DocumentCode
2190923
Title
A general method for multi-port active network reduction and realization
Author
Liu, Pu ; Qi, Zhenyu ; Aviles, Amalia ; Tan, Sheldon X D
Author_Institution
Dept. of Electr. Eng., California Univ., Riverside, CA, USA
fYear
2005
fDate
22-23 Sept. 2005
Firstpage
7
Lastpage
12
Abstract
This paper presents a novel compact modeling technique for linear(ized) active analog circuits. The new method is based on a recently proposed general s-domain hierarchical modeling and analysis method. In this work, based on the characteristics of analog circuits, we propose a constrained linear least square based method to optimize the order reduced admittance matrices with respect to both magnitude and phase responses up to the given frequency range. Theoretically we show that the hierarchical reduction does not change the reciprocity property of the linear circuit. For analog active circuits, which typically are non-reciprocal, so do the order-reduced circuit admittance matrices after hierarchical reduction. We propose a novel general multi-port, non-reciprocal network realization method. The resulting modeling algorithm can take in the SPICE netlists of active circuits and generate high-fidelity compact macro models of the active circuits with easily controlled model accuracy and complexity. The experimental results on two low-pass active filters demonstrate effectiveness of the proposed algorithm.
Keywords
SPICE; active networks; analogue circuits; circuit optimisation; electric admittance; least mean squares methods; matrix algebra; multiport networks; SPICE netlists; compact modeling technique; constrained linear least square based method; linear active analog circuits; low-pass active filters; magnitude response; multiport active network reduction; nonreciprocal network realization method; order reduced admittance matrices; phase response; s-domain hierarchical modeling; Active circuits; Active filters; Admittance; Analog circuits; Constraint optimization; Digital circuits; Integrated circuit interconnections; Least squares methods; Noise figure; Radio frequency;
fLanguage
English
Publisher
ieee
Conference_Titel
Behavioral Modeling and Simulation Workshop, 2005. BMAS 2005. Proceedings of the 2005 IEEE International
Print_ISBN
0-7803-9352-X
Type
conf
DOI
10.1109/BMAS.2005.1518179
Filename
1518179
Link To Document