Inversion-conforming data sets processing

• US 20030212528A1
• Filed: 01/21/2003
• Published: 11/13/2003
• Est. Priority Date: 01/25/2002
• Status: Active Grant

• Current Assignee: Chandler, Larry S.
• Original Assignee: Chandler, Larry S.
• Inventors: Chandler, Larry S.
• Granted Patent: US 7,107,048 B2
• Time in Patent Office: Days
• US Class Current: 702/179
• CPC Class Codes: G06F 17/17 Function evaluation by approximation methods, e.g. inter- or extrapolation, smoothing, least mean square method G06F17/18 takes precedence ; interpolation for numerical control G05B19/18

Abstract

Representations of data inversions are generated by the processing of least-squares and maximum likelihood estimates which are rendered in correspondence with inversion-conforming data sets being related to considered sets of sampled data-point coordinates and respective parametric approximative form. Inversion-conforming data sets are considered to be sets of coordinates for points which correspond to the projection of respective data points along corresponding coordinates onto a fitting approximation or onto the locus of an alternate approximating relationship which respectively corresponds to a data inversion or a considered estimate of the same. Slope-handling and alternate projection normalizing coefficients may be implemented to provide common units for combining orthogonal coordinate-oriented data-point projections for rendering respective inversions. Preliminary least-squares estimates may be adjusted in correspondence with characteristic dispersion models or directly implemented as initial estimates to render respective forms of maximum likelihood estimating in correspondence with multi-dimensional probability density functions. The variability in the displacement of inversion-conforming data sets from corresponding data-point coordinates is functionally related to respectively orthogonal components of dispersion accommodating variability. Bi-coupled measurement dispersions are evaluated to establish dispersion accommodating variabilities as related to both respective and antecedent orthogonal measurement dispersions.

Claims

1. A method for accessing, processing, and representing information whereby a data representation is generated in correspondence with a plurality of data-point projections,

each of said data-point projections extending from respective data-point coordinates to intersect an approximating relationship,

intersections of said data-point projections with said approximating relationship substantially establishing respective inversion-conforming data sets which comprise projected coordinates for points that conform to a corresponding data inversion at respective said intersections,

each of said sets of projected coordinates comprising a respective subset of said data-point coordinates and at least one parametrically determined variable measure,

each said determined variable measure being determined in correspondence with said approximating relationship and at least one data-point coordinate of said respective subset,

said approximating relationship establishing respective correspondence between said at least one data-point coordinate and said at least one parametrically determined variable measure,

said approximating relationship being determined in correspondence with a parametric approximative form and at least one parameter estimate, and

said data representation being rendered in correspondence with said parametric approximative form and at least one inversion parameter estimate,

said method comprising:

activating means for said accessing, processing, and representing information whereby at least one form of inversion-conforming data sets processing is effectuated;

accessing provided data;

establishing said parametric approximative form;

establishing at least one respective inversion parameter estimate;

determining said approximating relationship as related to said parametric approximative form and said at least one respective inversion parameter estimate; and

effectuating said at least one form of inversion-conforming data sets processing;

said means for accessing, processing, and representing information comprising;

a control system; and

said control system being configured for providing said activating, said effectuating, and said representing information;

said effectuating including implementing and effecting at least one form of parameter estimating in correspondence with said provided data whereby said at least one inversion parameter estimate is generated.

2. A method as in claim 1 wherein said inversion-conforming data sets processing includes the summing together of at least some addends in correspondence with each of at least two variable degrees of freedom,

said method including said summing,

each of said some addends being generated with respect to corresponding approximation-conforming data sets, and

said corresponding approximation-conforming data sets substantially representing respective intersections of at least two orthogonal data-point projections with said approximating relationship in correspondence with said provided data and said at least two variable degrees of freedom.

3. A method as in claim 1 wherein said activating includes activating means for rendering at least one correction to a considered data inversion,

said correction being rendered in correspondence with said considered data inversion and successive simulations of data,

said method including said rendering, and

said data representation including said correction;

said correction being generated by;

representing an initial estimate of said correction in correspondence with a respectively considered approximating relationship of corresponding parametric approximative form,

rendering said simulations by combining successive estimates of said correction with a characteristic representation of error deviations,

rendering inversions of said successive simulations by implementing similar processing techniques for successively processing said simulations that were used in processing to generate said considered data inversion,

rendering said successive estimates of said correction by combining prior said estimates with said considered data inversion and with respective inversions of said successive simulations, and

implementing and effecting at least some form of successive estimate approximating and thereby rendering inversions of said successive simulations to more closely replicate said considered data inversion.

4. A method as in claim 1 wherein said inversion-conforming data sets processing includes rendering the square of at least one projection normalizing coefficient,

said method including said rendering,

said square of at least one projection normalizing coefficient being substantially represented in proportion to the square of a respective slope-handling coefficient,

said slope-handling coefficient being represented as inversely proportional to a root of the absolute value of at least one -derivative of change in the local measure of said determined variable being considered with respect to corresponding change in the respective measure of at least one considered orthogonal variable,

the ratio of said change to said corresponding change being substantially represented in correspondence with said approximating relationship,

said slope-handling coefficient being determined at coordinates which substantially correspond to at least two considered elements of a respective inversion-conforming data set,

said effectuating encluding representing an estimated variability,

said estimated variability substantially representing possible variations in said determined variable measure as related to a representation of uncertainty in the sampling of at least one element of said respective subset of data-point coordinates, and

said representation of uncertainty being considered to characterize variations in possible outcomes of acquisition sampling of a form corresponding to the sampling used in acquiring the respective coordinate sample for said at least one element.

5. A method as in claim 2 wherein said effectuating includes representing an estimated variability,

said estimated variability substantially representing possible variations in said determined variable measure as related to a representation of uncertainty in the sampling of at least one element of said respective subset of data-point coordinates, and

said representation of uncertainty being considered to characterize variations in possible outcomes of acquisition sampling of a form corresponding to the sampling used in acquiring the respective coordinate sample for said at least one element.

6. A method as in claim 2 wherein said inversion-conforming data sets processing includes rendering the square of at least one projection normalizing coefficient,

said method including said rendering,

said square of at least one projection normalizing coefficient being substantially represented in proportion to the square of a respective slope-handling coefficient,

said slope-handling coefficient being represented as inversely proportional to a root of the absolute value of at least one derivative of change in the local measure of said determined variable being considered with respect to corresponding change in the respective measure of at least one considered orthogonal variable,

the ratio of said change to said corresponding change being substantially represented in correspondence with said approximating relationship, and

said slope-handling coefficient being determined at coordinates which substantially correspond to at least two considered elements of a respective inversion-conforming data set.

7. A method as in claim 6 wherein the value of said root substantially corresponds to the number of simultaneously considered variable degrees of freedom.

8. A method as in claim 6 wherein said effectuating includes representing an estimated variability,

said estimated variability substantially representing possible variations in said determined variable measure as related to a representation of uncertainty in the sampling of at least one element of said respective subset of data-point coordinates, and

said representation of uncertainty being considered to characterize variations in possible outcomes of acquisition sampling of a form corresponding to the sampling used in acquiring the respective coordinate sample for said at least one element.

9. A method as in claim 6 wherein said activating includes activating means for rendering at least one correction to a considered data inversion,

said correction being rendered in correspondence with said considered data inversion and successive simulations of data,

said method including said rendering, and

said data representation including said correction;

said correction being generated by;

representing an initial estimate of said correction in correspondence with a respectively considered approximating relationship of corresponding parametric approximative form,

rendering said simulations by combining successive estimates of said correction with a characteristic representation of error deviations,

rendering inversions of said successive simulations by implementing similar processing techniques for successively processing said simulations that were used in processing to generate said considered data inversion,

rendering said successive estimates of said correction by combining prior said estimates with said considered data inversion and with respective inversions of said successive simulations, and

implementing and effecting at least some form of successive estimate approximating and thereby rendering inversions of said successive simulations to more closely replicate said considered data inversion.

10. A data processing system comprising:

a control system;

means for accessing, processing, and representing information;

said control system being configured for providing said accessing, processing, and representing information; and

said control system being configured for generating at least one data inversion in correspondence with a parametric approximative form;

said processing including;

accessing information,

effectuating at least one form of inversion-conforming data sets processing, and

rendering at least one data representation in correspondence with said at least one data inversion;

said effectuating including;

accessing provided data,

establishing said parametric approximative form for said at least one data representation,

establishing at least one determined form for an approximating relationship as related to said parametric approximative form and at least one respective parameter estimate, and

implementing and effecting at least one form of parameter estimating in correspondence with said provided data whereby said at least one data inversion is generated;

said data inversion being generated in correspondence with a plurality of data-point projections,

each of said data-point projections extending from respective data-point coordinates to intersect said approximating relationship,

intersections of said data-point projections with said approximating relationship substantially establishing respective inversion-conforming data sets which comprise projected coordinates for points that conform to said data inversion at respective said intersections,

each of said sets of projected coordinates comprising a respective subset of said data-point coordinates and at least one correspondingly determined variable measure,

each said determined variable measure being determined in correspondence with said approximating relationship and at least one data-point coordinate of said respective subset, and

said approximating relationship establishing respective correspondence between said at least one data-point coordinate and said at least one correspondingly determined variable measure.

11. A data processing system as in claim 10 wherein said inversion-conforming data sets processing includes the summing together of at least some addends in correspondence with each of at least two variable degrees of freedom,

said effectuating including representing an estimated variability and rendering said summing,

each of said some addends being generated with respect to corresponding approximation-conforming data sets,

said corresponding approximation-conforming data sets substantially representing respective intersections of at least two orthogonal data-point projections with said approximating relationship in correspondence with said provided data and said at least two variable degrees of freedom,

said estimated variability substantially representing possible variations in said determined variable measure as related to a representation of uncertainty in the sampling of at least one element of said respective subset of data-point coordinates, and

said representation of uncertainty being considered to characterize variations in possible outcomes of acquisition sampling of a form corresponding to the sampling used in acquiring the respective coordinate sample for said at least one element.

12. A data processing system as in claim 10 wherein said inversion-conforming data sets processing includes summing together of at least some addends in correspondence with each of at least two variable degrees of freedom,

said effectuating including representing the square of at least one projection normalizing coefficient and rendering said summing,

each of said some addends being generated with respect to corresponding approximation-conforming data sets,

said corresponding approximation-conforming data sets substantially representing respective intersections of at least two orthogonal data-point projections with said approximating relationship in correspondence with said provided data and said at least two variable degrees of freedom,

said square of at least one projection normalizing coefficient being substantially represented in proportion to the square of a respective slope-handling coefficient,

said slope-handling coefficient being represented as inversely proportional to a root of the absolute value of at least one derivative of change in the local measure of said determined variable being considered with respect to corresponding change in the respective measure of at least one considered orthogonal variable,

the ratio of said change to said corresponding change being substantially represented in correspondence with said approximating relationship, and

said slope-handling coefficient being determined at coordinates which substantially correspond to at least two considered elements of a respective inversion-conforming data set.

13. A data processing system as in claim 12 comprising means for generating at least one weighting coefficient,

said effectuating including generating said at least one weighting coefficient,

said weighting coefficient being rendered directly proportional to said respective slope-handling coefficient,

said generating encluding representing an estimated variability,

said estimated variability substantially representing possible variations in said determined variable measure as related to a representation of uncertainty in the sampling of at least one element of said respective subset of data-point coordinates, and

said representation of uncertainty being considered to characterize variations in possible outcomes of acquisition sampling of a form corresponding to the sampling used in acquiring the respective coordinate sample for said at least one element.

14. A data processing system as in claim 12 comprising means for generating at least one correction to a considered data inversion,

said correction being rendered in correspondence with said considered data inversion and successive simulations of data,

said effectuating including generating said at least one correction, and

said at least one data representation including said correction;

said correction being generated by;

representing an initial estimate of said correction in correspondence with a respectively considered approximating relationship of corresponding parametric approximative form,

rendering said simulations by combining successive estimates of said correction with a characteristic representation of error deviations,

rendering inversions of said successive simulations by implementing similar processing techniques for successively processing said simulations that were used in processing to generate said considered data inversion,

rendering said successive estimates of said correction by combining prior said estimates with said considered data inversion and with respective inversions of said successive simulations, and

implementing and effecting at least some form of successive estimate approximating and thereby rendering inversions of said successive simulations to more closely replicate said considered data inversion.

15. A weighting coefficient generator comprising means for generating a representation of at least one weighting coefficient,

said weighting coefficient being rendered as substantially related to inverse proportion of the estimated variability of a parametrically determined variable measure,

said determined variable measure being restricted to the confines of an approximating relationship,

said approximating relationship being established in correspondence with a parametric approximative form and at least one parameter estimate,

said at least one parameter estimate being generated in correspondence with a plurality of approximation-conforming data sets,

each of said approximation-conforming data sets comprising a corresponding subset of respective data-point coordinates and at least one respectively determined variable measure,

each said respectively determined variable measure being determined in correspondence with said approximating relationship and at least one data-point coordinate of the corresponding said subset,

each said subset excluding a sample for the coordinate which corresponds to the respective said parametrically determined variable,

said estimated variability not being represented by the variance of a respectively considered data-point-sample acquisition for said variable,

said estimated variability not being represented as an effective variance of a respectively considered data-point-sample acquisition for said variable,

said estimated variability being represented in correspondence with possible variations in said parametrically determined variable measure,

said possible variations being associated with uncertainty in lateral components of variability being correspondingly restricted to said confines,

said lateral components of variability being related to variations in possible outcomes in the sampling acquisition of respective elements of corresponding subsets of data-point coordinates being considered in correspondence with restraints imposed by said approximating relationship,

said estimated variability substantially representing possible variations in said parametrically determined variable measure as related to uncertainty in the sampling of said at least one element of the corresponding said subset, and

representation of said uncertainty being related to variations in possible outcomes of orthogonal component acquisition sampling of a form corresponding to the sampling used in acquiring the respective coordinate sample for said at least one element;

said means for generating including;

a control system;

means for accessing, processing, and representing information; and

said control system being configured for providing said accessing, processing, and representing information;

said processing including;

accessing provided data,

representing said estimated variability in correspondence with said approximating relationship, and

rendering said weighting coefficient in correspondence with the inverse of said estimated variability.

16. A weighting coefficient generator as in claim 15 wherein said at least one weighting coefficient is rendered in direct proportion to the square of a projection normalizing coefficient,

said projection normalizing coefficient being substantially represented in proportion to a respective slope-handling coefficient,

said slope-handling coefficient being represented as inversely proportional to a root of the absolute value of at least one derivative of change in the local measure of said determined variable being considered with respect to corresponding change in the respective measure of at least one considered orthogonal variable,

the ratio of said change to said corresponding change being substantially represented in correspondence with said approximating relationship, and

said slope-handling coefficient being determined at coordinates which substantially correspond to at least two considered elements of a respective approximation-conforming data set.

17. A weighting coefficient generator as in claim 15 wherein said representing information includes:

representing at least one other estimate of variability in correspondence with said approximating relationship,

said at least one other estimate of variability substantially representing possible variations in a determined measure of at least one other variable as related to respectively considered variations in at least one element of an alternate subset of said data-point coordinates,

said at least one other estimate of variability not being represented by the variance of a respectively considered data-point-sample acquisition for said other variable,

said at least one other estimate of variability not being represented as an effective variance of a respectively considered data-point-sample acquisition for said other variable,

said determined measure of at least one other variable being restricted to said confines,

possible variations in said determined measure being associated with uncertainty in at least one alternate lateral component being correspondingly restricted to said confines,

representation of said uncertainty in said alternate lateral component being related to variations in possible outcomes in the sampling acquisition of at least one element of said alternate subset,

said alternate subset excluding a respective sample for the coordinate which corresponds to said at least one other variable, and

said approximating relationship establishing respective correspondence between parametrically determined measure of said at least one other variable and at least one element of said alternate subset;

said processing including;

generating said at least one other weighting coefficient in correspondence with the inverse of said at least one other estimate of variability,

rendering said at least one weighting coefficient and said at least one other weighting coefficient in proportion to corresponding squares of respective slope-handling coefficients,

said at least one weighting coefficient being substantially represented as inversely proportional to a root of the absolute value of at least one derivative of change in the measure of said parametrically determined variable,

said at least one derivative of change in the measure of said parametrically determined variable being considered with respect to a corresponding change in the measure of said at least one other variable,

said at least one other weighting coefficient being substantially represented as inversely proportional to a root of the absolute value of at least one derivative of change in the determined measure of said at least one other variable,

said at least one derivative of change in the determined measure of said at least one other variable being considered with respect to a corresponding change in the measure of said parametrically determined variable,

each said derivative of change being related to the locus of conforming points along respective confines of said approximating relationship,

each said derivative of change being evaluated in correspondence with at at least two respective coordinates, and

said respective coordinates substantially corresponding to considered elements of respective approximation-conforming data sets.

18. An output product comprising a data representation generated by a data processing system, and

said data representation being rendered in correspondence with at least one data inversion;

said data representation comprising;

processed information, and

said processed information establishing a descriptive correspondence of a determined parametric approximative form;

said output product being generated by;

activating means for accessing, processing, and representing information whereby at least one form of inversion-conforming data sets processing is effectuated,

accessing provided data,

effectuating said at least one form of inversion-conforming data sets processing, and

rendering said data representation in correspondence with said effectuating;

said effectuating including;

representing an approximating relationship as determined in correspondence with said parametric approximative form and at least one parameter estimate,

implementing and effecting at least one form of parameter estimating in correspondence with said provided data whereby at least one inversion parameter estimate is generated,

said at least one inversion parameter estimate being generated in correspondence with a plurality of data-point projections and said parametric approximative form,

each of said data-point projections extending from respective data-point coordinates to intersect an approximating relationship,

intersections of said data-point projections with said approximating relationship substantially establishing respective inversion-conforming data sets which comprise projected coordinates for points that conform to a respective data inversion at said intersections,

each of said sets of projected coordinates comprising a respective subset of said data-point coordinates and at least one correspondingly determined variable measure,

each said determined variable measure being determined in correspondence with said approximating relationship and at least one data-point coordinate of said respective subset, and

said approximating relationship establishing respective correspondence between said at least one data-point coordinate and said at least one correspondingly determined variable measure;

said data processing system comprising;

said means for accessing, processing, and representing information; and

means for effectuating said at least one form of inversion-conforming data sets processing;

said means for accessing, processing, and representing information comprising;

a control system, and

said control system being configured for providing said activating, said effectuating, and said representing information.

19. An output product as in claim 18 comprising a memory for storing data for access by an application program being executed on a processing system, said data representation being stored in said memory.

20. An output product as in claim 18 comprising at least one correction to a considered data inversion,

said activating including activating means for generating said at least one correction to said considered data inversion, and

said rendering including said generating;

said effectuating including;

said generating, and

summing at least some addends together in correspondence with each of at least two variable degrees of freedom,

each of said some addends being generated with respect to corresponding approximation-conforming data sets,

at least two of said corresponding approximation-conforming data sets substantially representing respective intersections of at least two orthogonal data-point projections with said approximating relationship in correspondence with said provided data and said at least two variable degrees of freedom,

said correction being generated in correspondence with said considered data inversion and successive simulations of data, and

said processed information comprising said at least one correction;

said correction being generated by;

representing an initial estimate of said correction in correspondence with a respectively considered approximating relationship of corresponding parametric approximative form,

rendering said simulations by combining successive estimates of said correction with a characteristic representation of error deviations,

rendering inversions of said successive simulations by implementing similar processing techniques for successively processing said simulations that were used in processing to generate said considered data inversion,

rendering said successive estimates of said correction by combining prior said estimates with said considered data inversion and with respective inversions of said successive simulations, and

implementing and effecting at least some form of successive estimate approximating and thereby rendering inversions of said successive simulations to more closely replicate said considered data inversion.

21. An output product as in claim IS wherein said effectuating includes rendering the square of at least one projection normalizing coefficient and summing at least some addends together in correspondence with each of at least two variable degrees of freedom,

each of said some addends being generated with respect to corresponding approximation-conforming data sets,

at least two of said corresponding approximation-conforming data sets substantially representing respective intersections of at least two orthogonal data-point projections with said approximating relationship in correspondence with said provided data and said at least two variable degrees of freedom,

said square of at least one projection normalizing coefficient being substantially represented in proportion to the square of a respective slope-handling coefficient,

said slope-handling coefficient being represented as inversely proportional to a root of the absolute value of at least one derivative of change in the local measure of respective said determined variable being considered with respect to corresponding change in the respective measure of at least one considered orthogonal variable,

the ratio of said change to said corresponding change being substantially represented in correspondence with said approximating relationship, and

said slope-handling coefficient being determined at coordinates which substantially correspond to at least two considered elements of a respective inversion-conforming data set.

22. An output product as in claim 21 wherein said effectuating includes representing an estimated variability,

said estimated variability substantially representing possible variations in said determined variable measure as related to a representation of uncertainty in the sampling of at least one element of said respective subset of data-point coordinates, and

said representation of uncertainty being considered to characterize variations in possible outcomes of acquisition sampling of a form corresponding to the sampling used in acquiring the respective coordinate sample for said at least one element.

23. A method for rendering a data representation of at least one correction to a considered data inversion of a determined parametric approximative form,

said correction being rendered in correspondence with a considered data inversion and successive simulations of data,

said correction being generated by:

representing an initial estimate of said correction in correspondence with a respectively considered approximating relationship of corresponding parametric approximative form,

rendering said simulations by combining successive estimates of said correction with a characteristic representation of error deviations,

rendering inversions of said successive simulations by implementing similar processing techniques for successively processing said simulations that were used in the processing of data to generate said considered data inversion,

rendering said successive estimates of said correction by combining prior said estimates with said considered data inversion and with respective inversions of said successive simulations, and

implementing and effecting at least some form of successive estimate approximating and thereby rendering inversions of said successive simulations to more closely replicate said considered data inversion.

said method comprising;

activating means for said accessing, processing, and representing information whereby said data representation is generated,

accessing provided data, and

effectuating at least one form of parameter estimating in correspondence with said provided data whereby said data representation is generated;

said means for accessing, processing, and representing information comprising;

a control system,

said control system being configured for providing said activating, said effectuating, and said representing, and

said method excluding rendering characteristic form iterations by a data processing system configured for implementing at least one form of inverse deviation variation weighting for representing the weight of respective function related deviations.