Data processing: structural design – modeling – simulation – and em – Simulating electronic device or electrical system – Circuit simulation
Reexamination Certificate
1999-10-19
2004-07-27
Jones, Hugh (Department: 2123)
Data processing: structural design, modeling, simulation, and em
Simulating electronic device or electrical system
Circuit simulation
C703S002000, C703S003000, C703S004000
Reexamination Certificate
active
06768976
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to an apparatus and a method for simulating an electric current flowing through an electronic appliances by solving simultaneous linear equations defined depending on an analytic frequency, and to a program storage medium storing a program for realizing the simulation apparatus, and more specifically to a simulation apparatus and method for quickly realizing the simulating process, and a program storage medium storing a program used to realizing the simulation apparatus.
There is a social rule that excess electric waves or noise higher than a predetermined level should not be discharged. Actually, they have been strictly regulated according to individual rules in each country in the world.
To satisfy such regulations on electric waves, various countermeasures such as shielding technology, filtering technology, etc. are used. However, it is necessary to develop an appropriate simulating technology to quantitatively figure out to what extent electric waves can be attenuated by each of the technologies.
With the above described background, the Applicant of the present invention has disclosed the invention of a simulating technology to compute the intensity of an electromagnetic field discharged from an electronic appliances in the moment method. To put the simulating technology for practical use, a technology of performing a high-speed simulating process should be established.
2. Description of the Related Art
A method of simulating electromagnetic field analysis can be the difference method, the finite element method, the moment method, etc.
Among them, the moment method only has to set the boundary plane of an analysis target discrete in a 2-dimensional array, and is expected as a method more practical than the difference method and the finite element method which require setting at maximum 4-dimensional time space discrete. A reference document is ‘H. N. Wang, J. H. Richmond and M. C. Gilreath: Sinusoidal reaction formulation for radiation and scattering from conducting surface IEEE TRANSACTIONS ANTENNAS PROPAGATION vol. AP-23 1975’.
In these simulating methods, electric currents flowing through each element of an electronic appliance can be simulated by solving simultaneous linear equations defined in a frequency domain. Thus, when the simulation is performed in a time domain in the simulation method, it is necessary to solve simultaneous linear equations on a number of frequencies for the same target as analysis targets. In addition, when there is a wave source having a number of frequencies, it is also necessary to solve simultaneous linear equations on a number of frequencies for the same target as analysis targets.
To solve the simultaneous linear equations at a high speed, the FFS method (Fast Frequency Stepping method) has been used in some cases as described by ‘G. Hoyler, R. Unbehauen, An Efficient Algorithm for The Treatment of Multiple Frequencies with The Method of Moments, Proceedings of EMC’ 96 ROMA pp. 368-371 (1996). In the FFS method, simultaneous linear equations are first derived using the lowest frequency as an analysis target. Then, the Cholesky factorization (A=CC
t
) is performed on the coefficient matrix of the simultaneous linear equations, and the simultaneous linear equations are solved in the direct method. Then, an analytic frequency is selected in the ascending order of frequencies, and simultaneous linear equations are derived using the analytic frequency as an analysis target. Then, the simultaneous linear equations are solved in the iterative method using a preconditioned matrix (C) obtained in the first direct method. This is the FFS method.
Described below is a numeric solution of simultaneous linear equations. A method of solving the simultaneous linear equations ‘Ax=b’ (coefficient matrix A=(a
ij
) is a complex symmetric matrix) as shown in
FIG. 1
can be the direct method or the iterative method.
The direct method can be followed by performing the Cholesky factorization on the coefficient matrix A (LU factorization on a symmetric matrix) as shown in
FIG. 2
, and a solution (x
i
) is obtained by the equation shown in FIG.
3
. The number of required computations is the order of O(n
3
)(n is the degree of the coefficient matrix A).
On the other hand, the iterative method referred to as a conjugate gradient method is described below. In the iterative method, a solution (x
i
) is obtained by obtaining the value ‘x(k+1)’ at the (k+1)th stage using the values ‘x(k), &agr;(k), p(k)’ at the k-th stage according to the algorithm shown in FIG.
4
. The number of required computations is the order of O(Kn
2
)(n is the degree of the coefficient matrix A, and K is the number occurrences having the maximum value of n). Therefore, it has the advantage over the direct method in quantity of computations.
In the iterative method, when the coefficient matrix A is A≈CC&tgr;, which is close to the Cholesky factorization, the original simultaneous linear equations ‘Ax=b’ are transformed as follows.
C
−1
AC
t*−1
C
t
x
=C
−1
b
where C
−1
is an inverse matrix of the matrix C,
C
t
is a transposed matrix of the matrix C, and
C
t*−1
is an inverse matrix of the transposed matrix of the matrix C.
At this time, the matrix C
−1
AC
t*−1
is very close to a unit matrix, and is expected to quickly converge. This is referred to as a preconditioned conjugate gradient method, and is performed by the algorithm shown in FIG.
5
.
When simultaneous linear equations are solved on a number of frequencies for the same target as analysis targets using the above described algorithms in the conventional methods, an FFS method has been used in which simultaneous linear equations are first derived using the lowest frequency as an analysis target, and the coefficient matrix of the simultaneous linear equations is processed in the Cholesky factorization (A=CC
t
), the simultaneous linear equations are solved in the direct method, then the analytic frequency is selected in the ascending order of frequencies, simultaneous linear equations are derived using the analytic frequency as an analysis target, and the simultaneous linear equations are solved in the iterative method using the preconditioned matrix (C).
However, according to the above described conventional technology, as the analytic frequency selected in the ascending order becomes farther from the lowest frequency, the preconditioned matrix deviates from a desired mode. As a result, the analysis time in the iterative method is prolonged, thereby arising the problem that a high-speed simulating process cannot be performed.
That is, as the analytic frequency selected in the ascending order becomes farther from the lowest frequency, the matrix C
−1
AC
t*−1
deviates from a unit matrix. As a result, the analysis time in the iterative method is prolonged, thereby arising the problem that a high-speed simulating process cannot be performed.
SUMMARY OF THE INVENTION
The present invention has been developed based on the above described background, and aims at providing a new simulating apparatus and method for quickly realizing a simulating process when an electric current flowing through electronic appliances is simulated by solving simultaneous linear equations defined depending on an analytic frequency in the moment method, etc., and at providing a new program storage medium storing a program for realizing the simulation apparatus.
The present invention is based on an apparatus and a method for simulating an electric current flowing through electronic appliances by solving simultaneous linear equations defined depending on an analytic frequency, or a computer-readable storage medium storing a program for directing a computer to perform the simulation.
According to the first embodiment of the present invention, the following configuration is designed.
First, an analytic frequency is selected from the frequency area to which the analytic freque
Homma Katsumi
Mukai Makoto
Tanaka Yoshirou
Craig Dwin M.
Jones Hugh
Staas & Halsey , LLP
LandOfFree
Apparatus and method for simulating electric current in... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Apparatus and method for simulating electric current in..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Apparatus and method for simulating electric current in... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3222775