• Journal of the Korean Institute of Telematics and Electronics D
• /
• v.36D no.5
• /
• pp.69-79
• /
• 1999

In this paper, we proposed a diffuion model of Ti diffused lithium niobate optical waveguide for fabricating waveguides with high refractive index and compared with conventional one. The achivement of low optical insertion loss between waveguide interface and single mode fibers was discussed as a function of Ti thickness for $\lambda$=1.55$\mu\textrm{m}$ The proposed diffusion method exhibited higher refractive index waveguide than conventional one for $\lambda$=0.6328$\mu\textrm{m}$ We have achieved the total fiber-waveguide-fiber insertion loss as low as 0.5dB/cm in z-cut and 1$\pm$0.5dB/cm in x-cut for both TM and TE mode of Mach-Zehnder interferometric waveguide in the range of Ti thickness 1000-1400$\AA$ for $\lambda$=1.55$\mu\textrm{m}$ From these results, this diffusion model for making a low loss waveguide can be used for low-power-modulators and switches.

• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.1122-1128
• /
• 2020

Bandwidth Extension refers to restoring and expanding a narrow band signal(NB) that is damaged or damaged in the encoding and decoding process due to the lack of channel capacity or the characteristics of the codec installed in the mobile communication device. It means converting to a wideband signal(WB). Bandwidth extension research mainly focuses on voice signals and converts high bands into frequency domains, such as SBR (Spectral Band Replication) and IGF (Intelligent Gap Filling), and restores disappeared or damaged high bands based on complex feature extraction processes. In this paper, we propose a model that outputs an bandwidth extended signal based on an autoencoder among deep learning models, using the residual connection of one-dimensional convolutional neural networks (CNN), the bandwidth is extended by inputting a time domain signal of a certain length without complicated pre-processing. In addition, it was confirmed that the damaged high band can be restored even by training on a dataset containing various types of sound sources including music that is not limited to the speech.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.11a
• /
• pp.713-716
• /
• 2007

오늘날 인간 프로젝트와 같은 종합적인 연구의 궁극적 목적을 달성하기 위해서는 이들 연구로부터 획득한 대량의 관련 데이터에 대해 새로운 현실적 의미를 부여할 수 있어야 한다. 따라서 현재의 마이크로어레이 기술을 이용해서 효과적으로 종양을 분류하기 위해서는 특정 종양 분류와 밀접하게 관련이 있는 정보력 있는 유전자를 선택하는 과정이 필수적이다. 본 논문에서는 암에 걸린 흰쥐 외피 기간 세포 분화 실험에서 얻어진 3840 유전자의 마이크로어레이 cDNA를 이용해 데이터의 정규화를 거쳐 유사성 척도 방법으로 정보력 있는 유전자들을 추출한 후, DT, NB, SVM, MLP 알고리즘을 이용하여 클래스 분류 모델을 구축하고, 성능을 비교분석하였다. 피어슨 적률 상관 계수를 이용하여 선택된 50 유전자들을 멀티퍼셉트론 분류기로 분류한 결과 94.8%의 정확도를 보여 가장 최적의 조합을 보였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.341-344
• /
• 2024

최근 랜섬웨어의 유포 증가로 인한 금전적 피해가 전세계적으로 급증하고 있다. 랜섬웨어는 사용자의 데이터를 암호화하여 금전을 요구하거나, 사용자의 중요하고 민감한 데이터를 파괴하여 사용하지 못하도록 피해를 입힌다. 이러한 피해를 막기 위해 파일의 API calls 이나, opcode 를 이용하는 탐지 및 분류 연구가 활발하게 진행되고 있다. 본 논문에서는 랜섬웨어를 효과적으로 탐지하기 위해 파일 PE 기능 값을 PCA 와 Wrapper 방법으로 데이터 전처리 후 머신러닝으로 학습하고, 학습한 모델을 활용하여 랜섬웨어를 정상과 악성으로 분류하는 방법을 제안한다. 제안한 방법으로 실험 결과 RF 는 98.25%, DT 96.25%, SVM 95%, NB 83%의 분류 정확도를 보였으며, RF 모델에서 가장 높은 분류 정확도를 달성하였다.

• KNOM Review
• /
• v.23 no.2
• /
• pp.18-28
• /
• 2020

With the development of the virtual community, the benefits that IT technology provides to people in fields such as healthcare, industry, communication, and culture are increasing, and the quality of life is also improving. Accordingly, there are various malicious attacks targeting the developed network environment. Firewalls and intrusion detection systems exist to detect these attacks in advance, but there is a limit to detecting malicious attacks that are evolving day by day. In order to solve this problem, intrusion detection research using machine learning is being actively conducted, but false positives and false negatives are occurring due to imbalance of the learning dataset. In this paper, a Random Oversampling method is used to solve the unbalance problem of the UNSW-NB15 dataset used for network intrusion detection. And through experiments, we compared and analyzed the accuracy, precision, recall, F1-score, training and prediction time, and hardware resource consumption of the models. Based on this study using the Random Oversampling method, we develop a more efficient network intrusion detection model study using other methods and high-performance models that can solve the unbalanced data problem.

• Korean Journal of Clinical Laboratory Science
• /
• v.52 no.2
• /
• pp.112-118
• /
• 2020

This study was undertaken to examine the effects and to investigate the relevant mechanisms of overexpressing stromal cell-derived factor-1 (SDF-1) produced by engineered mesenchymal stem cells, in a neurogenic bladder (NB) rat model. Sprague-Dawley (SD) rats (N=48) were randomly divided into 4 groups comprising 12 rats each: control group, Injury group, Injury+imMSC group, and Injury+SDF-1 eMSC group. Rats in the Injury+imMSC group were treated with imMSCs, whereas the Injury+SDF-1 eMSC group were administered SDF-1 eMSCs. After 4-weeks therapy, the bladder and pelvic nerve (PN) tissues were examined by subjecting to Masson's trichrome staining and immunofluorescence. Administration of SDF-1 eMSC resulted in improved smooth muscle content in the bladder tissue, significantly increased β-III tubulin expression of the PN, and enhanced SDF-1 expression (P<0.05). The bladder wall repair can be attributed to the overexpression of SDF-1 by SDF-1 eMSCs. Significantly increased SDF-1 expression was obtained in the Injury+SDF-1 eMSC group (P<0.05). The crushed PN also showed significant recovery in the Injury+SDF-1 eMSC group (P<0.05). In conclusion, our results indicate that SDF-1 eMSCs express more SDF-1 in vivo, thereby facilitating the repair of injured nerve and recovery of NB in rats.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.10
• /
• pp.2113-2120
• /
• 2011

In this paper, so as to improve spectral efficiency for cell-boundary users, we introduce a coordinated multi-point (CoMP) system, which is one of inter-cell cooperative transmission strategies studied in 3GPP long-term evolution-advanced (LTE-A) systems, and develop a system-level simulator to evaluate performance. To identify performance improvement of the system with inter-cell cooperative transmission, we select a 3GPP LTE system as a reference, which shows the highest performance among the existing mobile communication systems, and conduct a performance comparison. System-level simulation is performed based on widely-used OPNET tool. We implement modules including central unit (CU), CoMP eNodeB (CeNB), user equipment (UE), and multiple-input multiple-output (MIMO) channel model, while designing the inter-cell cooperative transmission system. Under WINNER wireless channel model and international telecommunication union (ITU) network model environments, we then evaluate the performance of edge users who belong to the lower 5% in terms of spectral efficiency. It is finally shown that throughput of the proposed CoMP system gets improved up to 2.5 times compared to that of the 3GPP LTE reference system.

• Journal of Digital Contents Society
• /
• v.19 no.2
• /
• pp.363-370
• /
• 2018

Ransomware detection has become a hot topic in computer security for protecting digital contents. Unfortunately, current signature-based and static detection models are often easily evadable by compress, and encryption. For overcoming the lack of these detection approach, we have proposed the dynamic ransomware detection system using data mining techniques such as RF, SVM, SL and NB algorithms. We monitor the actual behaviors of software to generate API calls flow graphs. Thereafter, data normalization and feature selection were applied to select informative features. We improved this analysis process. Finally, the data mining algorithms were used for building the detection model for judging whether the software is benign software or ransomware. We conduct our experiment using more suitable real ransomware samples. and it's results show that our proposed system can be more effective to improve the performance for ransomware detection.

• Journal of the Korea Convergence Society
• /
• v.13 no.3
• /
• pp.33-41
• /
• 2022

In this article, we proposed to predict natural gas (NG) leakage levels through feature selection based on a factor analysis (FA) of the integrating the Korean Meteorological Agency data and natural gas leakage data for considering complex factors. The paper has been divided into three modules. First, we filled missing data based on the linear interpolation method on the integrated data set, and selected essential features using FA with OrdinalEncoder (OE)-based normalization. The dataset is labeled by K-means clustering. The final module uses four algorithms, K-nearest neighbors (KNN), decision tree (DT), random forest (RF), Naive Bayes (NB), to predict gas leakage levels. The proposed method is evaluated by the accuracy, area under the ROC curve (AUC), and mean standard error (MSE). The test results indicate that the OrdinalEncoder-Factor analysis (OE-F)-based classification method has improved successfully. Moreover, OE-F-based KNN (OE-F-KNN) showed the best performance by giving 95.20% accuracy, an AUC of 96.13%, and an MSE of 0.031.

• Proceedings of the KWS Conference
• /
• 2005.11a
• /
• pp.201-203
• /
• 2005

A kinetic model for particle coarsening behavior in the weld heat affected zone (HAZ) was proposed. Unlike the conventional approach, where the mean-sized particle is considered to grow continuously, the proposed model considered the critical particle size which can be derived from the Gibbs-Thomson equation. In this study, the proposed particle coarsening model was applied to study the behavior of (Ti, Nb )(C, N) complex particle in the weld HAZ. The predicted particle size distributions using the proposed model were in agreement with the experimental results.