Categorization of Attributes

A Wafer does not contain meta data like units of quantities, interpolation methods (linear, logarithmic) or the data type (scalar, vector, ...) of an attribute. Instead, this information is stored in a separate database (input deck). This database defines several categories of attributes. An attribute is uniquely associated with its corresponding category via its name. The database must contain the names of all possible quantities that occur during a process flow. This data is called the configuration of the WAFER-STATE-SERVER and is made available in the Config class. A category is given by the following information:

Fig. 3.1 shows a part of the WAFER-STATE-SERVER configuration which depicts the attribute type class "Concentrations". During the reading process, names of encountered attributes are matched against the keywords in the members list. If the name of an attribute is not found an exception will be generated. Regular expressions can be used to match name combinations. In the shown example all attributes starting with the word e.g. Donors are configured as a Concentration. If the name of the attributes was found an interpolation object is selected according to the configured mechanism (interpolation and datatype). This interpolation object is used whenever the value of an attribute is requested on a position that requires an interpolation.

Figure 3.1: Configuration of attribute type concentrations. The example depicts the configuration of the attribute type Concentrations. The minval keyword defines the minimum value for this type of attribute. The members keyword holds a list with all possible names of concentration attributes. The interpolation and datatype keywords are used to select an interpolation mechanism. Unit defines the unit of the attribute type and invalidates defines conditions in which attributes of this type are removed from the Wafer.