Modeling the skin effect using efficient conduction mode techniques

• US 8,024,692 B2
• Filed: 05/02/2008
• Issued: 09/20/2011
• Est. Priority Date: 05/04/2007
• Status: Active Grant

• Current Assignee: Siemens Industry Software Limited (Siemens AG)
• Original Assignee: Mentor Graphics Corporation (Siemens AG)
• Inventors: Ortiz, Salvador, Suaya, Roberto
• Primary Examiner(s): LEVIN, NAUM B
• Application Number:
  US12/151,074
• Publication Number:
  US 20080276207A1
• Time in Patent Office: 1,236 Days
• Field of Search: 716/132, 716/136, 703/2, 703/14
• US Class Current: 716/132
• CPC Class Codes: G06F 2119/06 Power analysis or power optimisation
• CPC Class Codes: G06F 30/20 Design optimisation, verification or simulation optimisation, verification or simulation of circuit designs G06F30/30
• CPC Class Codes: G06F 30/367 Design verification, e.g. using simulation, simulation program with integrated circuit emphasis [SPICE], direct methods or relaxation methods

Abstract

Described herein are embodiments of methods for extracting various high frequency parameters for a circuit design. In one exemplary embodiment, circuit design information indicating at least a geometric layout of conductors in the circuit design and a desired frequency of operation for the circuit design is received. Conduction modes representing distribution functions for currents in the conductors at the desired frequency of operation are defined. A conduction mode matrix including matrix elements based on the defined conduction modes is generated. Values for one or more matrix elements are computed by decomposing integrands for calculating the matrix elements into simplified terms that are less computationally intensive than the integrands and computing the values of the simplified terms. The values for the one or more matrix elements can be stored (e.g., on one or more computer-readable media).

Claims

1. One or more computer-readable storage devices storing computer-executable instructions for causing a computer to perform a method, the method comprising:

receiving circuit design information, the circuit design information indicating at least a geometric layout of conductors in a circuit design and a desired frequency of operation for the circuit design;

defining conduction modes representing distribution functions for currents in the conductors at the desired frequency of operation;

generating a conduction mode matrix, the conduction mode matrix including matrix elements based on the defined conduction modes;

computing values for one or more matrix elements by decomposing integrands for calculating the matrix elements into simplified terms that are less computationally intensive than the integrands and computing values of the simplified terms; and

storing the values for the one or more matrix elements.

2. The one or more computer-readable storage devices of claim 1, wherein the method further comprises computing impedance values for each of the conductors by solving a matrix equation that includes the conduction mode matrix.

3. The one or more computer-readable storage devices of claim 1, wherein the act of computing the values for the matrix elements further comprises computing values for matrix elements corresponding to same conductor integrands using a first set of simplified terms and computing values for matrix elements corresponding to different conductor integrands using a second set of simplified terms, the second set of simplified terms being different than the first set of simplified terms.

4. The one or more computer-readable storage devices of claim 1, wherein the act of computing the values for the matrix elements further comprises classifying a selected one of the matrix elements as corresponding to either:

(a) an integrand having an exponential contribution along a first integration axis and an exponential contribution along a second integration axis;

(b) an integrand having a linear contribution along a first integration axis and an exponential contribution along a second integration axis;

or (c) an integrand having a linear contribution along a first integration axis and a linear contribution along a second integration axis.

5. The one or more computer-readable storage devices of claim 4, wherein the selected one of the matrix elements is classified as an integrand having a linear contribution along a first integration axis and an exponential contribution along a second integration axis, and wherein the act of computing values for the matrix elements further comprises applying a first type of simplified term to the selected one or more of the matrix elements to compute the exponential contribution and applying a second type of simplified term to the selected one or more of the matrix elements to compute the linear contribution.

6. The one or more computer-readable storage devices of claim 1, wherein the simplified terms for one or more of the matrix elements include one or more terms based on Gaussian quadrature rules.

7. The one or more computer-readable storage devices of claim 1, wherein the simplified terms for one or more of the matrix elements include a term derived by applying a polar coordinate system to the corresponding integrand of the one or more of the matrix elements.

8. The one or more computer-readable storage devices of claim 1, wherein the simplified terms for one or more of the matrix elements include one or more terms corresponding to Taylor expansions.

9. The one or more computer-readable storage devices of claim 1, wherein the simplified terms for one or more of the matrix elements include one or more terms corresponding to polynomial approximations.

10. The one or more computer-readable storage devices of claim 1, wherein the method further comprises assigning at least some of the computed values for the one or more matrix elements to other matrix elements based on conduction mode symmetries.

11. The one or more computer-readable storage devices of claim 1, wherein the act of defining conduction modes comprises defining one or more additional effective conduction modes for a respective conductor based on whether the width of the respective conductor causes a threshold value to be met.

12. The one or more computer-readable storage devices of claim 11, wherein the threshold value is a ratio between the width of the respective conductor and an amount the respective conductor overlaps with another conductor.

13. The one or more computer-readable storage devices of claim 1, wherein each matrix element corresponds to a respective conduction mode integrand that at least partially represents the distribution of currents within a respective conductor at the desired frequency of operation, and wherein the act of computing values for one or more matrix elements comprises computing a value for a respective one of the matrix elements by computing one or more linear contributions to the respective conduction mode integrand separately from one or more exponential contributions to the respective conduction mode integrand.

14. The one or more computer-readable media of claim 13, wherein the act of computing the value for the respective one of the matrix elements further comprises identifying the respective conduction integrand for the respective one of the matrix elements as either:

(a) an integrand having an exponential contribution along a first integration axis and an exponential contribution along a second integration axis;

(b) an integrand having a linear contribution along a first integration axis and an exponential contribution along a second integration axis;

or (c) an integrand having a linear contribution along a first integration axis and a linear contribution along a second integration axis.

15. The one or more computer-readable media of claim 13, wherein the method further comprises computing the exponential contributions to the respective conduction mode integrand using one or more terms based on Gaussian quadrature rules.

16. The one or more computer-readable media of claim 13, wherein the method further comprises computing the exponential contributions to the respective conduction mode integrand using one or more Taylor expansions.

17. The one or more computer-readable media of claim 13, wherein the method further comprises computing the exponential contributions to the respective conduction mode integrand using one or more polynomial approximations.

18. The one or more computer-readable media of claim 13, wherein the method further comprises computing the linear contributions to the respective conduction mode integrand using one or more terms based on Gaussian quadrature rules or using one or more terms corresponding to a polynomial interpolation.

19. The one or more computer-readable media of claim 13, wherein the method further comprises computing the linear contributions to the respective conduction mode integrand using one or more terms derived by applying a polar coordinate system to the respective conduction mode integrand and using one or more Taylor expansions.

20. A method, comprising:

receiving circuit design information, the circuit design information indicating at least a geometric layout of conductors in a circuit design and a desired frequency of operation for the circuit design;

defining conduction modes representing distribution functions for currents in the conductors at the desired frequency of operation;

generating a conduction mode matrix, the conduction mode matrix including matrix elements based on the defined conduction modes;

with a computer, computing values for one or more matrix elements by decomposing integrands for calculating the matrix elements into simplified terms that are less computationally intensive than the integrands and computing values of the simplified terms; and

storing the values for the one or more matrix elements.

21. The method of claim 20, further comprising:

computing impedance values for each of the conductors based at least in part on the computed values of the simplified terms; and

modifying a circuit design based at least in part on the computed impedance values.

22. The method of claim 21, further comprising manufacturing an integrated circuit using the circuit design.

23. The method of claim 20, further comprising computing impedance values for each of the conductors by solving a matrix equation that includes the conduction mode matrix.

24. The method of claim 20, wherein the act of computing the values for the matrix elements further comprises computing values for matrix elements corresponding to same conductor integrands using a first set of simplified terms and computing values for matrix elements corresponding to different conductor integrands using a second set of simplified terms, the second set of simplified terms being different than the first set of simplified terms.

25. The method of claim 20, wherein the act of computing the values for the matrix elements further comprises classifying a selected one of the matrix elements as corresponding to either:

(a) an integrand having an exponential contribution along a first integration axis and an exponential contribution along a second integration axis;

(b) an integrand having a linear contribution along a first integration axis and an exponential contribution along a second integration axis;

or (c) an integrand having a linear contribution along a first integration axis and a linear contribution along a second integration axis.

26. The method of claim 20, wherein the simplified terms for one or more of the matrix elements include one or more terms based on Gaussian quadrature rules, a term derived by applying a polar coordinate system to the corresponding integrand of the one or more of the matrix elements, one or more terms corresponding to Taylor expansions, or one or more terms corresponding to polynomial approximations.

27. The method of claim 20, wherein the method further comprises assigning at least some of the computed values for the one or more matrix elements to other matrix elements based on conduction mode symmetries.

28. The method of claim 20, wherein each matrix element corresponds to a respective conduction mode integrand that at least partially represents the distribution of currents within a respective conductor at the desired frequency of operation, and wherein the act of computing values for one or more matrix elements comprises computing a value for a respective one of the matrix elements by computing one or more linear contributions to the respective conduction mode integrand separately from one or more exponential contributions to the respective conduction mode integrand.