Interpolation techniques used for time alignment of multiple simulation models

Patent metadata

Field	Value
Publication number	US-9449122-B2
Application number	US-201314026222-A
Country	US
Kind code	B2
Filing date	Sep 13, 2013
Priority date	Jan 25, 2013
Publication date	Sep 20, 2016
Grant date	Sep 20, 2016

Abstract

A system for performing time conversions that includes a processor configured to generate a set of target time-series data from a set of source time-series data and a memory containing specification information relating to the set of target time-series data and also containing information relating to the set of source time-series data. The source time-series specification and the target time-series specification include time intervals between data values. The system also includes a time alignment algorithm used by the processor for converting the set of source time-series data to the set of target time-series data. The converting includes calculation of a set of cubic-spline interpolation constants and the cubic-spline constants.

First claim

What is claimed is:

1. A conversion system, comprising:
a processor configured to generate a set of target time-series data from a set of source time-series data;
a memory containing specification information relating to said set of target time-series data and also containing information relating to said set of source time-series data, said source time-series specification and said target time-series specification including time intervals between data values; and
a time alignment algorithm used by said processor for converting said set of source time-series data to said set of target time-series data, wherein said converting includes calculation of a set of cubic-spline interpolation constants,
said cubic-spline constants being calculated via a plurality of linear equations having a diagonal structure.

2. The system of claim 1 , wherein the processor includes a plurality of processors each configured to retrieve and store data in a plurality of memory devices.

3. The system of claim 1 , wherein said linear equations are calculated based on a set of cubic-spline interpolation constants.

4. The system of claim 3 , wherein said linear equations have a unique solution for calculating said cubic spline interpolation constant.

5. A time conversion system, comprising:
a computer having a plurality of processors configured to receive a specification request for generating a set of target time-series data from a set of source time-series data,
said computer configured for:
obtaining specification information relating to said set of target time-series data;
obtaining information relating to said set of source time-series data, said specifications for said source time-series data and said target time series data including time intervals between data values; and
converting said set of source time-series data to said set of target time-series data, wherein said converting includes calculating a set of cubic-spline interpolation constants by said plurality of processors.

6. The system of claim 5 , wherein data are retrieved and stored in a plurality of memory devices and portions of said data are stored in different processors.

7. The system of claim 6 , wherein said calculating is performed by said plurality of processors by assigning a portion of data to a subgroup of said processors such that said data portion is divided in a plurality of strata among different processors of said plurality of processors.

8. The system of claim 7 , wherein said calculating is performed based on a set of cubic-spline interpolation constants made based on a plurality of equations having a diagonal structure.

9. The system of claim 8 , wherein each processor of said plurality of processors is only assigned one stratum and said processors process data in said strata simultaneously.

10. The system of claim 9 , wherein said strata are overlapping to provide data flow between processors.

11. The system of claim 10 , wherein said processing of data continues until there are either no more data to process or a data stopping condition is encountered.

12. The system of claim 11 , wherein data are retrieved and stored in a memory that is in processing communication with said plurality of processors.

13. The system of claim 12 , wherein data are retrieved and stored in a plurality of memory devices and portions of said data are stored in different processors.

14. The system of claim 13 , wherein said calculating a set of cubic-spline interpolation constants is made based on a plurality of linear equations.

15. The system of claim 14 , wherein said linear equations have a diagonal structure.

16. The system of claim 15 , wherein said converting employs an algorithm that uses a plurality of linear equations to calculate said cubic-spline interpolation.

17. The system of claim 16 , wherein said converting employs an algorithm that uses a plurality of linear equations having a unique solution for calculating a cubic spline interpolation constant.

Assignees

• Globalfoundries Inc
• Globalfoundries Inc

Inventors

• Haas Peter J
• Sismanis John

Classifications

• G06F30/23

  using finite element methods [FEM] or finite difference methods [FDM] · CPC title

• G06F2119/12

  Timing analysis or timing optimisation · CPC title

• G06F30/20 Primary

  Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30) · CPC title

• G06F2217/84

  Physics · mapped topic

• G06F17/5009 Primary

  Physics · mapped topic