• The Journal of the Convergence on Culture Technology
• /
• v.5 no.2
• /
• pp.355-360
• /
• 2019

In this study, the structural stability of the railway level crossing system using rubber panel for high speed lines was investigated by applying the specification for wind load conditions (Train gust) of high speed train (300km/h and 360km/h). A finite element analysis using three-dimensional modeling was carried out by applying the field conditions that was installed with the complicated configuration of the rubber plate panel system. As a result of this study, the structural stability of the rubber plate panel system for high speed train gust was analytically verified.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.16 no.4
• /
• pp.270-275
• /
• 2016

We present an approach for acceleration actor-critic algorithm for reinforcement learning with continuous action space. Actor-critic algorithm has already proved its robustness to the infinitely large action spaces in various high dimensional environments. Despite that success, the main problem of the actor-critic algorithm remains the same-speed of convergence to the optimal policy. In high dimensional state and action space, a searching for the correct action in each state takes enormously long time. Therefore, in this paper we suggest a search accelerating function that allows to leverage speed of algorithm convergence and reach optimal policy faster. In our method, we assume that actions may have their own distribution of preference, that independent on the state. Since in the beginning of learning agent act randomly in the environment, it would be more efficient if actions were taken according to the some heuristic function. We demonstrate that heuristically-accelerated actor-critic algorithm learns optimal policy faster, using Educational Process Mining dataset with records of students' course learning process and their grades.

• Journal of Korean Institute of Industrial Engineers
• /
• v.21 no.1
• /
• pp.33-49
• /
• 1995

Simulated annealing(SA) has been admitted as a general purpose optimization technique which can be utilized for almost all kinds of combinatorial optimization problems without much difficulty. But there are still some weak points to be resolved, one of which is the slow speed of convergence. In this study, we carefully review various previous efforts to improve SA and propose some variations of SA which can enhance the speed of convergence to the optimum solution. Then, we apply the revised SA algorithms to the scheduling and hardware allocation problems occurring in high-level synthesis(HLS) of VLSI design. We confirm the efficiency of the proposed methods through several HLS examples.

• Journal of the Korean Society of Industry Convergence
• /
• v.3 no.1
• /
• pp.9-15
• /
• 2000

This paper presents a implementation of digital motion control system for induction motor vector drives using the 16bit DSP TMS320F240. The DSP controller enable enhanced real time algorithm and cost-effective design of intelligent controllers for motors which can be yield enhanced operation, fewer system components, lower system cost, increased efficiency and high performance. The system presented are speed and current sensing, sine look-up table and generated SVPWM by fully integrated control software. The developed system in a implementation are shown a good speed response and motion control characteristic results, and high performance features in general purposed 2.2[kW] machine. The system can be adapted variform motor drive system.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.10 no.4
• /
• pp.143-150
• /
• 2014

A high power amplifier(HPA) has widely been used in wireless communications. In order to have a maximum power efficiency, the HPA uses a saturation domain having nonlinear characteristics, which generates nonlinear distortions to decrease the transmission efficiency. In order to solve this problem, various methods to linearize HPA have been reported until now, and the pre-distorter method is very attractive among them. In this paper, a new enhanced adaptive pre-distorter method was proposed to be applied for compensating nonlinear distortions of HPA. Through computer simulations as well as the analytical analysis, it could be shown that it is possible for both conventional adaptive pre-distorter and proposed adaptive pre-distorter, to be applied for linearizing HPA. Also, the simulations results demonstrated that the proposed adaptive pre-distorter may have faster convergence speed and better capability of compensating the nonlinear distortion than the conventional adaptive pre-distorter with nearly equal computation complexity.

• Journal of Advanced Navigation Technology
• /
• v.28 no.2
• /
• pp.246-249
• /
• 2024

In this paper, we present a miniaturized differential line (DL) using inductance-enhanced corrugated ground planes (LCGP) for effective common-mode (CM) noise suppression in high-speed packages and printed circuit boards. The LCGP-DL demonstrates the CM noise suppression in the frequency range from 2.09 GHz to 3.6 GHz. Furthermore, to achieve the same low cutoff frequency, the LCGP-DL accomplishes a remarkable 23.2% reduction in size compared to a reference DL.

• The Journal of the Acoustical Society of Korea
• /
• v.34 no.5
• /
• pp.394-402
• /
• 2015

CMA (Constant Modulus Algorithm) is one of the well-known algorithms in blind acoustic channel equalization. Generally, CMA converges slowly and the speed of convergence is dependent on a step-size in the CMA procedure. Many researches have tried to speed up the convergence speed by applying a variable step-size to CMA, e.g. the orthogonal projection CMA and algebraic optimal step-size CMA. In this paper, we summarize these two algorithms, and we propose a new CMA with improved convergence performance. The improvement comes from the switching between the orthogonal projection CMA and algebraic optimal step-size CMA. In simulation results, we show the performance improvement in the time invariant channels as well as in time varying channel.

• Journal of Environmental Science International
• /
• v.5 no.6
• /
• pp.713-718
• /
• 1996

We identified two characteristic turbulent flow cases, weakening and strengthening, which appear at the downwind side. Observations were made two times, Dec. 2-3. 1995 and Feb. 13-14. 1996 at Pusan National University site located downwind side of Kumjeong mountain. Meteorological observation system, tethersonde, was adopted to present observation. In the case of the west wind which blows perpendicular to Sanghak mountain located westward from the site, the wind speed highly increased in exponential with height. Therefore, the low level wind speed was so weak just like Taylor(1988)'s review. While the wind speed was intensified at 200-400m layer when the northwest wind blows from the continental Siberian high. We suppose 기 is because of the strong vertical convergence of flow between the surface inversion layer and the upper one, and also the horizontal convergence along the saddle and valley between the two mountains, Kumjeong and Sanghak-because of Bernoulli's effect. The inversion layer existed at surface-l00m and 500-600m level and the strong wind existed at about 200-400m layer.

• Journal of Institute of Control, Robotics and Systems
• /
• v.9 no.11
• /
• pp.917-924
• /
• 2003

Recurrent neural networks have been successfully applied to communications channel equalization. Major disadvantages of gradient-based learning algorithms commonly employed to train recurrent neural networks are slow convergence rates and long training sequences required for satisfactory performance. In a high-speed communications system, fast convergence speed and short training symbols are essential. We propose decision feedback equalizers using a recurrent neural network trained with Kalman filtering algorithms. The main features of the proposed recurrent neural equalizers, utilizing extended Kalman filter (EKF) and unscented Kalman filter (UKF), are fast convergence rates and good performance using relatively short training symbols. Experimental results for two time-varying channels are presented to evaluate the performance of the proposed approaches over a conventional recurrent neural equalizer.

• Journal of Digital Convergence
• /
• v.13 no.9
• /
• pp.219-224
• /
• 2015

The broadcasting communications service will accelerate its development with the convergence of broadcasting media and internet service. In the field of CATV network related technology, only those service providers will be able to survive in future, and they can combine available services in the most effective and economical way. This research aims to explain the CATV status of Korea and China and its technology trends. It also analyzes CMC(Cable Modem Concentrator) technology, suggested as one of the high-speed internet technology. CMC technology has the advantages of enhancing the transmission speed while using the existing basic structure of HFC network and expanding service area by adding amplifiers within CMC. The distance between coaxial cables is getting shorter with more concentrated areas in large cities in Korea. However, in China, the demand for long distance transmission service is increasing. CMC technology satisfies both short and long distance service subscribers without any geographical limitations. With these advantages, CMC technology is expected to generate lots of economic benefits if applied for the CATV network in the area of China, Middle East, and Southeast Asia.