DocumentCode
3469892
Title
A Novel Modeling Methodology for Silicon-based RF Components
Author
Liu, Tao ; Zhang, Wenjun ; Yu, Zhiping
Author_Institution
Inst. of Microelectron., Tsinghua Univ., Beijing
fYear
2006
fDate
Oct. 2006
Firstpage
1392
Lastpage
1394
Abstract
A novel comprehensive and automatic wide-band artificial neural network (ANN)-based modeling methodology for radio-frequency (RF) components is presented. The methodology is applied to differential spiral inductors with various geometrical sizes. The established model is tested with 653 inductors and shows excellent accuracy over the frequency range of 1-20 GHz
Keywords
inductors; integrated circuit modelling; neural nets; radiofrequency integrated circuits; silicon; 1 to 20 GHz; Si; artificial neural network based modeling; differential spiral inductors; silicon-based radio-frequency components; Artificial neural networks; Circuit simulation; Computational modeling; Data preprocessing; Equivalent circuits; Inductors; Radio frequency; Sampling methods; Scattering parameters; Spirals;
fLanguage
English
Publisher
ieee
Conference_Titel
Solid-State and Integrated Circuit Technology, 2006. ICSICT '06. 8th International Conference on
Conference_Location
Shanghai
Print_ISBN
1-4244-0160-7
Electronic_ISBN
1-4244-0161-5
Type
conf
DOI
10.1109/ICSICT.2006.306193
Filename
4098419
Link To Document