RAN Updates Traffic Impact Dataset

This repository contains the dataset accompanying the paper:

"A First Look at Operational RAN Updates and Their Impact on Carrier Traffic Demands and Prediction"

This dataset provides 1,931 traffic time series measured at carriers affected by RAN modifications in a nationwide operational infrastructure.

Dataset Structure

The dataset is organized into two related tables linked by a unique identifier (UUID), following a normalized structure to reduce redundancy:

dataset/
├── traffic_dataset.parquet    # Normalized Hourly traffic measurements
├── info_dataset.parquet       # General carrier information
README.md
create_dataset.ipynb

Data Dictionary

info_dataset.parquet — Carrier Information Table

Contains one record per impacted carrier with static metadata.

Column	Type	Description
UUID	string	Unique identifier to use to correlate with the traffic table
antennaID	string	Anonymized antenna identifier
urbanization	string	Area classification: Rural, Suburban, Urban, Metropolitan, or Metropolitan Center
technology	string	Radio Access Technology of the impacted carrier (e.g., 4G)
event_type	string	Type of RAN update event (e.g., 4G addition, 5G addition — opening of new antenna at the same base station location)

traffic_dataset.parquet — Traffic Measurements Table

Contains hourly traffic time series for each carrier within a 120-day window centered on the RAN update event.

Column	Type	Description
UUID	string	UUID key linking to info_dataset.parquet
uxtime	int	Absolute timestamp in Unix time
deltaT	int	Time offset in seconds relative to the event period (negative = before event, positive = after event)
DL	float	Downlink traffic volume (standardized, z-score normalized)
UL	float	Uplink traffic volume (standardized, z-score normalized)

---

Example Usage

We will see how to load and plot a single timeseries

import pandas as pd
import matplotlib.pyplot as plt
from statsmodels.tsa.seasonal import seasonal_decompose
import matplotlib.pyplot as plt

# Load data
df_traffic = pd.read_parquet('traffic_dataset.parquet')
df_info = pd.read_parquet('info_dataset.parquet')

# Get a single carrier's time series
sample_uuid = df_info['UUID'].iloc[12]
carrier_traffic = df_traffic[df_traffic['UUID'] == sample_uuid].copy()
carrier_traffic = carrier_traffic.sort_values('deltaT').reset_index(drop=True)

# Convert deltaT to days and aggregate to daily level
carrier_traffic['days'] = carrier_traffic['deltaT'] / 3600 / 24
daily_traffic = carrier_traffic.groupby(carrier_traffic['days'].astype(int))['DL'].mean()

# Perform seasonal decomposition (period=7 for weekly seasonality)
decomposition = seasonal_decompose(daily_traffic, model='additive', period=7)

# Plot the decomposition
fig, axes = plt.subplots(4, 1, figsize=(14, 8), sharex=True)

axes[0].plot(daily_traffic.index, decomposition.observed, color='blue')
axes[0].axvline(x=0, color='red', linestyle='--', alpha=0.7, label='Event')
axes[0].set_ylabel('Observed')
axes[0].set_title(f'Time Series Decomposition (UUID: {sample_uuid[:8]}...)')
axes[0].legend()

axes[1].plot(daily_traffic.index, decomposition.trend, color='orange')
axes[1].axvline(x=0, color='red', linestyle='--', alpha=0.7)
axes[1].set_ylabel('Trend')

axes[2].plot(daily_traffic.index, decomposition.seasonal, color='green')
axes[2].axvline(x=0, color='red', linestyle='--', alpha=0.7)
axes[2].set_ylabel('Seasonal')

axes[3].plot(daily_traffic.index, decomposition.resid, color='red')
axes[3].axvline(x=0, color='red', linestyle='--', alpha=0.7)
axes[3].set_ylabel('Residual')
axes[3].set_xlabel('Days from Event')

plt.tight_layout()
plt.savefig('time_series_decomposition.png', dpi=300)
plt.show()

Data Processing Notes

Standardization

Traffic values (DL, UL) are z-score normalized using statistics from a 30-day reference period located two months before the RAN update:

Each time series spans a 120-day window centered on the RAN update event:

• 60 days before the update (including a 30-day reference period for normalization)
• 60 days after the update

Selection Criteria

Carriers in this dataset were selected based on:

1. Isolated updates: Only RAN updates occurring without concurrent modifications at the same site and azimuth
2. Statistical significance: Carriers passing a Wilcoxon signed-rank test (α = 0.01) confirming significant traffic change
3. Impact type: Carriers classified in "Cluster A" — those experiencing a significant drop in served traffic following the update

---

Citation

If you use this dataset in your research, please cite:

@inproceedings{ran_updates_2025,
  title     = {A First Look at Operational RAN Updates and Their Impact on Carrier Traffic Demands and Prediction},
  author    = {[Antonio Boiano, Nadezhda Chukhno, Zbigniew Smoreda, Alessandro E. C. Redondi, Marco Fiore]},
  booktitle = {[INFOCOM 2026]},
  year      = {2026}
}

---

License

This dataset is released under the Creative Commons Attribution 4.0 International License (CC BY 4.0).

You are free to:

• Share — copy and redistribute the material in any medium or format
• Adapt — remix, transform, and build upon the material for any purpose

Under the following terms:

• Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made.

---

Contact

For questions about the dataset, please contact: antonio.boiano@polimi.it, nadezda.chukhno@networks.imdea.org, marco.fiore@networks.imdea.org

For issues or contributions to this repository, please open a GitHub issue.

---

Acknowledge

The work involving the data collection, processing, and analysis was funded by CoCo5G (ANR-22-CE25-0016) of the French National Research Agency (ANR) as well as by the 6G-IRONWARE (CNS2023-143870) funded by MICIU/AEI /10.13039/501100011033 and by the European Union NextGenerationEU/PRTR, and by the 6G-AI-TANGO (GA 101206327) funded by the European Union.