Method and apparatus for using parameters to simulate an electronic circuit

• US 5,946,482 A
• Filed: 05/16/1997
• Issued: 08/31/1999
• Est. Priority Date: 05/16/1997
• Status: Expired due to Term

• Current Assignee: Keysight Technologies, Inc.
• Original Assignee: Hewlett-Packard Company (HP Inc.)
• Inventors: Troyanovsky, Boris, Barford, Lee A., Chang, Norman H.
• Primary Examiner(s): Teska, Kevin
• Assistant Examiner(s): Fiul, Dan
• Application Number:
  US08/857,427
• Time in Patent Office: 837 Days
• Field of Search: 364/578, 364/488-491, 324/754, 324/615, 330/149, 706/31, 395/500.34, 395/500.35, 395/500.36, 702/57, 702/66, 702/75, 702/76, 702/117-126
• US Class Current: 703/14
• 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

The present disclosure provides a method and apparatus for using frequency domain data, such as S-parameters, in a time-based simulator. S-parameters are either input to the simulator, or are empirically measured, at selected frequencies. Preferably, the selected frequencies are related to one another by a logarithmic scale, providing for determination of a system transfer function which is accurate across a very wide range of frequencies, from near zero hertz, to frequencies on the order of a hundred gigahertz. The transfer function preferably takes the form of a fitted polynomial, obtained using FDSI techniques. In addition, recursive convolution may be employed to operate in the time domain on inverse Laplace Transforms of the fitted transfer function and time-domain simulator test signals. This disclosure provides for circuit modeling and simulation which is accurate across a wide frequency range, which is stable for transfer functions of high order, and which is quickly and efficiently performed for large circuits.

Claims

1. A method of using parameters to simulate an electronic circuit, the electronic circuit having a plurality of measurement points, said method utilizing a digital processor and comprising:

selecting a plurality of test frequencies with which to measure frequency response of the electronic circuit at each of the measurement points;

extracting the parameters from the circuit which describe frequency response at the measurement points, for each of the test frequencies;

using the digital processor to, for each measurement point, analyze the parameters and fit at least one transfer function to the parameters to describe behavior the parameters using Frequency Domain System Identification, to thereby obtain at least one fitted transfer function; and

simulating the circuit using a fitted transfer functions.

2. A method according to claim 1, wherein using the digital processor to analyze parameters and fit a transfer function includes:

using the digital processor to detect whether the parameters vary sinusoidally in response to frequency; and

,

if the parameters vary sinusoidally in response to frequency, fitting a sum of complex exponentials to the parameters, to thereby obtain at least one fitted transfer function which is a sum of complex exponentials.

3. A method according to claim 1, wherein using the digital processor to analyze parameters and fit a transfer function includes fitting a rational polynomial to the parameters using Frequency Domain System Identification, to thereby obtain a fitted polynomial.

4. A method according to claim 3, wherein using the digital processor includes utilizing code stored on machine readible media, the code controlling the processor to analyze the parameters and fit at least one rational polynomial.

5. A method according to claim 3, wherein simulating the circuit includes adding in the effects of additional electronic elements, and simulating environmental performance of the circuit based both on a fitted polynomial and the additional electronic elements.

6. A method according to claim 3, wherein simulating the circuit includes:

obtaining inverse transformations of a fitted polynomial to obtain a time domain relation; and

using the digital processor to apply a recursive convolution between the time domain relation and simulator input signals to thereby simulate the circuit.

7. A method according to claim 6, wherein applying recursive convolution includes using a digital processor and code stored on machine readible media, the code causing the processor to perform the recursive convolution.

8. A method according to claim 6, further comprising a step for performing recursive convolution by multiplication with an exponential to eliminate time costs associated with integration over a large range.

9. A method according to claim 6, wherein applying recursive convolution includes using an application specific processor chip to apply recursive convolution.

10. A method according to claim 6, wherein applying recursive convolution includes:

obtaining an inverse Laplace transform of a fitted polynomials to obtain a time domain relation.

11. A method according to claim 1, wherein said method is applied to an actual circuit and extracting parameters includes:

discretely providing, as an input to the actual circuit, each one of the plurality of test frequencies, and electronically measuring response to the inputs, multiple times for each one of the plurality of test frequencies;

for each test frequency, averaging the corresponding responses at each of the measurement points to yield averaged responses; and

computing statistically averaged parameters from the averaged responses.

12. A method according to claim 1, wherein extracting parameters includes:

extracting scattering parameters corresponding to each of the measurement points, for each test frequency.

13. A method according to claim 1, wherein extracting parameters includes:

measuring response at each test frequency several times utilizing an electronic parameter extraction device which provides discrete test frequency inputs to an actual circuit, and automatically measures responses at each of the measurement points.

14. A method according to claim 1, wherein selecting test frequencies includes manually controlling a simulator to input desired frequencies into a physical circuit.

15. A method according to claim 1, wherein selecting test frequencies includes selecting test frequencies which are related to one another by substantially logarithmic scaling.

16. A method according to claim 15, wherein selecting test frequencies related to one another by substantially logarithmic scaling includes selecting at least four frequencies per decade.

17. A method according to claim 1, wherein selecting test frequencies includes selecting one of the frequencies to be no less than twice the highest expected frequency to be encountered any measurement point in circuit operation.

18. A method according to claim 17, wherein selecting no less than twice the highest expected frequency includes selecting a test frequency which is at least two decades above the cutoff frequency for the highest passband of the circuit.

19. An apparatus comprising code for controlling a machine to simulate a circuit based on parameters, and machine-readable media on which the code is stored, the parameters representing frequency-dependent response at each of a plurality of measurement points, said code directing a machine to:

fit at least one transfer function to the parameters, the transfer function representing behavior at the corresponding measurement point in response to frequency, to thereby obtain at least one fitted transfer function; and

simulate the circuit using a fitted transfer function;

wherein the code directs the machine to fit a transfer function using Frequency Domain System Identification.

20. An apparatus according to claim 19, wherein the code directs the machine to detect whether the parameters vary sinusoidally with respect to frequency and, if so, to fit a sum of complex exponentials to the parameters to thereby obtain a fitted transfer function.

21. An apparatus according to claim 19, wherein the code directs the machine to fit a rational polynomial to the parameters to thereby obtain a fitted polynomial, the fitted polynomial representing behavior in response to frequency.

22. An apparatus according to claim 21, wherein the code further directs the machine to simulate the circuit by:

taking an inverse Laplace Transform of a fitted polynomial; and

applying recursive convolution between the inverse Laplace Transform and time-based input signals.

23. An apparatus according to claim 22, further comprising means for performing recursive convolution by multiplication with an exponential to eliminate time costs associated with integration over a large range.

24. An apparatus according to claim 19, wherein the code directs the machine to fit a transfer function using parameters representing responses to unevenly spaced test frequencies.