Method of isolating sources of variance in parametric data
A method of isolating sources of variance in parametric data includes steps of: (a) cleaning a data set of measurements for a plurality of parameters; (b) generating a principal component analysis basis from the cleaned data set; (c) estimating an independent component analysis model from the principal component analysis basis; (d) calculating percentages of variance for the plurality of parameters explained by each component in the estimated independent component analysis model; (e) if the calculated percentages of variance indicate that a component is a minor component, then transferring control to step (f), else transferring control to step (g); (f) removing the minor component from the principal component analysis basis and transferring control to step (c); and (g) generating as output the estimated independent component analysis model wherein no component of the independent component analysis model is a minor component.
1. A method of isolating sources of variance in parametric data comprising steps of:
receiving a data set of measurements for a plurality of physical parameters;
cleaning the data set to remove measurements that may introduce error;
generating a principal component analysis basis from the cleaned data set;
estimating an independent component analysis model from the principal component analysis basis;
calculating percentages of variance for the plurality of parameters explained by each component in the estimated independent component analysis model;
removing a component from the principal component analysis basis when the calculated percentages of variance indicate that the component is a minor component; and
generating as output the estimated independent component analysis model excluding the minor component; and
identifying a physical mechanism corresponding to a significant component of the estimated independent component analysis model.
2. The method of claim 1 , the calculated percentages of variance indicating that a component is a minor component when a percentage of variance for each of the plurality of parameters explained by the component is less than a minimum percentage of variance for a single parameter.
3. The method of claim 1 , the calculated percentages of variance indicating that a component is a minor component when an average percentage of variance for the plurality of parameters explained by the component is less than a minimum average percentage of variance.
4. The method of claim 1 further comprising a step of calculating confidence intervals for rotation angles of the estimated independent component analysis model.
5. The method of claim 4 further comprising a step of labeling a component of the estimated independent component analysis model to correspond to a specific physical process.
6. The method of claim 5 further comprising a step of labeling the component as critical dimension of gate width effect on device drive current, critical dimension of gate width effect on device threshold voltage, critical dimension of gate length effect, implant dose effect, and gate oxide thickness effect.
7. The method of claim 5 further comprising a step of examining the labeled component to identify the corresponding physical mechanism associated with the specific physical process.
8. A computer readable storage medium tangibly embodying instructions for a computer that when executed by the computer implement a method for isolating sources of variance in parametric data, the method comprising steps of:
receiving a data set of measurements for a plurality of physical parameters;
cleaning the data set to remove measurements that may introduce error;
generating a principal component analysis basis from the cleaned data set;
estimating an independent component analysis model from the principal component analysis basis;
calculating percentages of variance for the plurality of parameters explained by each component in the estimated independent component analysis model;
removing a component from the principal component analysis basis when the calculated percentages of variance indicate that the component is a minor component; and
generating as output the estimated independent component analysis model excluding the minor component; and
identifying a physical mechanism corresponding to significant components of the estimated independent component analysis model.
9. The computer readable storage medium of claim 8 , the calculated percentages of variance indicating that a component is a minor component when a percentage of variance for each of the plurality of parameters explained by the component is less than a minimum percentage of variance for a single parameter.
10. The computer readable storage medium of claim 8 , the calculated percentages of variance indicating that a component is a minor component when an average percentage of variance for the plurality of parameters explained by the component is less than a minimum average percentage of variance.
11. The computer readable storage medium of claim 8 further comprising a step of calculating confidence intervals for rotation angles of the estimated independent component analysis model.
12. The computer readable storage medium of claim 11 further comprising a step of labeling a component of the estimated independent component analysis model to correspond to a specific physical process.
13. The computer readable storage medium of claim 12 further comprising a step of labeling the component as critical dimension of gate width effect on device drive current, critical dimension of gate width effect on device threshold voltage, critical dimension of gate length effect, implant dose effect, and gate oxide thickness effect.
14. The computer readable storage medium of claim 12 further comprising a step of examining the labeled component to identify the corresponding physical mechanism associated with the specific physical process.