• The Journal of the Korea Contents Association
• /
• v.9 no.11
• /
• pp.289-296
• /
• 2009

In this study, I presented power control unit with potential use in the magnetic stimulation of biological systems. The effect of the magnetic stimulation depends on the geometry and orientation of the induced electric field as well as on the current pulse waveform delivered by the stimulator coil. TMS is achieved from the outside of the head using pulses of electromagnetic field that induce an electric field in the brain. There are numerous possibities in the applications TMS, such as diagnosis and therapy through the brain stimulation. These factors are very important to define the equipment requirements and characteristics in that the topology of the power supply and the size and geometry of the coil. The proposed solution is the generation of current pulses with variable amplitude and duration, according to a user defined input. Another solution is the topology that uses elements to store and transfer energy from the power source to the load. In addition to proposed topology, an adequate control strategy and right set of the power circuit parameters made possible to obtain unipolar waves and bipolar waves.

• Journal of Advanced Navigation Technology
• /
• v.13 no.1
• /
• pp.48-53
• /
• 2009

The CMOS RF front-end for Global Positioning System(GPS)are implemented in 1P8M CMOS $0.13{\mu}m$ process. The LNAs consist of LNA1 with high gain and low NF, and LNA2 with low gain and high IIP3 for supporting operation with active and passive antenna. the measured performances of both LNAs are 16.4/13.8 dB gain, 1.4/1.68 dB NF, and -8/-4.4 dBm IIP3 with 3.2/2 mA form 1.2 V supply, respectively. The quadrature downconversion mixer is followed by transimpedance amplifier with gain controllability from 27.5 to 41 dB. The front-end performances in LNA1 mode are 39.8 dB conversion gain, 2.2 dB NF, and -33.4 dBm IIP3 with 6.6 mW power consumption.

• Journal of the Korea Computer Graphics Society
• /
• v.23 no.3
• /
• pp.95-103
• /
• 2017

Horizon correction is a crucial stage for image composition enhancement. In this paper, we propose a deep learning based method for estimating the slanted angle of a photograph and correcting it. To estimate and correct the horizon direction, existing methods use hand-crafted low-level features such as lines, planes, and gradient distributions. However, these methods may not work well on the images that contain no lines or planes. To tackle this limitation and robustly estimate the slanted angle, we propose a convolutional neural network (CNN) based method to estimate the slanted angle by learning more generic features using a huge dataset. In addition, we utilize multiple adaptive spatial pooling layers to extract multi-scale image features for better performance. In the experimental results, we show our CNN-based approach robustly and accurately estimates the slanted angle of an image regardless of the image content, even if the image contains no lines or planes at all.

• Proceedings of the Korean Reliability Society Conference
• /
• 2000.11a
• /
• pp.257-267
• /
• 2000

Mostly, the economic analyses for replacement of major components of nuclear power Plants(NPPs) have been performed in deterministic ways. However, the analysis results are more or less affected by the uncertainties associated with input variables. Therefore, it is desirable to use a probabilistic economic analysis method to properly consider uncertainty of real problem. In this paper, the probabilistic economic analysis method and decision analysis technique are briefly described. The probabilistic economy analysis method using decision analysis will provide efficient and accurate way of economic analysis for the repair and/or replace mai or components of NPPs.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.42 no.6 s.336
• /
• pp.67-74
• /
• 2005

Crypto-systems have difficulties in designing hardware due to the various standards. We propose a programmable and configurable architecture for cryptography coprocessors to accommodate various crypto-systems. The proposed architecture has a 32 bit I/O interface and internal bus width, and consists of a programmable finite field arithmetic unit, an input/output unit, a register file, and a control unit. The crypto-system is determined by the micro-codes in memory of the control unit, and is configured by programming the micro-codes. The coprocessor has a modular structure so that the arithmetic unit can be replaced if a substitute has an appropriate 32 bit I/O interface. It can be used in many crypto-systems by re-programming the micro-codes for corresponding crypto-system or by replacing operation units. We implement an elliptic curve crypto-processor using the proposed architecture and compare it with other crypto-processors

• 전자공학회논문지 IE
• /
• v.43 no.3
• /
• pp.22-27
• /
• 2006

This paper presents the control system of high-efficiency automotive 35W metal-halide lamp ballast using the new control method and planar transformer. In this paper, the electronic ballast is designed so that digital control method by microcontroller can be applied to the electronic ballast for the lamp requirement and peripheral environment in order that metal-halide lamp, which has the complicated transient features, is to be applied to the automobile even if it has superior features in brightness, color rendering, light efficiency, and lifespan compared to the conventional halogen lamp. Also, the efficiency increase of the ballast is devised by being varied the switching frequency of Flyback Converter following the battery input voltage of the automobile. Being designed for high-frequency switching transformer of converter in planar form, reduction of loss, weight, overall size are realized and efficient power control in the automobile that had the limited energy and the limited space of battery is devised. The results of the proposed system is verified through various experiment results.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.27 no.9
• /
• pp.825-833
• /
• 2016

This paper introduces a direct-conversion CMOS RF transmitter for the IEEE 802.15.4 standard with a low-power high-gain up-conversion mixer designed in $0.18{\mu}m$ process. The designed RF DCT(Direct Conversion Transmitter) is composed of differential DAC(Digital to Analog Converter), passive low-pass filter, quadrature active mixer and drive amplifier. The most important characteristic in designing RF DCT is to satisfy the 2.4 GHz Zigbee standard in low power. The quadrature active mixer inside the proposed RF DCT provides enough high gain as well as sufficient linearity using a gain boosting technique. The measurement results for the proposed transmitter show very low power consumption of 7.8 mA, output power more than 0 dBm and ACPR (Adjacent Channel Power Ratio) of -30 dBc.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.9 no.1
• /
• pp.1-18
• /
• 2007

This study is about an arrangement method of trapezoidal pre-cast segment lining that can be applied in shield tunnel construction. Trapezoidal segment lining is formed by assembling tapered pre-cast concrete segments taking advantage of the tapered shape of pre-cast segments upon delivery on site. By calculating tapering of the segments manufactured in single type and rotating the segments when putting them together, a variety of tunnel alignments can be arranged in the most efficient way. Once the design criteria and tunnel alignment (straight or curved) is analyzed, the sequence of assembling trapezoidal segments in compliance with tunnel alignment will be computed. On site an operator can utilize the softwareto automatically determine sequential arrangement of trapezoidal segments. When the actual arrangement of segmental lining is different from the computed output, the operator can input the actually measured values to coincide the computerized calculation with the real status of assembly. Then the adjustment will be the basis of subsequent arrangement of segments, thus the continuity of work can be guaranteed.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.10
• /
• pp.2341-2349
• /
• 2015

The commonly used Goldschmidt's floating-point divider algorithm performs two multiplications in one iteration. In this paper, a tentative error corrected K'th Goldschmidt's floating-point number divider algorithm which performs K times multiplications in one iteration is proposed. Since the number of multiplications performed by the proposed algorithm is dependent on the input values, the average number of multiplications per an operation in single precision and double precision divider is derived from many reciprocal tables with varying sizes. In addition, an error correction algorithm, which consists of one multiplication and a decision, to get exact result in divider is proposed. Since the proposed algorithm only performs the multiplications until the error gets smaller than a given value, it can be used to improve the performance of a divider unit. Also, it can be used to construct optimized approximate reciprocal tables.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.5
• /
• pp.592-601
• /
• 2016

The recurrent neural network (RNN) is a deep learning model which is suitable to sequential or length-variable data. The Long Short-Term Memory (LSTM) mitigates the vanishing gradient problem of RNNs so that LSTM can maintain the long-term dependency among the constituents of the given input sequence. In this paper, we propose a LSTM based language model which can predict following words of a given incomplete sentence to generate a complete sentence. To evaluate our method, we trained our model using multiple Korean corpora then generated the incomplete part of Korean sentences. The result shows that our language model was able to generate the fluent Korean sentences. We also show that the word based model generated better sentences compared to the other settings.