• The Transactions of the Korea Information Processing Society
• /
• v.4 no.11
• /
• pp.2861-2873
• /
• 1997

An FPU(Floating Point unit) is the principle component in high performance computer and is placed on a chip together with main processing unit recently. As a Processing speed of the FPU is accelerated, the rounding stage, which occupies one of the floating point Processing steps for floating point operations, has a considerable effect on overall floating point operations. In this paper, by studying and analyzing the processing flows of the conventional floating point adder/subtractor, multipler and divider, which are main component of the FPU, efficient rounding mechanisms are presented. Proposed mechanisms do not require any additional execution time and any high speed adder for rounding operation. Thus, performance improvement and cost-effective design can be achieved by this approach.

• The Korean Journal of Pharmacology
• /
• v.21 no.2
• /
• pp.90-98
• /
• 1985

1) In the study of the forced-swimming test in mice (FSM), the duration of immobility posture was dose-dependently shortened by ${\alpha}_2$-agonists, clonidine and guanabenz. BH-T 933 and oxymetazoline also decreased it . Xylazine rather increased the immobility duration at low dose. 2) ${\alpha}_1$-Agonists, cirazoline, amidephrine and methoxamine, however, showed inconsistent effect on the immobility duration (ID). 3) The decrease in ID by clonidine and guanabenz was antagonized by pretreatment with yohimbine, idazoxan and phentolamine (${\alpha}_2$antagonist), but not by prazosin and corynanthine (${\alpha}_1$-antagonist) .4) The ID in the FSM was shortened dose-dependently by d-amphetamine, and it was also antagonized by yohimbine, but not by prazosin. 5) In the mice pretreated with either ${\alpha}$-methyl-p-tyrosine or reserpine, or with combination of both, the decrease in ID was still evoked by clonidine. 6) When the mice were chronically treated with antidepressants (desipramine and imipramine), or with electroconvulsive shock, clonidine still decreased the ID as it did in the control. 7) These results provided the evidences to hypothesize that the change of the ID in the FSM is closely related with the postsynaptie ${\alpha}_2$-adrenoceptor located on the central noradrenergic neuron body. Furthermore, it is assumed that this escape-directed behavior enhanced by ${\alpha}_2$-adrenoceptor agonist may be the result in some analogy with the incentive of drives which are directed toward the self-preservation.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2008.06a
• /
• pp.538-538
• /
• 2008

금속계열의 박막을 평탄화하기위해서는 슬러리에 함유된 산화제에 의해 부동태층의 형성이 선행되어야 한다. 따라서 본 논문에서는 Copper 박막의 표면을 부동태층으로 형성시키고 CMP공정을 하기위해 산화제에 dipping을 시켰으며 삼화제의 종류는 $H_2O_2$, MSW2000B, $KIO_3$로 하고 dipping 시간은 30초, 60초, 90초, 3분, 10분으로 하여 시간과 산화제 종류에 따른 부동태층의 변화를 연구하였다. 부동태층의 관찰은 FESEM을 이용하여 표면과 단면을 관찰하였고 부동태층의 조성비율은 EDX를 이용하여 조사하였다. MSW 2000B의 경우는 부동태층이 덩어리 모양으로 형성되었으며 포화현상은 3분에 일어났다. 반면에 $H_2O_2$의 경우는 부동태층이 침상 모양으로 형성되었으며 포화현상은 90초에 일어났다. 산화제에 의해 부동태층을 형성시킨 후 POLI-450을 이용하여 평탄화공정을 진행하였으며 CMP공정조건은 부동태층의 연질상태임을 감안하여 헤드 스피드 20rpm, 플레이튼 스피드 10rpm, 슬러리 주입속도 90ml/min, 공정온도는 상온으로 하여 진행하였다. $H_2O_2$를 산화제로 사용하여 dipping을 하고 CMP를 하였을 경우에 균일한 박막을 확보 할 수 있었으며 CMP 공정 후 copper 박막의 균일성은 FESEM을 이용하여 관찰 하였다.

• Journal of the Korean Chemical Society
• /
• v.28 no.5
• /
• pp.271-278
• /
• 1984

Ellipsometric and reflectance measurements were made with magneto-optically self-nulling ellipsometer on the iron surface being passivated. The passivation was induced by abruptly changing potential of the mechanically polished high purity iron from the reduction potential to the oxidation potential in basic solutions. From the differences in the optical paramates(${\Delta},\;{\psi}$) and reflectance (R) between the reduced (film-free) and oxidized (film-covered) states, the thickness(${\tau}$) and optical constants (n, k) of the film in the early stage of its formation were computed as functions of pH and time. From the computed values, it was deduced that the properties of the anodic film did not undergo a drastic change with time which would indicate a transformation of the film before effective passivity is attained, and that the film reached its stady state within a few second. The thickness of anodic film was $14\;{\sim}\;23{\AA}$. The anodic films also seemed to have small values of optical absorption coefficient. The film formed in high pH environments had thinner and denser structure than that formed in low pH.

• Korean Journal of Food Science and Technology
• /
• v.41 no.4
• /
• pp.413-416
• /
• 2009

The sap of the white birch is one of the most widely used herbal medicines to treat gastrointestinal tract, jaundice, and various periodontal diseases. The purpose of this study was to measure the effect of white birch sap on in vivo immune agents in mice. The treatment of white birch sap on the forced swimming test has been used as a screening model for new immune enhancement agents. The hematological examination was significantly different in neutrophil, lymphocyte, and monocyte counts between treatment and control groups. Moreover the white birch sap changed the phagocytic index, immobility duration time, and leukocyte counts in mice.

• 한국HCI학회:학술대회논문집
• /
• 2006.02a
• /
• pp.766-771
• /
• 2006

현재 많은 신경망의 하드웨어 구현은 부동 소수점 연산에 비해서 적은 면적과 빠른 수행시간을 가지는 고정소수점 연산을 많이 사용하지만, 소프트웨어에서는 일반적으로 높은 정확도를 가지는 부동소수점 연산을 사용한다. 신경망의 하드웨어 구현에서 많이 사용하는 고정소수점 연산은 부동소수점 연산에 비해서 빠른 처리속도와 적은 면적으로써 쉽게 하드웨어 구현에 용이하지만, 부동소수점 연산에 비해서 낮은 정확도와 기존의 부동소수점 연산을 사용하는 소프트웨어 신경망을 쉽게 적용할 수 없는 단점을 가진다. 본 논문에서는 부동소수점 연산을 사용하여 문자 추출 MLP의 데이터 변환 없이 적용할 수 있는 전체 파이프라이닝 설계 구조를 제안한다. 제안된 설계방법은 신경망의 전체 구조를 입력층과 은닉층을 링크 병렬화 방법과 은닉층과 출력층을 뉴런 병렬화 방법을 개선하여 쉽게 파이프라이닝 구조로 설계함으로써 신경망 처리는 은닉층 뉴런수와 동일한 주기로 처리되며, 기존의 문자추출 소프트웨어 신경망을 제안된 하드웨어 설계방법으로 구현하였을 때 11배의 빠른 성능을 나타낸다.

• Journal of the Korean Chemical Society
• /
• v.61 no.6
• /
• pp.320-327
• /
• 2017

In a borate buffer solution, the growth kinetics and the electronic properties of passive film on cobalt were investigated, using the potentiodynamic method, chronoamperometry, and single-frequency electrochemical impedance spectroscopy. It was found out that the unstable passive film ($Co(OH)_2$) and CoO of Co formed in the low electrode potential changes to $Co_3O_4$ and CoOOH while the electrode potential increases. And the composition of the passive films was varied against the applied potential and oxidation time. The oxide film formed during the passivation process of cobalt has showed the electronic properties of p-type semiconductor, which follow from the Mott-Schottky equation.

• Journal of the Korean Geotechnical Society
• /
• v.32 no.12
• /
• pp.59-67
• /
• 2016

The strength of frozen soils is affected by size and shape of particles, and the amount of ice and unfrozen water. The objective of this study is to characterize the unfrozen water content and the unconfined compressive strength of the frozen soils according to the degree of saturations and silt fractions. The specimens are mixtures of sand, silt, and water. The silt fractions (SF), which are the ratio of the silt weight ($W_{silt}$) to the sand weight ($W_{sand}$), are 10% and 30%. In addition, the degrees of the saturation are 5%, 10%, 15%, and 20%. The specimens are frozen under the temperature of $-10^{\circ}C$ conditions. The uniaxial compression tests are conducted for 24 hours, 48 hours, and 72 hours after freezing to determine proper freezing time. The freezing time of 24 hours is chosen because the unconfined compressive strengths of specimens after 24 hours freezing times are similar to each other. Furthermore, the unfrozen water content is monitored during freezing using the TDR system. The unfrozen water content increases with the increase of the silt fraction and degree of saturation. The unconfined compressive strength of the frozen soils exponentially increases with increasing the degree of saturation. This study shows that the amount of ice has more influence on the strength of the frozen soils than the amount of unfrozen water.

• Journal of the Korean Chemical Society
• /
• v.26 no.6
• /
• pp.369-377
• /
• 1982

The technique of combined-measurement of reflectance and ellipsometric parameters was used for studying the anodic film formed on nickel surface in basic solutions. An ellipsometer was automated for transient measurements by way of modulating the plane-polarized light with the Faraday effect. Surface film was formed electrochemically by applying a potential step from the reduction potential range to the passivation range on a polished, high-purity, polycrystalline nickel specimen. From that instant, the changes in the reflectance(r) and the ellipsometric parameters(${\Delta},{\Psi}$) of the surface film were recorded by the automatic ellipsometer. Three exact simultaneous equations including these optical signals, ${\Delta},{\Psi}$ and r were solved numerically with a computer in order to determine the optical properties, n, k, and the thickness, ${\tau}$, of the surface film. From the computed results which showed dependence on pH and time, it was found that passivation of nickel can be effectively attained by surface film thinner than $15{\AA}$ and this passivation film has a small optical absorption coefficient. It seemed that a high pH environment enhances the rate of passivation and is favorable for a denser structure of the surface film. The experimental evidence is in accordance with the hypothesis that the composition of the passive film can be approximated by $Ni(OH)_2$ in the early stage of passivation and that as time passes the composition changes partially toward that of NiO through dehydration.

• The KIPS Transactions:PartA
• /
• v.13A no.3 s.100
• /
• pp.185-190
• /
• 2006

JPEG 2000 is used as an alternative to solve the blocking artifact problem with the existing still image compression JPEG algorithm. However, it has shortcomings such as longer floating point computation time and more complexity in the procedure of enhancing the image compression rate and decompression rate. To compensate for these we implemented with hardware the JPEG 2000 algorithm's filter part which requires a lot of floating point computation. This DWT Filter[1] chip is designed on the basis of Daubechies 9/7 filter[6] and is composed of 3-stage pipeline system to optimize the performance and chip size. Our implemented Filter was 7 times faster than software based Filter in the floating point computation.