http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
소자 파라메터 자동 추출 시스템을 이용한 LDD MOSFET의 채널유효길이와 직렬저항의 추출
강유선,김충원 조선대학교 동력자원연구소 1997 動力資源硏究所誌 Vol.19 No.1
In this paper, we present a new algorithm that decide the effective channel length and serial resistance which are not independent from the LDD MOSFET gate voltage. By using the algorithm, automatic device parameter extraction system has been developed under the IBM-PC environment. The system is connected to a PC through the interface board, and can modulate the voltages of LDD MOSFET's drain and gate. The modulated voltage can be saved as a file when it is changed. Overall developed programs have been written with C-Languag
개선된 신경망 학습 알고리즘을 이용한 한국어 인식에 관한 연구
윤석주,강유선,김충원 조선대학교 동력자원연구소 1994 動力資源硏究所誌 Vol.16 No.1
In this paper, we presented an improved error backpropagation algorithm. The algorithm has designed to selectively releam the pattern that generated the largest error. An artificial neural network with our algorithm was applied to recognition of Korean monophthongs, /a/, /e/, /i/, /o/, /u/, speken by a man. The artificial neural network is implemented with one hidden layer and the LPC Cepstrum coefficients of 12 orders as feature parameters. The result shows that the presented algorithm increases the efficiency of iterations of the learning algorithm significantly about 67.5% and the ratio of voice recognition in 92%.