This dataset provides normalized real-world measurements of energy consumption and operational data from a large-scale 5G network deployment. It includes eight days of measurements collected from more than 1,000 RRU/AAUs, covering 12 different hardware products.

The dataset is intended to support research in energy-efficient mobile networks, network optimization, and data-driven modeling of 5G systems.

📂 Dataset Structure

The dataset is organized into three main CSV files:

1. BSinfo.csv – Base Station Basic Information

Contains static information about the deployed base stations and their cells.

Fields:

BS: Identifier of the base station
CellName: Cell identifier (multiple cells can be configured per BS, named CellX)
RUType: Radio unit product name
Mode: Transmission mode
Bandwidth: Normalized cell bandwidth
Frequency: Normalized cell frequency
Antennas: Number of antennas
TXpower: Maximum transmit power of the cell

2. CLstat.csv – Cell-Level Statistics

Contains hourly counters describing the operational status of each cell.

Fields:

Time: Timestamp of the measurement
BS: Identifier of the base station
CellName: Cell identifier
Load: Load of the cell (share of used resources)
EnergySavingMode: Intensity of activation of energy-saving mechanisms

3. ECstat.csv – Energy Consumption

Contains hourly measurements of energy consumption per base station.

Fields:

Time: Timestamp of the measurement
BS: Identifier of the base station
Energy: Energy consumption of the base station

📊 Data Summary

Measurement period: 8 days
Number of RRUs/AAUs: > 1,000
Hardware diversity: 12 different product types
Granularity: Hour-level measurements for load, counters, and energy consumption

🧪 Applications

This dataset can be used for:

Analysis of energy consumption patterns in 5G networks
Evaluation of energy-saving methods and load-dependent consumption
Development of predictive models for network optimization
Benchmarking of AI/ML approaches for green networking

Downloads last month: 16