Pkg: dulogpy
Purpose: Handle data retrived by the duolog bio-logging system (e.g. range and proximity data)
Author: Dr. Jesus R. Hernández Montero
Contact: jesus.hdezmontero@gmail.com
Last uptade: 2023.03.30

dulogpy

This is a package with functions for handling files retrived by dulog. Dulog is a bio-logging system based on wireless sensors developed by Niklas Duda. This bio-loging system delivers interactions among components in terms of proximity between them.

Dulog components

The dulog bio-logging system has the following components (please refer to the developer's manual to read the details about each component):

• Ground Nodes
  • Base Station (BS)
  • Download Station (DS)
• Mobile Nodes (MN)
• Stationary Nodes (SN)

Retrived data

The data retrived can be categorized into:

• Range data: this data is delivered by GN and indicates the detection of a mobile or stationary node within its detectin range. The tables are saved in the files named "RANGE.CSV". BS as well DS deliver this data but tables have subtle differences.

• Proximity data: this data indicates the interaction between MN-MN, MN-SN and SN-SN. Both ground nodes can save this data in the file "MN_DATA.CSV". This data have redundancy.

dulogpy functions

The functions provide in this package are for loading the data (Range and Proximity)into your local environment ready to use for analysis. The functions are the following

mn_range_loader()

This function is for load range data delivered by the BS and DS. The functions does the following:

1. Parse datetime columns
2. Transform datatypes
3. Sort records chronologically
4. Sets the date and time as an index

Function arguments:

• path: is the directory of the file where the "RANGE.CSV" file is stored.
• is_ds: indicates if the files comes from a Download Station or not. The default is True. If you are loading range data from a Base Station this parameter should be set False.

mn_data_loader()

This function is for loading the proximity data delivered by ground nodes. The function does the following:

1. Parse datetime columns.
2. Transform datatypes.
3. Handle potential duplicates.
4. Sort records chronologically.
5. Sets the date and time as index.

Function arguments:

• path: is the directory of the file where the "MN_DATA.CSV" is stored.

no_redundancy()

This functions deals with redundancy of the data between dyads. Since Node A meets B and node B meets A, this data is redundant. This function deliver pair-wise interactions of specified nodes. This function uses the output DataFrame delivered by the mn_data_loader() function to do the following:

1. Captures interactions between specified nodes (e.g. A-B and B-A).
2. Fill missing values where data is not redundant.
3. Get the maximum RSSI values between nodes (returned in the column RSSI_max).
4. Specify the dyad in string format in the colum dyad.
5. If specified, returns interactions between a time window.

Function arguments

• df: DataFrame where the output of the mn_data_loader() is stored.
• Node1: string of the first node of a given dyad (e.g. 'A')
• Node2: string of the second node of a given dyad (e.g. 'B')
• start: datetime string specifying the start of the observation window, it should be specified in the following format YYYY-MM-DD HH:MM:SS. Default value is None.
• end: datetime string specifying the end of the observation window, it should be specified in the following format YYYY-MM-DD HH:MM:SS. Default value is None.

save_csv()

This is a basic function to store, into your working directory, DataFrames as .csv files delivered by mn_range_loader(), mn_data_loader() and no_redundancy().

Function arguments:

• df: DataFrame to be saved as .csv
• name: name of the file, do not include '.csv' e.g. 'my_file'
• index: specify is the index should be saved as a column or kept as index. The default value is True saving the index as colum. Saving the index as column allows to use the output file in R.