• 산업경영시스템학회지
• /
• 제39권2호
• /
• pp.37-45
• /
• 2016

Control chart is representative tools of statistical process control (SPC). It is a graph that plotting the characteristic values from the process. It has two steps (or Phase). First step is a procedure for finding a process parameters. It is called Phase I. This step is to find the process parameters by using data obtained from in-controlled process. It is a step that the standard value was not determined. Another step is monitoring process by already known process parameters from Phase I. It is called Phase II. These control chart is the process quality characteristic value for management, which is plotted dot whether the existence within the control limit or not. But, this is not given information about the economic loss that occurs when a product characteristic value does not match the target value. In order to meet the customer needs, company not only consider stability of the process variation but also produce the product that is meet the target value. Taguchi's quadratic loss function is include information about economic loss that occurred by the mismatch the target value. However, Taguchi's quadratic loss function is very simple quadratic curve. It is difficult to realistically reflect the increased amount of loss that due to a deviation from the target value. Also, it can be well explained by only on condition that the normal process. Spiring proposed an alternative loss function that called reflected normal loss function (RNLF). In this paper, we design a new control chart for overcome these disadvantage by using the Spiring's RNLF. And we demonstrate effectiveness of new control chart by comparing its average run length (ARL) with ${\bar{x}}-R$ control chart and expected loss control chart (ELCC).

• 대한금속재료학회지
• /
• 제49권4호
• /
• pp.329-333
• /
• 2011

To develop wide-band noise absorbers with a special design for low-frequency performance, this study proposes a tin oxide (Sn-O) thin films as the noise absorbing materials in a microstrip line. Sn-O thin films were deposited on polyimide film substrates by reactive sputtering of the Sn target under flowing $O_{2}$ gas, exhibiting a wide variation of surface resistance (in the range of $10^{0}-10^{5}{\Omega}$) depending on the oxygen partial pressure during deposition. The microstrip line with characteristic impedance of $50\Omega$ was used for the measurement of noise absorption by the Sn-O films. The reflection parameter $(S_{11})$ increased with a decrease of surface resistance due to an impedance mismatch at the boundary between the film and the microstrip line. Meanwhile, the transmission parameter $(S_{21})$ diminished with a decrease of surface resistance resulting from an Ohmic loss of the Sn-O films. The maximum noise absorption predicted at an optimum surface resistance of the Sn-O films was about $150{\Omega}$. For this film, greater power absorption is predicted in the lower frequency region (about 70% at 1 GHz) than in conventional magnetic sheets of high magnetic loss, indicating that Ohmic loss is the predominant loss parameter for the conduction noise absorption in the low frequency band.

• 대한전자공학회논문지SD
• /
• 제39권3호
• /
• pp.41-48
• /
• 2002

진행파형 광 검출기의 대역폭 제한 요소로 천이시간(transit time), 광파와 마이크로파 사이의 속도 부정합, 광 흡수계수, impedance 매칭, 분산(dispersion), 마이크로파 손실 등과 같은 요소들 사이의 복합적인 영향에 의해 결정되어진다. 본 논문에서는 기존에 제시된 바 있는 속도 부정합 임펄스 응답에 천이시간을 고려하여 새로이 진행파형 광 검출기에서의 임펄스 응답을 제시하였다. 이 임펄스 응답을 이용하여 1.55㎛ 진행파형 CPW 광 검출기에서 천이시간, 광과 마이크로파 사이의 속도 부정합, 그리고 광 구속인자가 소자의 대역폭에 미치는 영향에 대하여 분석하였고, 정확한 계산을 위해 마이크로파의 위상속도는 3차원 FDTD를 이용하여 계산했다. 대역폭 개선을 위해 진성 흡수영역의 두께를 줄여가며 그 영향을 살펴보았고, α=0.2㎛ -1 인 경우 천이시간과 광 구속인자의 영향에 의해 결정되는 최적의 진성 영역 두께는 0.2㎛가 되는 것을 본 논문에서 제시한다.

• 대한전자공학회논문지SD
• /
• 제40권8호
• /
• pp.580-587
• /
• 2003

본 논문에서는 진행파형 전계흡수 변조기 (TW-EAM: traveling-wave electroabsorption modulator)의 선형성을 분석하기 위한 새로운 모델을 제시한다. TW-EAM은 소자의 길이, 마이크로파 손실 (microwave loss, ML), 그리고 임피던스 부정합에 의한 내부반사(internal reflection, IR) 등이 소자의 선형성에 영향을 미친다. 소자의 길이의 증가는 혼변조 왜곡 (intermodulation distortion, IMD)이 최소가 되는 전원전압의 크기를 감소시킨다. ML의 증가는 3차 혼변조 왜곡 (third-order IMD, IMD3)의 감소와 동시에 출력신호의 전력도 감소시킨다. IR은 입력주파수의 파장과 소자의 길이에 따라 각기 다른 IMD 특성을 나타낸다. ML 또는 IR에 의한 SFDR (spurious-free dynamic-range)의 변화는 거의 없었으며, TW-EAM의 IR을 이용하면 ML에 의한 신호 전력의 감쇄를 보상해 줄 수 있음도 알 수 있었다. 결과적으로 50 GHz 대역의 RF-광통신용 TW-EAM은 길이가 0.8 mm이고 출력단의 임피던스 부정합을 이용하면서 최소의 손실을 가지는 구조가 적당함을 알 수 있었다.

• 한국전기전자재료학회논문지
• /
• 제11권10호
• /
• pp.924-930
• /
• 1998

In this study, we describe the properties of an electro-optical modulator with CPW(coplanar waveguide) electrode fabricated on $LiNvO_3$ optical waveguide, that is applicable to optical communications. These optical modulators have features that use a buffer layer to reduce velocity mismatch between microwaves and optical wave as well as employ CPW to improve impedance and propagation mode mismatch between the electrode and the external circuits. And an annealed proton exchange technique for obtaining low-loss optical waveguides and good reproducibility was employed. Taking into consideration the mentioned background, to achieve the lower driving voltage of optical modulator, we have re-disigned the longer interaction length. And their device properties are discussed also. As a result, fabricated optical modulators of good 10Gps operation and low voltage(5.6V of the halfwave lengfth voltage) at an $1.5{\mu}m$ wavelength are achived with good reproducibility.

• 한국태양에너지학회 논문집
• /
• 제30권6호
• /
• pp.102-107
• /
• 2010

a-Si solar cell has relatively dominant drift current when compared with crystalline solar cell due to the high internal electric field. Such drift current make an impact on the PV module in the local shading. In this paper, the a-Si PV module output characteristics of shading effects was approached in terms of process condition, because of the different deposition layer of thin film lead to rising the resistance. We suggested design condition to ensure the long-term durability of the module with regard to the degradation factors such as hot spot by analyzing the module specification. The result shows a remarkable difference on module uniformity for each shading position. In addition, the unbalanced power loss due to power mismatch of each module could intensify the degradation.

• Journal of Advanced Marine Engineering and Technology
• /
• 제30권4호
• /
• pp.505-513
• /
• 2006

A linear power amplifier is particularly emphasized on the system using a linear modulations, such as 16QAM and QPSK with pulse shaping. because intermodulation distortion which causes adjacent channel interference and co-channel interference is mostly generated in a nonlinear power amplifier. In this paper, parameters of a linearization loop, such as an amplitude imbalance a phase imbalance and a delay mismatch, are briefly analyzed to get a specific cancellation performance and linearization bandwidth. Experimental results are presented for IMT-2000 frequency band. The center frequency of the feedforward amplifier is 2140 MHz with 60 MHz bandwidth. When the average output power of feedforward amplifier is 20 Watt. the intermodulation cancellation performance is more than 21 dB. In this case, the output power of feedforward amplifier reduced 3.5 dB because of extra delay line loss and coupling loss. The feedforward amplifier efficiency is more than 7.2 % for multicarrier signals, 59 dBc for ACPR.

• 한국마린엔지니어링학회:학술대회논문집
• /
• 한국마린엔지니어링학회 2005년도 전기학술대회논문집
• /
• pp.1176-1185
• /
• 2005

A linear power amplifier is particularly emphasized on the system using a linear modulations, such as 16QAM and QPSK with pulse shaping, because intermodulation distortion which causes adjacent channel interference and co-channel interference is mostly generated in a nonlinear power amplifier. In this paper, parameters of a linearization loop, such as an amplitude imbalance, a phase imbalance and a delay mismatch, are briefly analyzed to get a specific cancellation performance and linearization bandwidth. Experimental results are presented for IMT-2000 frequency band. The center frequency of the feedforward amplifier is 2140 MHz with 60 MHz bandwidth. When the average output power of feedforward amplifier is 20 Watt, the intermodulation cancellation performance is more than 21 dB. In this case, the output power of feedforward amplifier reduced 3.5 dB because of extra delay line loss and coupling loss. The feedforward amplifier efficiency is more than 7.2 % for multicarrier signals, 59 dBc for ACPR.

• 한국항해항만학회지
• /
• 제28권2호
• /
• pp.149-154
• /
• 2004

A linear power amplifier is particularly emphasized on the system using a linear modulations, such as 16QAM and QPSK with pulse shaping, because intermodulation distortion which causes adjacent channel interference and co-channel interference is mostly generated in a nonlinear power amplifier. In this paper, parameters of a linearization loop, such as an amplitude imbalance, a phase imbalance and a delay mismatch, are briefly analyzed to get a specific cancellation performance and linearization bandwidth Experimental results are presented for IMT-2000 frequency band The center frequency of the feedforward amplifier is 2140MHz with 60MHz bandwidth When the average output power of feedforward amplifier is 20 Watt, the intermodulation cancellation performance is more than 28dB. In this case, the output power of feedforward amplifier reduced 3.5dB because of extra delay line loss and coupling loss. The feedforward amplifier efficiency is more than 7% for multicarrier signals.

• 말소리와 음성과학
• /
• 제12권3호
• /
• pp.65-71
• /
• 2020

특징 정규화는 음성 특징 파라미터들의 통계적인 특성의 정규화를 통해 훈련 및 테스트 조건 사이의 환경 불일치의 영향을 감소시키는 방법으로서 기존의 Gaussian mixture model-hidden Markov model(GMM-HMM) 기반의 음성인식 시스템에서 우수한 성능개선을 입증한 바 있다. 하지만 심층신경망(deep neural network, DNN) 기반의 음성인식 시스템에서는 환경 불일치의 영향을 최소화 하는 것이 반드시 최고의 성능 개선으로 연결되지는 않는다. 본 논문에서는 이러한 현상의 원인을 과도한 특징 정규화로 인한 정보손실 때문이라 보고, 음향모델을 훈련 하는데 유용한 정보는 보존하면서 환경 불일치의 영향은 적절히 감소시켜 음성인식 성능을 최대화 하는 특징 정규화 방식이 있는 지 검토해보고자 한다. 이를 위해 평균 정규화(mean normalization, MN)와 평균 및 분산 정규화(mean and variance normalization, MVN)의 절충 방식인 평균 및 지수적 분산 정규화(mean and exponentiated variance normalization, MEVN)를 도입하여, 잡음 및 잔향 환경에서 분산에 대한 정규화의 정도에 따른 DNN 기반의 음성인식 시스템의 성능을 비교한다. 실험 결과, 성능 개선의 폭이 크지는 않으나 분산 정규화의 정도에 따라 MEVN이 MN과 MVN보다 성능이 우수함을 보여준다.