) concentration of Nakdong river estuary with LSTM model. we aimed to figure out a optimal model condition and appropriate predictor for prediction in dissolved oxygen concentration with model parameter and predictor as cases. Model parameter case study results showed that Epoch = 300 and Sequence length = 1 showed higher accuracy than other conditions. In predictor case study, it was highest accuracy where DO and Temperature were used as a predictor, it was caused by high correlation between DO concentration and Temperature. From above results, we figured out an appropriate model condition and predictor for prediction in DO concentration of Nakdong river estuary.

• Nuclear Engineering and Technology
• /
• v.54 no.8
• /
• pp.3027-3033
• /
• 2022

Background: In this study, various types of deep-learning models for predicting in vitro radiosensitivity from gene-expression profiling were compared. Methods: The clonogenic surviving fractions at 2 Gy from previous publications and microarray gene-expression data from the National Cancer Institute-60 cell lines were used to measure the radiosensitivity. Seven different prediction models including three distinct multi-layered perceptrons (MLP), four different convolutional neural networks (CNN) were compared. Folded cross-validation was applied to train and evaluate model performance. The criteria for correct prediction were absolute error < 0.02 or relative error < 10%. The models were compared in terms of prediction accuracy, training time per epoch, training fluctuations, and required calculation resources. Results: The strength of MLP-based models was their fast initial convergence and short training time per epoch. They represented significantly different prediction accuracy depending on the model configuration. The CNN-based models showed relatively high prediction accuracy, low training fluctuations, and a relatively small increase in the memory requirement as the model deepens. Conclusion: Our findings suggest that a CNN-based model with moderate depth would be appropriate when the prediction accuracy is important, and a shallow MLP-based model can be recommended when either the training resources or time are limited.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.2
• /
• pp.71.2-71.2
• /
• 2021

Using multi-epoch spectra of active galactic nuclei (AGN) obtained from the Sloan Digital Sky Survey, we identify 16 spectrally variable sources, for which the spectral shapes of broad emission lines significantly vary with a time scale of yrs. Out of them, 3 AGNs are already known as changing-look (CL) AGNs by previous studies. 6 AGNs are newly identified as CL AGNs from our study. A majority of these AGNs are relatively faint and their variability in the continuum is small, which may explain their non-detection in the previous studies. 7 sources are known as binary AGN candidates based on the systematic velocity offset between broad emission lines and narrow emission lines. For those sources and 3 CL AGNs, we find that the peak of broad emission lines had been shifted up to a few thousands km/s for ~10 years, implying that those can be promising candidates for pc-scale binary AGNs or recoiling black holes. We plan to conduct multiwavelength follow-up studies to nail down the physical origin of the velocity shift.

• Smart Media Journal
• /
• v.13 no.3
• /
• pp.27-35
• /
• 2024

This study deals with a method of combining image generation using Semi Supervised Learning based on GAN (Generative Adversarial Network) and image classification using ResNet50. Through this, a new approach was proposed to obtain more accurate and diverse results by integrating image generation and classification. The generator and discriminator are trained to distinguish generated images from actual images, and image classification is performed using ResNet50. In the experimental results, it was confirmed that the quality of the generated images changes depending on the epoch, and through this, we aim to improve the accuracy of industrial accident prediction. In addition, we would like to present an efficient method to improve the quality of image generation and increase the accuracy of image classification through the combination of GAN and ResNet50.

• Journal of Digital Convergence
• /
• v.14 no.8
• /
• pp.245-252
• /
• 2016

The purpose of this study is to make up for missing of weather informations from ASOS and AWS using artificial neural networks. We collected temperature, relative humidity and wind velocity for August during 5-yr (2011-2015) and sample designed artificial neural networks, assuming the Seoul weather station was missing. The result of sensitivity study on number of epoch shows that early stopping appeared at 2,000 epochs. Correlation between observation and prediction was higher than 0.6, especially temperature and humidity was higher than 0.9, 0.8 respectively. RMSE decreased gradually and training time increased exponentially with respect to increase of number of epochs. The predictability at 40 epoch was more than 80% effect on of improved results by the time the early stopping. It is expected to make it possible to use more detailed weather information via the rapid missing complemented by quick learning time within 2 seconds.