Extended kalman filter incorporating offline metrology

Abstract

An algorithm uses offline metrology to control a process by passing information from an outer control loop to an inner control loop, extended Kalman filter estimator. The inner control loop operates online, and the outer control loop operates asynchronously with respect to the inner control loop. The online control loop is updated for each subsequent process. The offline metrology is optionally updated for each subsequent process.

1 . A method comprising: 
 estimating a state variable for a physical process by use of an extended Kalman Filter (EKF) that operates in an online control loop;    applying a control statement to the physical process from the online control loop; and    sending an information statement from an offline metrology control loop to the online control loop, wherein the information statement optionally changes estimating the state variable.    
   
   
       2 . The method according to  claim 1 , wherein the offline metrology control loop includes an EKF engine.  
   
   
       3 . The method according to  claim 1 , wherein the online control loop and the offline metrology control loop are operated asynchronously.  
   
   
       4 . The method according to  claim 1 , wherein the online control loop generates at least two control cycles before receipt of the information statement from the offline metrology control loop.  
   
   
       5 . A method of planarizing a substrate comprising: 
 in an inner loop: 
 estimating state variables of a substrate planarizing cycle, wherein the estimating the state variables includes use of an extended Kalman filter (EKF) engine; and  
 estimating the endpoint of the substrate planarizing cycle based upon the estimating of the state variables; and  
   in an outer loop: 
 asynchronously comparing at least one of a state variable included in the state variables and an endpoint to an offline metrology database; and  
 sending an information statement from the outer loop to the inner loop.  
   
   
   
       6 . The method according to  claim 5 , wherein estimating the state variables includes: 
 providing an initial estimate of the state variables, wherein the state variables include at least one of the depth of thickness of a substrate layer, an erosion rate of the substrate layer, and an endpoint thickness of the substrate layer.    
   
   
       7 . The method according to  claim 5 , wherein estimating the state variables includes: 
 calculating total reflectance of a substrate that includes an array region and a periphery region;    calculating an instantaneous change in reflectance of the substrate relative to a depth being planarized;    measuring actual reflectance;    updating the state variables by an EKF estimate; and    estimating whether the actual reflectance is within a given variance of measured reflectance.    
   
   
       8 . The method according to  claim 7 , wherein estimating whether the actual reflectance is within a given variance includes: 
 if the estimated reflectance is not within the given variance: 
 providing a new estimate of the state variables; and  
 recalculating the total reflectance of the substrate;  
   and if the estimated reflectance is within the given variance: 
 calculating an endpoint time;  
 determining whether the elapsed physical process time is equal to a most recently calculated endpoint time by making a query;  
 and if the query receives an affirmative response: 
 terminating the CMP process;  
 
   and if the query receives a negative response: 
 providing a new estimate of the state variables; and  
 recalculating the total reflectance of the substrate.  
   
   
   
       9 . The method according to  claim 7 , wherein estimating whether the reflectance is within a given variance includes: 
 if the estimated reflectance is not within the given variance: 
 providing a new estimate of the state variables; and  
 recalculating the total reflectance of the substrate;  
   and if the estimated reflectance is within the acceptable variance: 
 estimating the planarizing status of the substrate; and  
 altering at least one process parameter of the substrate planarizing cycle.  
   
   
   
       10 . The method according to  claim 5 , wherein estimating the state variables of the substrate planarizing cycle includes providing an initial estimate of the state variables, wherein the state variables include at least one of a depth of thickness of a substrate layer, an erosion rate of the substrate layer, and an endpoint thickness.  
   
   
       11 . A method of controlling a process, comprising: 
 estimating a state variable for a physical process by use of an extended Kalman Filter (EKF) that operates in an online control loop;    applying a control statement to the physical process from the online control loop; and    sending an information statement from an offline metrology control loop to the online control loop, wherein the information statement optionally changes estimating the state variable, wherein the sending includes initiating an interrupt of the online control loop.    
   
   
       12 . The method according to  claim 11 , wherein initiating the interrupt includes holding the information statement in a buffer until the online control loop is at a decision node.  
   
   
       13 . The method according to  claim 11 , wherein initiating the interrupt includes holding the information statement in a buffer until the online control loop has finished a cycle.  
   
   
       14 . The method according to  claim 11 , wherein estimating the state variable for the physical process includes: 
 estimating state variables of a substrate planarizing cycle, wherein estimating the state variables includes the use of the EKF; and    estimating the endpoint of the substrate planarizing cycle based upon estimating the state variables.    
   
   
       15 . The method according to  claim 14 , wherein the offline metrology control loop is operated to include: 
 asynchronously comparing at least one of a state variable included in the state variables and an endpoint to an offline metrology database; and    sending an information statement from the offline metrology control loop to the online control loop.    
   
   
       16 . A process of fabricating and assembling a microelectronic system comprising: 
 processing a substrate with a two-control loop control system, wherein the two-control loop control system includes an online control loop and an offline metrology control loop; and    assembling at least a portion of the substrate into an integrated microelectronic device.    
   
   
       17 . The process of fabricating according to  claim 16 , wherein the online control loop includes an extended Kalman filter estimator.  
   
   
       18 . The process of fabricating according to  claim 16 , wherein the offline metrology control loop includes an extended Kalman filter.  
   
   
       19 . The process of fabricating according to  claim 16 , wherein the online control loop and the offline metrology control loop operate asynchronously.  
   
   
       20 . The process of fabricating according to  claim 16 , wherein assembling the microelectronic device includes assembling a circuit module.  
   
   
       21 . The process of fabricating according to  claim 16 , wherein assembling the microelectronic device includes assembling a memory module.  
   
   
       22 . The process of fabricating according to  claim 16 , wherein assembling the microelectronic device includes assembling a memory system.  
   
   
       23 . The process of fabricating according to  claim 16 , wherein assembling the microelectronic device includes assembling a computer system.  
   
   
       24 . A method of calculating a present condition, comprising: 
 estimating a state variable for a physical process by use of an extended Kalman Filter (EKF) that operates in an online control loop;    applying a control statement to the physical process from the online control loop; and    sending an information statement from an offline metrology control loop to the online control loop, wherein the information statement is used in estimating the state variable for a next iteration of the online control loop.    
   
   
       25 . The method according to  claim 24 , wherein the offline metrology control loop includes an EKF engine.  
   
   
       26 . The method according to  claim 24 , wherein the online control loop and the offline metrology control loop are operated asynchronously.  
   
   
       27 . The method according to  claim 24 , wherein the online control loop generates at least two control cycles before receipt of the information statement from the offline metrology control loop.  
   
   
       28 . The method according to  claim 24 , wherein estimating the state variable for the next iteration of the online control loop includes calculating a process parameter refinement.  
   
   
       29 . The method according to  claim 28 , wherein calculating the process parameter refinement includes applying a process parameter improvement to a physical process.  
   
   
       30 . The method according to  claim 29 , wherein the physical process includes an integrated microelectronic device process.  
   
   
       31 . The method according to  claim 29 , wherein the physical process includes a chemical mechanical polishing process.