• 전자공학회논문지S
• /
• 제35S권12호
• /
• pp.103-108
• /
• 1998

디지털 자기기록 장치에서의 용량의 증가는 필연적으로 심각한 인접 심벌간의 간섭 (Intersymbol interference : ISI)과 비선형 왜곡현사을 겪게된다. 본 논문에서는 디지털 자기기록 채널의 심각한 심벌간의 간섭과 비선형 왜곡현상을 보상해 주기 위한 방법으로 신경망을 이용한 등화기를 부분 삭제 모델로 나타난 자기기록 채널에 적용하였다. 신경망을 이용한 결정궤환 등화기 (Neural Decision Feedback Equalizer : NDFE)와 일반적인 결정궤환 등화기 (Decision Feedback Equalizer : DFE)간의 성능을 컴퓨터를 이용한 모의 실험을 통해 비교하였다. 실험 결과, 신경망을 이용한 결정궤환 등화기가 기존의 결정궤환 등화기보다 디지털 자기기록 장치에서 기록 밀도가 증가함에 따라 같은 신호 대 잡음비 (Signal-to-Noise Ratio : SNR)에서 우수한 비트 오류 확률 (bit error probability or bit error ratio : BER) 성능을 보였다.. 또한 같은 밀도의 경우에도 비선형 왜곡 현상이 강할수록 신경망을 이용한 결정궤환 등화기가 보다 안정적이며 우수한 성능을 보였다.

• 한국분무공학회지
• /
• 제5권3호
• /
• pp.9-16
• /
• 2000

In a SI engine, three-way catalyst converter has the best efficiency when A/F ratio is near the stoichiometry. The feedback control using oxygen sensors in the commercial engine has limits caused by the system delays. So it is necessary to control fuel quantity in accordance with intake air amount in order to reduce exhaust emission and improve the specific fuel consumption. Precise A/F ratio control requires measurement of air amount with respect to the cylinder and injection fuel according to the air amount In this paper, we applied nonlinear fuel injection model and developed the algorithm of A/F ratio control. This algorithm includes the methods of measurement of transient air mass flowing into each cylinder, of calculation of injection pulse width for measured air mass, and the method of feedback and engine control by using lambda sensor. Also we developed control program for IBM-PC by using C++ Builder, and tested it in the commercial engine.

• Journal of Communications and Networks
• /
• 제16권1호
• /
• pp.1-9
• /
• 2014

Block diagonalization (BD) has been proposed as a simple and effective technique in multiuser multiple-input multiple-output (MU-MIMO) broadcast channels. However, when channel state information (CSI) knowledge is limited at the transmitter, the performance of the BD may be degraded because inter-user interference cannot be completely eliminated. In this paper, we propose an efficient CSI quantization technique for BD precoded systems with limited feedback where users supported by a base station are selected by dynamic scheduling. First, we express the received signal-to-interference-plus-noise ratio (SINR) when multiple data streams are transmitted to the user, and derive a lower bound expression of the expected received SINR at each user. Then, based on this measure, each user determines its quantized CSI feedback information which maximizes the derived expected SINR, which comprises both the channel direction and the amplitude information. From simulations, we confirm that the proposed SINR-based channel quantization scheme achieves a significant sum rate gain over the conventional method in practical MU-MIMO systems.

• Journal of Power Electronics
• /
• 제12권6호
• /
• pp.886-894
• /
• 2012

This paper presents a secondary-side, dual-mode feedback LLC resonant controller IC with dynamic PWM dimming for LED backlight units. In order to reduce the cost, master and slave outputs can be generated simultaneously with a single LLC resonant core based on dual-mode feedback topologies. Pulse Frequency Modulation (PFM) and Pulse Width Modulation (PWM) schemes are used for the master stage and slave stage, respectively. In order to guarantee the correct dual feedback operation, Phased-Locked Loop (PLL)-based automatic duty control circuit is proposed in this paper. The chip is fabricated using $0.35{\mu}m$ Bipolar-CMOS-DMOS (BCD) technology, and the die size is $2.5mm{\times}2.5mm$. The frequency of the gate driver (GDA/GDB) in the clock generator ranges from 50 to 425 kHz. The current consumption of the LLC resonant controller IC is 40 mA for a 100 kHz operation frequency using a 15 V supply. The duty ratio of the slave stage can be controlled from 40% to 60% independent of the frequency of the master stage.

• 천문학회보
• /
• 제44권2호
• /
• pp.36.4-36.4
• /
• 2019

Ricci et al. (2017, Nature, 549, 488) discovered a lack of high accretion rate, obscured Active Galactic Nuclei (AGN) in the hard X-ray selected Swift/BAT local AGN survey. This was interpreted as radiative pressure driven AGN feedback clearing its immediate vicinity composed of dusty gas (having an effectively low Eddington limit in the order of 0.01-0.1), and governing the level of nuclear obscuration. As we find Eddington-limited accretion and high extinction values among obscured, luminous AGN (quasars) however, it may be that the local X-ray AGN and the distant quasars undergo different feedback mechanisms in clearing their surroundings. In this study, we simply compare the obscuring column density and Eddington ratio values for quasars selected by various methods, including X-ray obscured, optically blue, infrared red/luminous, and submillimeter bright AGN. We find obscured quasars lying on the column density-Eddington ratio diagram previously unoccupied by Ricci et al., suggesting that radiative pressure is insufficient to clear its dusty structure at high luminosity, or that the dust in obscured quasars are more extended than the low luminosity counterparts to become fully transparent. We discuss alternative feedback scenarios that may be more relevant for obscured quasars.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 한국자동제어학술회의논문집(국내학술편); 포항공과대학교, 포항; 24-26 Oct. 1996
• /
• pp.1004-1007
• /
• 1996

Various auto-tuning methods using relay feedback are presented recently. They are composed of the consecutive procedures identifying Nyquist critical point using relay feedback and designing PID controller by one point of Nyquist plot mapping. This paper suggests a strategy to get the knowledges of Nyquist critical point and the neighborhood point of it using relay feedback. The parameters of PID controller are established by dominant pole approximation based on these knowledges. Designers can make use of the damping ratio as a time domain specification. So design flexibilities are taken in view of stability and performance of the system response considering practical system condition.

• Journal of Communications and Networks
• /
• 제13권4호
• /
• pp.339-344
• /
• 2011

Base station coordination has received much attention as a means to reduce the inter-cell interference in cellular networks. However, this interference reducing ability comes at the expense of increased feedback, backhaul load and computational complexity. The degree of coordination is therefore limited in practice. In this paper, we explore the trade-off between capacity and feedback load in a cellular network with coordination clusters. Our main interest lies in a scenario with multiple fading users in each cell. The results indicate that a large fraction of the total gain can be achieved by a significant reduction in feedback. We also find an approximate expression for the distribution of the instantaneous signal to interference-plus-noise ratio (SINR) and propose a new effective scheduling algorithm.

• ETRI Journal
• /
• 제32권4호
• /
• pp.548-554
• /
• 2010

The transcutaneous energy transmission system (TETS) composed of a Class-E amplifier may operate at a state away from the optimum power transmission due to the load variation. By introducing the feedback-loop technique, the TETS can keep the optimum state with constant output voltage by adjusting the important design parameters, that is, the duty ratio and frequency of the driving signal and the supply voltage. The relations between these adjusted parameters and the load are investigated. The effectiveness of the feedback technique is validated through a design example with a variable load parameter. The experimental results show that the Class-E amplifier in the feedback loop can keep operating at the optimum state under the condition of up to 50 percent variation of the load value.

• 조명전기설비학회논문지
• /
• 제17권4호
• /
• pp.36-44
• /
• 2003

전동기와 부하가 강성이 작은 축으로 연결되는 산업용 전동기 구동시스템에 있어서 전동기의 관성과 부하의 관성에 의해 축에 비틀림 진동이 발생한다. 이러한 문제를 해결하기 위하여 기존에 2자유도 H$_{\infty}$제어기가 제안되었다. 그러나 이 제어기는 구현하는데 어려움이 있다. 본 논문에서는 부분적인 상태궤환과 공진비제어를 이용한 2관성계의 H$_{\infty}$제어가 제안되었다. 제안된 제어기는 간단하면서도 외란 및 진동억제 성능을 만족한다.어기는 간단하면서도 외란 및 진동억제 성능을 만족한다.

• 컴퓨터교육학회논문지
• /
• 제7권1호
• /
• pp.89-96
• /
• 2004

일반적으로 형성 평가는 교수-학습 활동이 진행 중에 투입되는 평가활동으로 학습 정보에 대한 피드백과 교정 및 교수 방법의 개선을 위한 평가이다. 본 논문에서는 형성평가에서 학습자 개인의 평가 결과에 따라 서로 다른 피드백을 제공하는 시스템을 제안한다. 이 시스템은 지식, 이해, 적용 등의 영역별 기준에 따라 제작된 문항을 통해 각 학습자를 평가하고 이 결과를 기반으로 구분된 학습자 유형에 따라 서로 다른 피드백을 개별 학습자에게 제공하며 학습자는 이를 통해 부족한 영역을 보완하게 된다. 교수자에게는 문항별 정답 비율뿐만 아니라 학생 개개인에게 제공되었던 피드백 내용도 알려주어 교수방법을 개선하도록 한다.