http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
송홍복 東義大學校産業技術開發硏究所 2010 産業技術硏究誌 Vol.24 No.-
This paper suggested an algorithm that uses a multiplier, ' ω bit x ω bit = 2(원문을참조하세요) bit', to process (원문을참조하세요) integer division of 2 ω bit integer N and ω bit integer D. An algorithm suggested of the research, when the divisor D (원문을참조하세요), approximate value of (원문을참조하세요)which satisfies (원문을참조하세요), is defined as over reciprocal number and the dividend N is segmented in small word more than ω -3 bit, and partial quotient is calculated by multiplying over reciprocal number in each segmented word, and quotient of double precision integer division is evaluated with sum of partial quotient. The algorithm suggested in this paper doesn't require addidional correction, because it can calculate correct reciprocal number.
NP 同値類 대표함수를 이용한 AND-EXOR 最少 論理式의 表現
宋洪復,金聖大,金明起 동아대학교 공과대학 부설 한국자원개발연구소 1990 硏究報告 Vol.14 No.1
The paper presents a catalog of AND-EXOR expressions for representative functions of four-variable NP-equivalence classes. Minimality is defined as minimizing first the number of product terms and then the total number of literal in the expression. Also, the properties of minimum expressions are discussed. Using this as a base, we compare minimum expressions of AND-OR type two-variable circuit with minimum expressions of AND-EXOR type two-variable circuit which used algorithm in this paper. As a result if was found that in the case of AND-OR type minimum expressions, number of product terms is under 8, and in the case of AND-EXOR type minimum expressions all functions are formed in which number of product terms is under 6, and generally number of product term is considerably small to realize four variable function toward AND-EXOR type minimum expression circuit.
宋洪復,金聖大,金明起 동아대학교 공과대학 부설 한국자원개발연구소 1990 硏究報告 Vol.14 No.1
An optimization problem of AND-EXOR PLA's with imput decoders can be regarded as a minimization problem of Exclusive-Or Sum-Of-Products expressions (ESOP's) for multiple-valued input two-valued output functions. In this paper, we propose a minimization algorithm for ESOP's. The algorithm is based on an iterative improvement. Five rules are used to replace a pair of products with another one. We minimize many ESOP's for arithmetic circuits. In most cases, ESOP's required fewer products than SOP's to realize same functions.
CoNbZr/Cu/CoNbZr 다층박막의 습식 식각 거동에 관한 연구
윤재홍,김현식,송재성,이영생,민복기 國立 昌原大學校 産業技術硏究所 1997 産技硏論文集 Vol.11 No.-
We manufactured CNZ/Cu/CNZ multi-layers by sputtering methode and formed pattern on the deposited multi-layers. In this study, we fabricated new etchant for the etching to formed pattern by the wet chemical etchant. it was searching for the best condition of etchant. Cu was etched selectively independence in the concentration of basic solution but amorphous CNZ thin film did not. The etchant was achieved by basic solution(17.5 mol%) and HF(20mol%) mixing, which etched CNZ/Cu/CNZ multi-layers in the same. Also, the etchant etched CNZ/Cu/CNZ multi-layers by the three-step. It was shown that the etched cross-section have isotropic structure and excellent etching characteristics.
Polaron Hopping in the K_(2)O · B_(2)O_(3) · V_(2)O_(5) Glasses
차유정,강재필,홍성덕,김복수,김영훈,서용문,송승기 명지대학교 자연과학연구소 2005 자연과학논문집 Vol.24 No.-
K_(2)O · B_(2)O_(3) · V_(2)O_(5) 계 유리의 조성비에 따른 밀도, 굴절율, 전이온도, 천이금속이온비율 등과 직류 전기 성질을 100~450K 온도영역에서 연구하였다. 실험결과를 비정질 반도체에 관한 전기전도에 대한 이론 모델을 이용하여 분석하였다. 고온 영역에서는 직류전도도의 온도 의존성을 Mott의 포논-도움껑충뛰기 전도모델로 분석할 수 있었다. 활성화에너지는 0.28~0.33 eV로 시료내 V_(2)O_(5) 의 양이 증가할수록 감소하는 경향을 보인다. 저온 영역에서 직류 전기전도도는 Mott의 가변거리 껑충뛰기(variable-range hopping)전기전도도와 잘 일치하며, N(E_(F))값은 국지화 상태에서와 같다. The dc conductivity of K_(2)O-B_(2)O_(3)-V_(2)O_(5) glasses has been investigated in the temperature range 100~450K and physical parameters such as density, refractive indices, glass transition temperature and fraction of reduced transition-metal ion, were observed for various compositions. The experimental results are analyzed with reference to theoretical models proposed for electrical conduction in amorphous semiconductors. The analysis shows that at high temperature region the temperature dependence of the dc conductivity is consistent with Mott's model of phonon-assisted hopping of small polarons. The values of activation energy are 0.28~0.33 eV and decrease with increasing V_(2)O_(5) content in the sample. The lowtemperature dc conductivity data can be well fitted to Mott's variable-range hopping conductivity and the values of N(E_(F)) are also reasonable for localized states.
A study on the synthesis method of multi-valued Carry Look-Ahead Adders
Song, Hong Bok 東義大學校 産業技術開發硏究所 1996 産業技術硏究誌 Vol.10 No.-
This paper present a synthesis method of multi-valued carry look-ahead adders(MCLA) and the circuit design. In multi-valued operations. parallel additions and operations without carry propagation can be using the redundant number system. That is, the conversion of a redundant number to a standard number can be done by using some functional blocks such as threshold decoders. AND and OR gates, arithmetic adders and decoders. Carries occurs, however, if the resulting number should be represented as the standard R-radix number. We have shown the circuit configuration of a MCLA with CMOS current mode circuits.
다치 논리 함수의 ESOP 최소화 알고리즘에 관한 연구
송홍복(Song Hong Bok) 한국정보처리학회 1997 정보처리학회논문지 Vol.4 No.7
This paper presents an algorithm simplifying the ESOP function by several rules. The algorithm is repeatedly performing operations based on the state of each terms by the product transformation operation of two functions and thus it is simplifying the ESOP function through the reduction of the product terms. Through the minimization of the product terms of the multi-valued input binary multi-output function, an optimization of the input has been done using EXOR PLA with input decoder. The algorithm when applied to four valued arithmetic circuit has been used for a EXOR logic circuit design and the the two bits input decoder has been used for a EXOR-PLA design. It has been found from a computer simulation(IBM PC486) that the suggested algorithm can reduce the product terms of the output function remarkably regardless of the number of input variables when the variable AND-EXOR PLA is applied to the poperation circuit.