Method of using control models for data compression

• US 6,789,052 B1
• Filed: 10/24/2000
• Issued: 09/07/2004
• Est. Priority Date: 10/24/2000
• Status: Expired due to Fees

• Current Assignee: GlobalFoundries, Inc.
• Original Assignee: Advanced Micro Devices, Inc.
• Inventors: Toprac, Anthony John
• Primary Examiner(s): PALADINI, ALBERT WILLIAM
• Application Number:
  US09/695,495
• Time in Patent Office: 1,414 Days
• Field of Search: 703/2, 702/85, 700/29, 700/30
• US Class Current: 703/2
• CPC Class Codes: H03M 7/30 Compression speech analysis-synthesis for redundancy reduction G10L19/00; for image communication H04N; Expansion; Suppression of unnecessary data, e.g. redundancy reduction

Abstract

A method is provided for compressing data. The method comprises collecting processing data representative of a process. The method further comprises modeling the collected processing data using a control model. The method also comprises applying the control model to compress the collected processing data.

Claims

1. A method comprising:

collecting processing data representative of a process;

applying a control model to model the collected processing data; and

applying the control model to compress the collected processing data.

2. The method of claim 1, wherein modeling the collected processing data using the control model comprises building the control model.

3. The method of claim 2, wherein building the control model comprises fitting the collected processing data using at least one of polynomial curve fitting, least squares fitting, polynomial least squares fitting, non polynomial least squares fitting, weighted least squares fitting, weighted polynomial least squares fitting, weighted non polynomial least squares fitting, Partial Least Squares (PLS), and Principal Components Analysis (PCA).

4. The method of claim 3 further comprising:

scaling at least a portion of the collected processing data to generate mean values and mean scaled values for the collected processing data;

calculating Scores from the mean scaled values for the collected processing data using at most first, second, third and fourth Principal Components derived from a Principal Components Analysis (PCA) control model using archived data sets; and

saving the Scores and the mean values.

5. The method of claim 4, wherein building the Principal Components Analysis (PCA) control model further comprises saving Loadings corresponding to at most first, second, third and fourth Principal Components derived from a plurality of processing data representative of previous processes.

6. The method of claim 5, wherein saving Loadings corresponding to at most the first, second, third and fourth Principal Components comprises using an eigenanalysis method.

7. The method of claim 5, wherein saving Loadings corresponding to at most the first, second, third and fourth Principal Components comprises using a singular value decomposition method.

8. The method of claim 5, wherein saving Loadings corresponding to at most the first, second, third and fourth Principal Components comprises using a power method.

9. The method of claim 4 further comprising:

reconstructing the data representative of the process using the Scores and the mean values.

10. The method of claim 4, wherein scaling the at least the portion of the collected processing data to generate the mean values and the mean scaled values for the collected processing data comprises generating the mean values and the mean scaled values for the collected processing data from the Principal Components Analysis (PCA) control model built using a plurality of processing data representative of previous processes.

11. A computer readable, program storage device encoded with instructions that, when executed by a computer, perform a method, the method comprising:

collecting processing data representative of a process;

applying a control model to model the collected processing data; and

applying the control model to compress the collected processing data.

12. The computer readable, program storage device of claim 11, wherein modeling the collected processing data using the control model comprises building the control model.

13. The computer readable, program storage device of claim 12, wherein building the control model comprises fitting the collected processing data using at least one of polynomial curve fitting, least squares fitting, polynomial least squares fitting, non polynomial least squares fitting, weighted least squares fitting, weighted polynomial least squares fitting, weighted non polynomial least squares fitting, Partial Least Squares (PLS), and Principal Components Analysis (PCA).

14. The computer readable, program storage device of claim 13 further comprising:

scaling at least a portion of the collected processing data to generate mean values and mean scaled values for the collected processing data;

calculating Scores from the mean scaled values for the collected processing data using at most first, second, third and fourth Principal Components derived from a Principal Components Analysis (PCA) control model using archived data sets; and

saving the Scores and the mean values.

15. The computer readable, program storage device of claim 14, wherein building the Principal Components Analysis (PCA) control model further comprises saving Loadings corresponding to at most first, second, third and fourth Principal Components derived from a plurality of processing data representative of previous processes.

16. The computer readable, program storage device of claim 15, wherein saving Loadings corresponding to at most the first, second, third and fourth Principal Components comprises using an eigenanalysis method.

17. The computer readable, program storage device of claim 15, wherein saving Loadings corresponding to at most the first, second, third and fourth Principal Components comprises using a singular value decomposition method.

18. The computer readable, program storage device of claim 15, wherein saving Loadings corresponding to at most the first, second, third and fourth Principal Components comprises using a power method.

19. The computer readable, program storage device of claim 14 further comprising:

reconstructing the data representative of the process using the Scores and the mean values.

20. The computer readable, program storage device of claim 14, wherein scaling the at least the portion of the collected processing data to generate the mean values and the mean scaled values for the collected processing data comprises generating the mean values and the mean scaled values for the collected processing data from the Principal Components Analysis (PCA) control model built using a plurality of processing data representative of previous processes.

21. A computer programmed to perform a method, the method comprising:

collecting processing data representative of a process;

applying a control model to model the collected processing data; and

applying the control model to compress the collected processing data.

22. The computer of claim 21, wherein modeling the collected processing data using the control model comprises building the control model.

23. The computer of claim 22, wherein building the control model comprises fitting the collected processing data using at least one of polynomial curve fitting, least squares fitting, polynomial least squares fitting, non polynomial least squares fitting, weighted least squares fitting, weighted polynomial least squares fitting, weighted non polynomial least squares fitting, Partial Least Squares (PLS), and Principal Components Analysis (PCA).

24. The computer of claim 23 further comprising:

scaling at least a portion of the collected processing data to generate mean values and mean scaled values for the collected processing data;

calculating Scores from the mean scaled values for the collected processing data using at most first, second, third and fourth Principal Components derived from a Principal Components Analysis (PCA) control model using archived data sets; and

saving the Scores and the mean values.

25. The computer of claim 24, wherein building the Principal Components Analysis (PCA) control model further comprises saving Loadings corresponding to at most first, second, third and fourth Principal Components derived from a plurality of processing data representative of previous processes.

26. The computer of claim 25, wherein saving Loadings corresponding to at most the first, second, third and fourth Principal Components comprises using an eigenanalysis method.

27. The computer of claim 25, wherein saving Loadings corresponding to at most the first, second, third and fourth Principal Components comprises using a singular value decomposition method.

28. The computer of claim 25, wherein saving Loadings corresponding to at most the first, second, third and fourth Principal Components comprises using a power method.

29. The computer of claim 24 further comprising:

reconstructing the data representative of the process using the Scores and the mean values.

30. The computer of claim 24, wherein scaling the at least the portion of the collected processing data to generate the mean values and the mean scaled values for the collected processing data comprises generating the mean values and the mean scaled values for the collected processing data from the Principal Components Analysis (PCA) control model built using a plurality of processing data representative of previous processes.

31. A method comprising:

collecting processing data (xi,yi), where i=1, 2, . . . , M, representative of a process;

applying a control model to model the collected processing data said control model having at most N+1 parameters, wherein 2M is greater than N+1; and

applying the control model to compress the collected processing data to have a data compression ratio of at least about 2M;

(N+1).

32. The method of claim 31, wherein modeling the collected processing data using the control model comprises building the control model.

33. The method of claim 32, wherein building the control model comprises fitting the collected processing data using at least one of polynomial curve fitting, least squares fitting, polynomial least squares fitting, non polynomial least squares fitting, weighted least squares fitting, weighted polynomial least squares fitting, weighted non polynomial least squares fitting, Partial Least Squares (PLS), and Principal Components Analysis (PCA).

34. The method of claim 33 further comprising:

scaling at least a portion of the collected processing data to generate mean values and mean scaled values for the collected processing data;

calculating Scores from the mean scaled values for the collected processing data using at most first, second, third and fourth Principal Components derived from a Principal Components Analysis (PCA) control model using archived data sets; and

saving the Scores and the mean values.

35. The method of claim 34, wherein building the Principal Components Analysis (PCA) control model further comprises saving Loadings corresponding to at most first, second, third and fourth Principal Components derived from a plurality of processing data representative of previous processes.

36. The method of claim 35, wherein saving Loadings corresponding to at most the first, second, third and fourth Principal Components comprises using an eigenanalysis method.

37. The method of claim 35, wherein saving Loadings corresponding to at most the first, second, third and fourth Principal Components comprises using a singular value decomposition method.

38. The method of claim 35, wherein saving Loadings corresponding to at most the first, second, third and fourth Principal Components comprises using a power method.

39. The method of claim 34 further comprising:

reconstructing the processing data representative of the process using the Scores and the mean values.

40. The method of claim 34, wherein scaling the at least the portion of the collected processing data to generate the mean values and the mean scaled values for the collected processing data comprises generating the mean values and the mean scaled values for the collected processing data from the Principal Components Analysis (PCA) control model built using a plurality of processing data representative of previous processes.

41. A system, comprising:

a processing tool to process a semiconductor wafer;

a controller comprising a control model to acquire and model processing data relating to said processing of said semiconductor wafer, said control model to compress said modeled processing data.

42. The system of claim 41, wherein said controller is adapted to build said control model, wherein said controller to perform a fitting of said collected processing data using at least one of polynomial curve fitting, least squares fitting, polynomial least squares fitting, non polynomial least squares fitting, weighted least squares fitting, weighted polynomial least squares fitting, weighted non polynomial least squares fitting, Partial Least Squares (PLS), and Principal Components Analysis (PCA).

43. The system of claim 42, wherein said controller further comprising:

means for scaling at least a portion of the collected processing data to generate mean values and mean scaled values for the collected processing data;

means for calculating Scores from the mean scaled values for the collected processing data using at most first, second, third and fourth Principal Components derived from a Principal Components Analysis (PCA) control model using archived data sets; and

means for saving the Scores and the mean values.

44. The system of claim 43, wherein said means for scaling the at least the portion of the collected processing data to generate the mean values and the mean scaled values for the collected processing data further comprises means for generating the mean values and the mean scaled values for the collected processing data from said Principal Components Analysis (PCA) control model built using a plurality of processing data representative of previous processes.

45. An apparatus, comprising:

a controller comprising a control model to acquire and model processing data relating to a process performed on a semiconductor wafer, said control model to compress said processing data.

46. The apparatus of claim 45, wherein said controller is adapted to build said control model, wherein said controller to perform a fitting of said collected processing data using at least one of polynomial curve fitting, least squares fitting, polynomial least squares fitting, non polynomial least squares fitting, weighted least squares fitting, weighted polynomial least squares fitting, weighted non polynomial least squares fitting, Partial Least Squares (PLS), and Principal Components Analysis (PCA).

47. The apparatus of claim 46, wherein said controller further comprising:

means for scaling at least a portion of the collected processing data to generate mean values and mean scaled values for the collected processing data;

means for calculating Scores from the mean scaled values for the collected processing data using at most first, second, third and fourth Principal Components derived from a Principal Components Analysis (PCA) control model using archived data sets; and

means for saving the Scores and the mean values.

48. The apparatus of claim 46, wherein said means for scaling the at least the portion of the collected processing data to generate the mean values and the mean scaled values for the collected processing data further comprises means for generating the mean values and the mean scaled values for the collected processing data from said Principal Components Analysis (PCA) control model built using a plurality of processing data representative of previous processes.