• Journal of Surface Science and Engineering
• /
• v.32 no.2
• /
• pp.93-99
• /
• 1999

Surface polymer layer formed on the silicon wafer during the oxide overetching using $C_4F_8$/ helicon wave plasmas and their characteristics were investigated using spectroscopic elipsometry, X-ray photoelectron spectroscopy, and secondary ion mass spectrometry. Overetch percentage and dc-self bias voltage were varied to investigate the effects on the characteristics of the polymers remaining on the overetched silicon surface. The increase of bias voltage from -80 volts to -120 volts increased the C/F ratio and carbon bondings such as C-C, $C-CF_x$/, and C-Si in the polymer while reducing the thickness of the polymer layer. However, the increase of the overetch percentage from 50% to 100% did not change the composition of the polymer layer and the carbon bondings in the polymer layer remained same even though it increased the polymer thickness. The polymer layer formed at the higher dc-self bias voltage was more difficult to be removed by the following various post-etch treatments compared to that formed at the longer overetch percentage.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.40 no.11
• /
• pp.83-94
• /
• 2003

In this paper. a highly linear and low glitch CMOS current mode digital-to-analog converter (DAC) by self calibration bias circuit is proposed. The architecture of the DAC is based on a current steering 6＋4 segmented type and new switching scheme for the current cell matrix, which reduced non-linearity error and graded error. In order to achieve a high performance DAC . novel current cell with a low spurious deglitching circuit and a new inverse thermometer decoder are proposed. The prototype DAC was implemented in a 0.35${\mu}{\textrm}{m}$ n-well CMOS technology. Experimental result show that SFDR is 60 ㏈ when sampling frequency is 32MHz and DAC output frequency is 7.92MHz. The DAC dissipates 46 mW at a 3.3 Volt single power supply and occupies a chip area of 1350${\mu}{\textrm}{m}$ ${\times}$750${\mu}{\textrm}{m}$.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2001.05a
• /
• pp.35-37
• /
• 2001

The Al etching was studied employing negative ions generated in the downstream $Cl_2$ plasma. In order to etch the Al film practically on an insulator covered electrode coupled with RF power, reduction of the negative self bias voltage (Vdc) was examined using a magnetic filter which trapped electrons. Addition of $SF_6$ and $H_2$ to a $Cl_2/BCl_3$ mixture reduced significantly Vdc.

• Journal of the Society of Naval Architects of Korea
• /
• v.40 no.5
• /
• pp.1-9
• /
• 2003

To predict the powering performance of full scale ships from the towing tank tests, resistance, propeller open water and self-propulsion tests are conducted. Model tests inevitably include the experimental error defined as the sum of two types of uncertainties, bias and precision errors. The induced errors in each element of model test are propagated through various routes and correlated with one another. The correlation coefficients are very important in the uncertainty analysis. The coefficient gives a direction(increase or decrease) for a value of error in individual elements. If the coefficient is not used accurately, the error bounds of the individual elements are overestimated or underestimated. In this study, the new methodology is applied to the uncertainty analysis of HMRI's towing tank tests, thus error bounds of each element is suggested and verified by several repetitive experiments.

• Journal of the Korean Vacuum Society
• /
• v.3 no.2
• /
• pp.179-185
• /
• 1994

음극에 DLC 필름이 놓여져 있는 용량결합형의 RF 플라즈마 장치에서 Ar가스에 의한 방전에서 발생된 입자의 분포를 레이져 산란에 의하여 관측하였다. 발생된 입자들은 플라즈마와 sheath의 경계면 에서 높은 밀도의 구름을 이루었으며 시간에 따라 주기적인 분포의 변화가 반복되었다, 입자 구름의 발 생은 플라즈마 물성의 변화를 야기하였으며 그 결과로 심한 self-bias 전위의 감소현상이 관측되었다. 입자 구름분포의 시간에 따른 변화와 같은 주기의 self-bias 플라즈마전위의 진동현상이 가열된 fast-scanning langmuir 탐침에 의하여 관측되었다. 이결과는 입자 표면에로의 음전하 누적에 따른 전체 음전하의 이동도 감소에 의한 것으로 해석된다. 또한 방출 분광법에 의하여 입자오염상태의 Ar 플라즈 마와 정상상태의 Ar 플라즈마의 방출 선세기의 변화를 관측하였는데 입자구름 오염시의 2차전자 차폐 현상에 의해 높은 문턱 에너지를 가진 Ar II 선의 세기 감소현상이 나타났다.

• Proceedings of the Korea Electromagnetic Engineering Society Conference
• /
• 2002.11a
• /
• pp.29-33
• /
• 2002

본 논문은 자기 바이어스(self bias)를 이용한 PCS 대역용 하이브리드 전력 저잡음 증폭기(power LNA) 모듈에 관한 것으로 GaAs p-HEMT 칩을 세라믹 기판에 실장하여 와이어 본딩과 주변 매칭을 통해 고주파 손실을 줄이고 온도 변화에 대한 안정성과 1.2㏈의 저잡음, 21~23㏈m의 P$_1$㏈를 실현하였다. 10mm$\times$10mm 크기로 표면 실장이 되도록 단자를 cut-line 형태로 모듈화 하여 안정성과 신뢰성을 향상시켰고 또한 저가격화를 실현하였다.

• Proceedings of the KIEE Conference
• /
• 1995.11a
• /
• pp.323-325
• /
• 1995

An Inductively Coupled Plasma(ICP) was employed to prepare the Diamond-Like Carbon film with $CH_4$ gas. We observed the changes of mechanical, optical properties and internal stress of the films according to the variation of discharge power and negative-self bias. When weak magnetic field is applied, the properties of film are observed to change drastically. In magnetized case, the micro-hardness and the internal stress increase up to critical point and droped down in marked contrast to unmagnetized case. It suggests that large amount of ion flux exists due to high dissociation rate of the reactive radicals in plasma with magnetic field as reported elsewhere. As a result of FT-IR absorption measurement it could be confirmed that the $CH_x$ bonding and the micro-hardness and the internal stress decreased with the increase of negative-self bias.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.7
• /
• pp.2407-2424
• /
• 2022

Recently, Transformer has made great progress in video retrieval tasks due to its high representation capability. For the structure of a Transformer, the cascaded self-attention modules are capable of capturing long-distance feature dependencies. However, the local feature details are likely to have deteriorated. In addition, increasing the depth of the structure is likely to produce learning bias in the learned features. In this paper, an improved Transformer structure named TransDCS (Transformer with Dynamic Convolution and Shortcut) is proposed. A Multi-head Conv-Self-Attention module is introduced to model the local dependencies and improve the efficiency of local features extraction. Meanwhile, the augmented shortcuts module based on a dual identity matrix is applied to enhance the conduction of input features, and mitigate the learning bias. The proposed model is tested on MSRVTT, LSMDC and Activity-Net benchmarks, and it surpasses all previous solutions for the video-text retrieval task. For example, on the LSMDC benchmark, a gain of about 2.3% MdR and 6.1% MnR is obtained over recently proposed multimodal-based methods.

• Korean Journal of School Psychology
• /
• v.16 no.2
• /
• pp.207-230
• /
• 2019

This study examined the effect of Attention Feedback Awareness and Control Training(A-FACT) on attention bias and generalized anxiety symptoms in college students. A total of 31 college students with at least 10 points on the Generalized Anxiety Disorder 7-item (GAD-7) scale or at least 56 points on the Korean version of the Penn State Worry Questionnaire (K-PSWQ) with attention bias were randomly assigned to one of three groups: A-FACT( n = 11), Attention Bias Modification (ABM)(n = 10) and Active Placebo Control (APC)(n = 10). Participants in A-FACT group received real-time feedback on attention bias based on their Baseline Neutral Response time(BNR) during A-FACT using a dot probe task. Participants in the ABM group received standard ABM, and those in the APC performed a dot probe task that they were informed was a program to reduce attention bias, but feedback was not provided. A total of eight sessions was conducted twice a week over a 4-week period. After every two sessions, GAD-7, K-PSWQ and K-STAI were rated. The effect of attention bias modification training was rated by changes in the Attention Bias Score(ABS), and in GAD-7, K-PSWQ and K-STAI scores. The results of repeated measure ANOVA indicated that the A-FACT group showed a significant decrease in ABS as well as in GAD-7, K-PSWQ and K-STAI scores compared to the other groups. Current results suggest that self-regulatory control of attention, that is, recognition of bias through feedback in A-FACT, may be effective in alleviating attention bias and generalized anxiety symptoms by recognizing bias through feedback on bias in attention bias modification training.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2006.02a
• /
• pp.118-118
• /
• 2006