Stochastic Optimization of the RLC Model of the Package Output in Order to Increase Its Reliability at High Frequencies

The scattering matrix of a multipole (or S- parameters) is widely used to describe electrical circuits in the form of a black box. The application of S -parameters allows us to characterize the circuit in any frequency range. However, the standard circuit simulators do not work with S -parameters. I...

Full description

Saved in:
Bibliographic Details
Published in:Russian microelectronics 2021-12, Vol.50 (7), p.528-533
Main Authors: Belov, E. N., Shvets, A. V.
Format: Article
Language:English
Subjects:
Algorithms
Capacitance
Circuits
Electrical Engineering
Engineering
Frequency ranges
High frequencies
Inductance
Mathematical models
Model accuracy
Multipoles
Optimization
Parameters
S matrix theory
Simulators
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The scattering matrix of a multipole (or S- parameters) is widely used to describe electrical circuits in the form of a black box. The application of S -parameters allows us to characterize the circuit in any frequency range. However, the standard circuit simulators do not work with S -parameters. In addition, the more outputs a circuit has and the larger the frequency range that needs to be characterized the larger the matrix that has to be solved. This paper considers the classical model of the package output based on its own capacitance, inductance, and capacitance of the connection between adjacent outputs, which can be built based on the existing S -parameters. However, as this model cannot take into account the RF and microwave effects, it is inaccurate at high frequencies. The optimization method for a classical model based on a stochastic optimization algorithm is presented in order to achieve greater model accuracy at high frequencies. The use of this optimization makes it possible to increase the accuracy of the model at high frequencies by a factor of more than 3 in comparison with the classical RLC model.
ISSN:1063-7397
1608-3415
DOI:10.1134/S1063739721070039