Collision Detection Dataset in Time Windows (BIN)

This dataset was designed for the development and evaluation of binary classification models focused on collision event detection. Each record in the dataset represents a ‘time window’ or ‘BIN,’ containing 24 samples (P1_PDP to P24_PDP) that describe the transmission events.

Dataset Structure:

The dataset is provided in two separate files:

RACH_XTRAIN_YTRAIN_SIMULATION_9010_ETU70: This file contains the training set, combining the 24 feature columns (P1_PDP to P24_PDP) and the target column (Label).

Dimensions: 25 columns and 850,077 rows.

RACH_XTEST_YTEST_SIMULATION_9010_ETU70: This file contains the test set, also with the 24 feature columns and the target column (Label).

Dimensions: 25 columns and 57,988 rows.

Columns:

P1_PDP to P24_PDP: These 24 columns contain numerical data, where each Px_PDP represents a sample within a time window (‘BIN’). The values reflect transmission events.

Label: This is the target column, indicating the class of the event within the respective time window.

0: Represents the absence of a collision (normal state or no incident).

1: Represents the occurrence of a collision.