Method for multi-target vector fitting

Abstract

The invention discloses a method for multi-target vector fitting, which comprises the following steps: 1) processing a given data set to construct a single-target fitting data set; 2) determining an initial pole, and establishing a basic matrix by using the initial pole and the frequency sampling points; 3) separating a real part and an imaginary part of the basic matrix to construct a real number basic matrix; 4) carrying out simplified QR decomposition on the real number basic matrix to obtain the first M columns of a complete Q matrix, recording the first M columns as a Qs matrix, and storing the Qs matrix; 5) constructing and solving a pole coefficient solving equation to obtain a pole coefficient; 6) constructing a pole solution characteristic equation by using the pole coefficient and the initial pole, and calculating to obtain a new pole; and 7) calculating each matrix element residue corresponding to the new pole to complete vector fitting. According to the method for multi-target vector fitting, the complexity of multi-target vector fitting calculation is greatly reduced, and the calculation efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of passive device modeling, in particular to a method for multi-objective vector fitting. Background technique [0002] Vector fitting is widely used in the modeling process of passive devices. Especially in the modeling of transmission lines, people use the W-element or S parameter method to give the physical description of the transmission line. When the signal passes through the transmission line, in order to correctly calculate the output signal, it is usually necessary to use vector fitting to obtain the transmission line in the frequency domain. impulse response. [0003] In the process of realizing vector fitting, the user needs to provide the data of the frequency spectrum at discrete points, and the user provides the data set (ω i ,Y i ), i=1,…,N, where, ω i is the i-th frequency point, Y i is ω i The value of the corresponding physical quantity, which can be an S parameter, a Y parameter, etc...

AI Technical Summary

Problems solved by technology

However, since all matrix elements use the same pole, the fitting accuracy is not as good as the local fitting method, but in many problems, its accuracy can meet the needs of the problem

In addition, in the traditional method, the calculation of the global fitting method needs to solve The system of linear equations for a dense matrix has a time complexity of For the local fitting method, its time complexity is The ratio of the two is about If M>>1, The ratio of the two can be simplified to Obviously, when N p When it is large, the calculation time of the global fitting method is greatly increased

Images