UE Statistics Time-Series (CQI) in LTE Networks

Citation Author(s):
Ilias
Chatzistefanidis
University of Thessaly
Nikos
Makris
University of Thessaly, CERTH Greece
Virgilios
Passas
University of Thessaly, CERTH Greece
Thanasis
Korakis
University of Thessaly, CERTH Greece
Submitted by:
Ilias Chatziste...
Last updated:
Thu, 09/29/2022 - 10:11
DOI:
10.21227/ec7p-xq38
Data Format:
Links:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

This dataset includes real-world Channel Quality Indicator (CQI) values from UEs connected to real commercial networks in Greece. In total, we collected CQI data from 74 cars that drive through a specific road in the city of Volos, Greece. This dataset is part of our following work:

 

As wireless networks become denser and more heterogeneous different paths can be considered in order to reach each multi-homed UE, offering optimal performance. 5G and beyond networks feature contributions related to the dynamic programming of the network, from the operator side, in order to optimally allocate resources in the network. In our work, we consider such a case, where network access is provided to the end-users via heterogeneous (3GPP and non-3GPP) Distributed Units (DUs), converging to a single Central Unit (CU), and programmable on the fly with external interfaces. We employ Machine Learning (ML) methods in order to forecast the Quality of Service (QoS) that a wireless client will get from the network in the near future based on the Channel State Information (CSI) metric. Subsequently, we appropriately steer the traffic over the different heterogeneous DUs for ensuring that the network meets the needs of the UEs. We design, develop, deploy and evaluate our method in a real testbed environment, using emulated mobility. Our results show that the overall throughput of each UE can be drastically improved compared to existing allocation mechanisms. This work is based on the submitted data in this repository, where we collect Channel Quality Indicator (CQI) data from real commercial networks in city Volos, in Greece. Specifically, we create a dataset with CQI data from 74 cars driving through a specific road in the city.

Instructions: 

To utilize this dataset effiently, we propose using the respective github repository: https://github.com/ilias-chatzistefanidis/HetNets-steering

In this repo you can find:

  • Data Folder: In this foler, we have the CQI data collected from real cars in a route in city Volos, Greece. There are two files, the train.csv which contains CQI data from 73 cars and used as training data. The other file is the test.txt which contains the CQI data from the experiment in our experimental environment.
  • Figures Folder: In this folder, we have several important figures that are used by the notebook.
  • Notebook Folder: In this folder, we have the notebook that includes the utilized machine learning pipeline used to develop a Bi-LSTM model. It is a reproducable notebook that is easily executed in Google Colab or locally. In the notebook, you can find useful functions and information on the pre-processing, deep learning model development, evaluation and extra stuff.
  • utils.py file: This file includes the functions developed in the notebook for easier usage.