• Proceedings of the KIEE Conference
• /
• 2003.07d
• /
• pp.2082-2084
• /
• 2003

A practical real-time LOS rate estimator is proposed to handle the time-varying measurement noise statistics. To calculate the optimal Kalman gain, the algebraic transformation method is taken into account. By using the algebraic transformation, the differential algebraic Riccati equation(DARE) regarding estimation error covariance is replaced by the simple algebraic Riccati equation(ARE). The proposed LOS estimation filter gain is only a function of relative range. Consequently, the proposed method is computationally very efficient and suitable for embedded environment.

• Proceedings of the PSK Conference
• /
• 2002.10a
• /
• pp.412.1-412.1
• /
• 2002

Purpose. To prepare PVP-based solid dispersions containing lovastatin (LOS) and solubilizers (sodium lauryl sulfate. Tween80. oleic acid) to enhance dissolution of practically insoluble LOS. Methods. Solid dispersions containing LOS were prepared by dissolving two different organic solvent systems (acetone/ethanol or acetonitrile/ethanol). Results. The stickiness and flowability of solid dispersion powders were dependent on the composition and ratio of the solubilizers. (omitted)

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.11a
• /
• pp.164-166
• /
• 2022

Long-tail problem은 class 별로 sample의 개수에 차이가 있어 성능에 안 좋은 영향을 미치는 것을 말한다. 본 논문에서는 cost-sensitive learning 중 Class-Balanced Loss를 이용해 성능을 개선하여 Long-tail problem을 해결하려고 한다. 먼저, balanced data set과 imbalanced data set의 성능 차이를 살펴보도록 할 것이다. 그 후, Class-Balanced Loss를 3가지 버전으로 이용해 그 성능을 측정하고 분석해 볼 것이다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.2
• /
• pp.841-854
• /
• 2020

The paper explains the method to process, analyze and predict traffic patterns in Los Angeles county using Big Data and Machine Learning. The dataset is used from a popular navigating platform in the USA, which tracks information on the road using connected users' devices and also collects reports shared by the users through the app. The dataset mainly consists of information about traffic jams and traffic incidents reported by users, such as road closure, hazards, accidents. The major contribution of this paper is to give a clear view of how the large-scale road traffic data can be stored and processed using the Big Data system - Hadoop and its ecosystem (Hive). In addition, analysis is explained with the help of visuals using Business Intelligence and prediction with classification machine learning model on the sampled traffic data is presented using Azure ML. The process of modeling, as well as results, are interpreted using metrics: accuracy, precision and recall.

• International Journal of Internet, Broadcasting and Communication
• /
• v.12 no.1
• /
• pp.116-121
• /
• 2020

With the recent development of the 4th Industrial Revolution, efforts have been made to carry out communication in a smart factory, farm, etc. at low cost and reliably. Methods for utilizing empty frequencies using TVWS(TV White Space) have been studied which can be used locally within 30 km. However, there have not been many studies on Path-Loss model considering Korean environment. In this paper, the Path-Loss model is divided into LOS (Line Of Sight) and NLOS (None LOS). In case of LOS model, we checked the difference between Free space model, Friis model, 2-ray model and Hata model with measured data. In the case of NLOS model, we checked the difference between Lee Model, ITU-R526-3 Model with measured data. In order to overcome the difference in Korean environment, we derived a model that can be applied in LOS and NLOS and prove its usefulness through performance evaluation through simulation.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.5 no.3
• /
• pp.152-156
• /
• 2010

Non-line-of-sight (NLOS) propagation is one of the challenges in radio positioning. Distinguishing the transmission status of the communication as line-of-sight (LOS) or NLOS is of great importance for the wireless communication systems. This paper focuses on the identification of NLOS based on time-of-arrival (TOA) distance estimates and the received signal strength (RSS) measurements. We set a path loss threshold based on the joint TOA and RSS based NLOS detection method to determine LOS or NLOS. Simulation results show that the proposed method ensures the correct of detection for the LOS condition and can improve the NLOS identification for the weak noise and long distance.

• The Korean Journal of Health Service Management
• /
• v.10 no.1
• /
• pp.39-51
• /
• 2016

Objectives : This study investigated the health care utilization patterns of workers' compensation insurance(WCI) pneumoconiosis patients with excessively long hospital stays. Methods : The discharge summary data of 3,094 WCI pneumoconiosis patients were analyzed. The study sample was divided into 3 groups based on the length of stay(LOS). Health care utilization patterns were compared among the groups with logistic regression analysis, and the LOS determinants were identified with linear regression analysis. Results : The average LOS of the 222 long stay group patients was 1,448 days. Patients in this group tended to use private general hospitals, were admitted through the emergency room and discharged without the consent of a doctor. Conclusions : Many of the long LOS patients will maintain their inpatient status for the rest of their lives. For quality of life and efficient use of health care resources, policy makers need to establish a policy that enables patients to receive outpatient care in appropriate living conditions outside the hospital.

• Korean Journal of Optics and Photonics
• /
• v.6 no.4
• /
• pp.373-378
• /
• 1995

Bending losses of bent optical waveguides are calculated by virtue of the finite difference-beam p propagation method in the cylindrical coordinate system. In order to minimize the radiating losses of bent optical waveguides, we apply the trench structure to the bent waveguides and perform the a analysis to keep track of: 1) the influence of curvature radius on the bending loss without the trench, 2) the influence of curvature radius and refractive index difference on the bending loss with the trench, 3) the influence of the trench width on the bending loss.

• Proceedings of the KIPE Conference
• /
• 2005.07a
• /
• pp.140-142
• /
• 2005

SMPS의 손실에는 switching loss, conduction loss, core loss가 있다. 최근 SMPS에서는 switching loss를 줄여 효율을 높이고자 반도체 스위치 2개를 사용하는 공진형 구조가 증가하고 있다하지만 공진형 구조는 반도체 스위치에서 소비되는 conduction loss로 의해 기존의 컨버터에 비해 무부하시 전력 소모가 크다. 그러나 최근 시장은 무부하시 소비되는 대기전력의 규제가 이슈가 되고 있다. 본 논문에서는 active clamped flyback converter에서 무부하시 반도체 스위치의 conduction loss의 감소를 위해 Clamp 회로의 보조 스위치는 동작시키지 않고, flyback converter로만 동작하도록 설계하여 무부하시에는 기존의 flyback converter의 동작과 같이 도통 손실이 급속히 줄도록 하였다. 또한 스위칭 손실을 줄이기 위해 주 스위치의 동작 주파수를 감소시켜 SMPS의 무부하시의 소모를 감소시켰다. 70W급 SMPS의 제작과 실험을 통해 위의 방법을 증명하고자 한다.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.26 no.1
• /
• pp.72-82
• /
• 2023

The EOTS(Electro Optical Tracking System) must have stabilization performance to provide high-quality images under disturbance environment. In this paper, we present a controller that can minimize the LOS error and has a simple structure. Hence, to evaluate the performance of this controller, analysis in the frequency domain and LOS error measurement are performed. In order to measure the LOS error without a 'rate table' that requires a lot of facility investment, we propose a design method for disturbance model that considers the operating environment of the EOTS. Finally, the performance of the stabilization algorithm is evaluated by the proposed disturbance model.