README.md

metadata

license: apache-2.0
language:
  - en
tags:
  - telecommunications
  - mobile
  - internet
  - 4G
pretty_name: Synthetic Mobile Performance Dataset
size_categories:
  - 10M<n<100M

Dataset Card: Synthetic Mobile Network Performance

Dataset Description

This dataset contains synthetically generated mobile signal measurements designed to mirror real-world data in the UK. The data represents geolocated signal quality metrics from mobile devices, capturing a range of environmental and temporal conditions over several months in 2025.

All data has been anonymized, aggregated, and processed to protect user privacy. The synthetic dataset has undergone pre-processing, including coordinate validation and outlier removal.

Fields

The dataset contains the following columns:

• timestamp: (date) The date and time of the measurement, recorded in UTC.
• unique_cell: (string) A unique identifier for the mobile network cell to which the device was connected.
• measurement_type_name: (string) The type of measurement recorded. This column appears to be null in the provided sample.
• in_outdoor_state: (string) A categorical label indicating the predicted environment of the device, with values such as "Surely Indoor," "Probably Indoor," and "Surely Outdoor."
• value: (float) A numerical value associated with the measurement. The specific KPI this represents is not defined.
• latitude: (float) The latitude of the device's location at the time of measurement.
• longitude: (float) The longitude of the device's location at the time of measurement.
• signal_level: (float) The received signal strength, likely measured in dBm.

Potential Uses

This dataset is suitable for a variety of analyses, including:

• Network Coverage Mapping: Visualizing signal strength across different geographic areas to identify zones with strong or weak coverage.
• Performance Analysis: Correlating signal quality with factors like location (indoor/outdoor), time of day, and cell tower.
• Mobility Pattern Simulation: Understanding how user movement impacts network performance.
• Machine Learning Model Training: Developing models to predict signal quality based on location and environmental factors.