DoNext

<h1> DoNext: An Open Measurement Data Set for Machine Learning-driven 5G Mobile Network Analysis</h1>

This dataset has been published in conjunction with an article in <em>IEEE Transactions on Machine Learning in Communications and Networking</em>.
All data files have been compressed due to space constraints on this server.
Usernames and exact timestamps have been removed from the mobile data for privacy reasons.
<h1> Getting started </h1>

The dataset can be divided into mobile and static measurements.
Mobile measurements consist of passive, latency and data rate measurements.
They were conducted with Android phones in vehicles around Dortmund, Germany (folder: mobile) and a dedicated modem on a suspension railroad track at TU Dortmund University (folder: mobile hbahn).

While the static measurements only contain passive and latency measurements, they are taken over an extended period of time and cover all times of day (folder: static).

<h2> File Format </h2>

The files are stored in human-readable .csv files.
Semicolons (";") are used as separators.
Common spreadsheet software like 'MS Excel' or 'Libre Office Calc' can open these files.
However, the size of the files may affect performance.

The associated article describes and analyzes the data in detail (see "Acknowledgments").
CSV columns and data types used are described in the included REAME.md file.

<h3>Further Information about the Data</h3>
Due to privacy concerns, the exact time and measurement user is stripped of the mobile measurements with android phones.

In the case of static measurements, there is a separate table specifying the location of each static measurement.

Data of neighboring cells is indicated with the suffix "_neighboring".
Neighboring measurement columns are a subset of the cell data columns.

<h2>Load the Data with Python3</h2>

We recommend the use of *Python3* to work with the data.
With the help of the Python library *'pandas'*, you can load the data as follows:

<pre><code>import pandas as pd

dataframe = pd.read_csv("filename.csv", sep=";")
# Print general information about the dataframe and its columns
dataframe.info()</code></pre>

You can use <em>pip</em> to install <em>pandas</em>.
<pre><code>python3 -m pip install pandas</code></pre>

<h1>Support</h1>
If you have any issues with the supplied data, please feel free to contact one of the article's authors.

<h1>Acknowledgments</h1>
The data acquisition was funded by the Federal Ministry of Transport and Digital Infrastructure (BMVI) in the context of the project <em>Virtual integration of
decentralized charging infrastructure in cab stands</em> (VIZIT) under the funding reference 16DKVM006B and by the Ministry of Economic Affairs, Industry, Climate Action and Energy of the state of North Rhine–Westphalia (MWIKE NRW) under the grant number 005–01903–0047.

If you use the data or the <em>ConMon Android</em> application, please cite the TMLCN article:

<pre><code>@Article{Schippers.etal/2025a,
Author = {Hendrik Schippers and Melina Geis and Stefan B{\"o}cker and Christian Wietfeld},
Title = {{DoNext: A}n Open-Access Measurement Dataset for Machine Learning-Driven {5G} Mobile Network Analysis},
Journal = {{IEEE} Transactions on Machine Learning in Communications and Networking},
Volume = {3},
Number = {},
Pages = {585-604},
Year = {2025},
Month = apr,
Doi = {10.1109/TMLCN.2025.3564239},
Keywords = {5G new radio; 6G; dataset; machine learning; predictive QoS; multi-MNO; channel modelling; transfer learning},
Project = {6GEM, CC5G.NRW, VIZIT},
} </code></pre>