Data Level¶
Explanation¶
You can see here where the different data levels are found in the workflow of the logger component in the graphics .
Level 0¶
Level 0 data is the term used to describe the raw data that is usually stored in file form, which are generated directly by the data logger. The structure of Level 0 data is predefined by the manufacturer and can usually only be adapted to a limited extent. Simple comma-separated tables in ASCII format (CSV) are often used. However, proprietary binary formats or XML are also used.
Level 1¶
Level 1 data is generated by the processes in the DMP and represents the interpreted raw data. Manufacturer-specific parsers are used to read in the Level 0 data and convert the individually structured raw data into a standardized format. In addition to the acquisition of the measured values, the measured data is also linked to the metadata stored in the sensor configuration. Furthermore, Level 1 data can be regenerated via virtual sensors, manipulated by measurement functions or checked for plausibility in a rudimentary way by specifying limits.
Measurement functions¶
When reading in measured values, they can be changed by measurement functions or new values can be calculated within virtual sensors.
The measurement function is formulated in Javascript. For mathematical operations the Math Builtin is available. In the sensor configuration there is the possibility to use the function edit in editor. Above the input field measurement function there is an info button, which also reveals a simple example when hovered over with the mouse.
Level 2¶
Level 2 corresponds to data for which quality flags are stored in addition to the measured values. These quality flags can be generated in different ways and linked to Level 1 data. Depending on the process used, Level 2 data is divided into Level 2a and Level 2b.
Level 2a¶
For Level 2a data, the Level 1 measured values are directly linked to quality flags. This can be done either by manual flagging in the DMP web interface or by importing the flags and the identification characteristics of the measured values ("LEVEL1_VALUE_ID"). The reimport format for Level 2a quality characteristics is specified under manual upload.
Level 2b¶
Unlike Level 2a data, Level 2b data do not necessarily have to go through the complete workflow from the Level 0 to Level 1 record in the DMP. Level 2b data can be processed independently. A reference to Level 1 data sets is not necessarily required, but recommended. In general, Level 2b data should also not be manipulated, cleaned, or padded but should simply represent raw data and quality flags in a consistent format. The reimport format for Level 2b data and quality attributes is specified under manual upload.
Level 3¶
Level 3 data are manipulated, cleaned or padded data that no longer allow a direct reference to the originally recorded Level 1 or Level 2 data without restrictions. Currently, the DMP does not provide an explicit possibility to process Level 3 data.