TIME WEIGHTED MOVING AVERAGE FILTER

• US 20070239285A1
• Filed: 04/06/2006
• Published: 10/11/2007
• Est. Priority Date: 04/06/2006
• Status: Active Grant

• Current Assignee: MediaTek, Inc.
• Original Assignee: GlobalFoundries, Inc.
• Inventors: Good, Richard, Schulze, Uwe, Chamness, Kevin
• Granted Patent: US 7,542,880 B2
• Time in Patent Office: Days
• US Class Current: 700/1
• CPC Class Codes: G05B 13/048 using a predictor
• CPC Class Codes: G05B 19/41865 characterised by job scheduling, process planning, material flow
• CPC Class Codes: G05B 2219/32053 Adjust work parameter as function of other cell
• CPC Class Codes: G05B 2219/37532 Synchronized data acquisition
• CPC Class Codes: G05B 2219/45031 Manufacturing semiconductor wafers
• CPC Class Codes: Y02P 90/02 Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Abstract

A method for estimating a state associated with a process includes receiving a state observation associated with the process. The state observation has an associated process time. A weighting factor to discount the state observation is generated based on the process time. A state estimate is generated based on the discounted state observation. A system includes a process tool, a metrology tool, and a process controller. The process tool is operable to perform a process in accordance with an operating recipe. The metrology tool is operable to generate a state observation associated with the process. The process controller is operable to receive the state observation, the state observation having an associated process time, generate a weighting factor to discount the state observation based on the process time, generate a state estimate based on the discounted state observation, and determine at least one parameter of the operating recipe based on the state estimate.

Claims

1. A method for estimating a state associated with a process, comprising:

receiving a state observation associated with the process, the state observation having an associated process time;

generating a weighting factor to discount the state observation based on the process time; and

generating a state estimate based on the discounted state observation.

2. The method of claim 1, wherein the weighting factor comprises an exponential weighting factor.

3. The method of claim 2, wherein the exponential weighting factor comprises

4. The method of claim 1, wherein generating the state estimate further comprises generating the state estimate using a time-weighted exponentially weighted moving average.

5. The method of claim 1, wherein generating the state estimate further comprises generating the state estimate using a recursive time-weighted exponentially weighted moving average.

6. The method of claim 5, wherein generating the state estimate further comprises:

discounting a weighting factor from a previous state update as a function of the elapsed time since the previous state update; and

generating the updated state estimate as a function of the discounted state observation, the discounted weighting factor, and a previous state estimate.

7. The method of claim 6, further comprising updating the weighting factor based on the discounted weighting factor and the weighting factor used to discount the state observation.

8. The method of claim 1, wherein the state observation comprises a first state observation, the process time comprises a first process time, and the method further comprises:

receiving a second state observation associated with the process, the second state observation having a second associated process time earlier than the first process time;

discounting the second state observation based on the second process time; and

updating the state estimate based on the discounted second state observation.

9. The method of claim 1, further comprising:

receiving a plurality of state observations associated with the process, each state observation having a process time;

discounting each state observation as a function of its associated process time; and

generating the state estimate based on the discounted state observations.

10. The method of claim 9, wherein generating the state estimate further comprises generating the state estimate using the equation

11. The method of claim 9, wherein generating the state estimate further comprises generating the state estimate using the equation

12. The method of claim 9, wherein generating the state estimate further comprises generating an updated state estimate responsive to the receipt of each state observation, and wherein generating each updated state estimate further comprises:

generating a new weighting factor for the updated state estimate as a function of the process time associated with the state observation being incorporated into the updated state estimate;

discounting a weighting factor associated with a previous updated state estimate as a function of the elapsed time since the generation of the previous updated state estimate; and

generating the updated state estimate as a function of the new weighting factor, the state observation being incorporated into the updated state estimate, the discounted weighting factor associated with the previous updated state estimate, and the previous updated state estimate.

13. The method of claim 12, further comprising removing a selected one of the state observations from the state estimate.

14. The method of claim 13, wherein removing the selected state observation further comprises:

generating a removal weighting factor for the selected state observation as a function of the associated process time; and

removing the selected state observation using the selected state observation and the removal weighting factor.

15. The method of claim 14, wherein removing the selected state observation further comprises removing the selected state observation using the equation

16. The method of claim 1, further comprising controlling the process based on the state estimate.

17. The method of claim 12, wherein the process comprises a semiconductor fabrication process, controlling the process further comprises determining at least one parameter of an operating recipe for semiconductor fabrication process, and the method further comprises processing semiconductor devices subsequent to determining the at least one parameter of the operating recipe.

18. A computer implemented method for filtering data, comprising:

receiving a plurality of data elements, each data element having an associated time stamp;

generating a weighting factor to discount each data element based on the associated time stamp; and

generating a data element estimate based on the discounted data elements.

19. The method of claim 18, wherein generating the data element estimate further comprises generating an updated data element estimate responsive to the receipt of each data element, and wherein generating each updated data element estimate further comprises:

generating a new weighting factor for the updated data element estimate as a function of the time stamp associated with the data element being incorporated into the updated data element estimate;

discounting a weighting factor associated with a previous updated data element estimate as a function of the elapsed time since the generation of the previous updated data element estimate; and

generating the updated data element estimate as a function of the new weighting factor, the data element being incorporated into the updated data element estimate, the discounted weighting factor associated with the previous updated data element estimate, and the previous updated data element estimate.

20. A system, comprising:

a process tool operable to perform a process in accordance with an operating recipe;

a metrology tool operable to generate a state observation associated with the process; and

a process controller operable to receive the state observation, the state observation having an associated process time, generate a weighting factor to discount the state observation based on the process time, generate a state estimate based on the discounted state observation, and determine at least one parameter of the operating recipe based on the state estimate.

21. A system comprising:

means for receiving a state observation associated with a process, the state observation having an associated process time;

means for generating a weighting factor to discount the state observation based on the process time; and

means for generating a state estimate based on the discounted state observation.