Statistical Data Format File (.sdf)
SDF file is an ASCII format file used save the data in the different spreadsheets used in the Statistical Analysis window, including:
| • |
Parameter variance explained |
| • |
Dominant parameter-based equations |
so that they can be used in a different sessions of the Statistical analysis.
A typical SDF file has different data blocks corresponding to each of the spreadsheet mentioned above. Each data block has an identifier to mark the beginning and the end. The beginning is marked by BEGIN_<data-type> and the end by END. The begin identifiers are BEGIN_PARMDATA, BEGIN_CORRELATION, BEGIN_FACTOR_LOADINGS, BEGIN_PARAMETER_VARIANCE, BEGIN_EQN_FACTOR, BEGIN_EQN_DOMPARM.
Comments begin with the character "!".
Parameters Data Block
| • |
Each line or row in the data block represent a single sample of a particular device. |
| • |
Each column specifies either an Attribute or a Model Parameter. |
| • |
The first line in the block represents the header and contains information regarding the data block. |
| • |
Each attribute or parameter value can be either space or tab separated. |
| • |
"~#A" appended to an attribute name, specifies that the data in the column rep resent and attribute as opposed to a data. |
| • |
Deactivated columns are identified by "~#C" appended to the parameter name. |
| • |
An attribute column cannot be a deactivated column. |
| • |
Data filtered rows are identified by "~#F" in the first column of the row. |
| • |
Attribute filtered rows by "~#AF". |
| • |
Deactivated rows by "~#R". |
Correlation Data Block
| • |
The first line in the block has the correlated parameter names. |
| • |
The first cell or word in a row also has the correlated parameter name. |
| • |
All other values represent a correlation value. |
Factor Loading Data Block
This data block is divided into two sections:
| • |
Dominant Parameter/Factor Group details |
The first section has 6 lines, identifying:
| • |
Default Dominant Parameter row number |
| • |
Current Dominant Parameter row number |
| • |
Dominant parameter type: 0 - Default, 1 - User defined |
| • |
Default Factor Group column number |
| • |
Current Factor Group column number |
| • |
Factor Group type: 0 - Default, 1 - User defined |
The second section contains the factor values and the parameter names.
Parameter Variance Explained Data Block
This data block always has only 4 lines excluding the markers.
Factor-based Equations and Parameter-based Equations
These two data blocks are similar and contain the equation coefficient values. The first line is the Parameter line.
Example
Portion of the bsim3.sdf file in ${ICCAP_ROOT}/examples/icstat is reproduced below:
! @(#) $Source: /wlv/src/iccap500/mdl/icstat_examples/bsim3.sdf,v $ $Revision: 1 .4 $
! $Date: 97/06/09 16:24:24 $
! Date/Time: Wed May 21 11:17:46 1997
! Filename: /tmp_mnt/hpdisk15/lrich/bsim3new.sdf
!******************************************************************************
! IC-CAP Statistical Analysis Data Set File
! File Format:
! * Comments begin with !
! * A begin and end of a member-set data block is marked by BEGIN_<data-set Type >
! and END. Eg. BEGIN_PARMDATA, END.
! * Parameter rows are identified by the keyword PARAMETER
! * Identifiers for the different data-sets
! * Raw data data-set: BEGIN_PARMDATA
! * Correlation data-set: BEGIN_CORRELATION
! * Factor Loadings data-set: BEGIN_FACTOR_LOADINGS
! * Parameter Variance Explained data-set: BEGIN_PARAMETER_VARIANCE
! * Equation data-set: EQUATIONS
! Parameters Data Set:
! ---------------------
! * Each Row in the raw data-set matrix represent a single sample
! * Each field in a single sample data-set are separated by one or more space(s) /tab
! * deactivated rows are identified by the Delimiter: ~#R in the first column
! * deactivated columns are identified by the Delimiter: ~#C appended to the <pa ram-name>
! * Attribute Columns are identified by the Delimiter: ~#A appended to the <para m-name>
! * Data filtered rows are identified by the Delimiter: ~#F in the first column
! * Attribute filtered rows are identified by the Delimiter: ~#AF in the first column
!******************************************************************************
BEGIN_PARMDATA
PARAMETER LotID~#A WaferID~#A Module~#A VTH0~#C TOX K1 K2 PDIBL1 ~#R JAMES 5188A039
SAPT_7I 0.669 1.643e-08 0.873 -0.0648 0.21 ~#F JAMES 5188A039 SAPT_7I 0.67 1.64e-08 0.863
-0.0619 1.53
~#AF JAMES 5188A039 SAPT_7I 0.666 1.638e-08 0.867 -0.0624 4.77 JAMES 5188A039 SAPT_7I 0.664
1.642e-08 0.863 -0.0658 10.5 JAMES 5188A039 SAPT_7I 0.661 1.642e-08 0.858 -0.064 0.719 JAMES
5188A039 SAPT_7I 0.658 1.639e-08 0.851 -0.0596 3.12 JAMES 5188A039 SAPT_7I 0.656 1.637e-08
0.85 -0.0599 0.401 JAMES 5188A039 SAPT_7I 0.652 1.634e-08 0.857 -0.0625 0.17 JAMES 5188A039
SAPT_7I 0.66 1.641e-08 0.859 -0.0628 3.58
...
END
|
|
