Skip to main content
Publications lead hero image abstract pattern


Publication Date

Manuscript Submission Deadline

Feature Topic

Call for Papers

Submit a Paper

In the era of 5G and 6G, the telecommunication industry has evolved to provide services for new types of Internet of Things (IoT) devices in addition to basic mobile phone or internet services, including extended reality devices, sensors, and ground and aerial robots, to name a few. With the deployment of these new services,  it is difficult for the wireless network to support ubiquitous connections with diverse quality-of-service (QoS) requirements. Despite the remarkable success of the model-based design and analysis for wireless networks, it turns out that they are not always adequate for dynamic wireless environments with diverse QoS requirements. To address this, the data-driven methods of Machine Learning (ML) is expected to enable fundamentally new intelligent design and decision-making in wireless networks.  The intelligence supported by ML solutions will allow for the pattern recognition from time-series data, the network anomalies detection and prediction, the network design automation, and performance optimization in real-time, hence creating a self-optimizing and self-updating networks.

However, the research advancements with ML for wireless communications and networks inherently relies on the availability of data sets to test the results and attempts generalizations. One of the main bottlenecks for such research is the current limited availability of datasets from either practical simulations or experimental testbeds that can be considered as reference or standard data sets. Creating standard reference datasets for research purposes, ranging from low-level physical layer measurements to telecommunication network analyses, is expensive, especially for research organizations, while the commercial datasets from telecommunication operators are mostly inaccessible. Thus, in this Special Issue, we aim to attract manuscripts that propose data sets for experimental testbeds and ML methods in wireless communications and networking, which have the potentiality to become reference or standard data sets for research purposes. These manuscripts should encourage growth in the use of experimental methods and the use and analysis of data sets.

All submissions must be based on high-quality research that has already been published or accepted in a peer-reviewed venue, and this must be clearly indicated. Thus, this submission does not need to verify the underlying research. Instead, a submission should focus on the description of the experimental setup and the value of the considered public available data set.

This Special Issue submission must provide a description of the necessary lab equipment and the steps for performing a physical experiment.  Experiments do not need to be described at the level of a user manual, but should have substantially more details than any previous work using the same setup and the level of detail should enable the work to be reproducible by someone who is not an expert in the area.

An original data set from the experiment must be provided. This data set should be of demonstrable value for the IEEE communications community. There should also be a brief example(s) of how to use the data, and this should also be novel (and in particular distinct from previous works that used the method). Data sets should be placed on IEEE DataPort with open access (open access uploads are currently free for IEEE members) and be available as part of the review process. Complementarily, data and code can be uploaded to CodeOcean.

As a sample submission, please refer to the following published paper.

AI/ML Data sets for

  • 5G/6G testbeds and trials
  • distributed AI/ML over communication networks
  • distributed multi-agent reinforcement learning
  • edge learning in wireless networks
  • federated learning and communications
  • integrated sensing and communication
  • intelligent reflecting surfaces 
  • learn to transmit and receive
  • link layer
  • MAC layer
  • mobility and network management
  • molecular networks
  • optical networks
  • over-the-air computation
  • physical layer
  • privacy and security issues
  • resource management and network optimization
  • semantic communications
  • wireless communications and networks to support AI/ML services

Submission Guidelines

Manuscripts should conform to the standard format as indicated in the Information for Authors section of the Manuscript Submission Guidelines. Please, check these guidelines carefully before submitting since submissions not complying with them will be administratively rejected without review.

All manuscripts to be considered for publication must be submitted by the deadline through Manuscript Central. Select the “FT–2220 / Data Sets for Machine Learning in Wireless Communications and Networks” topic from the drop-down menu of Topic/Series titles. Please observe the dates specified here below noting that there will be no extension of submission deadline.

Manuscripts should be submitted through our Author Portal by selecting the Submit a Paper option at the top of this page:

Important Dates

Manuscript Submission Deadline: 16 November 2022
Decision Notification: 28 February 2023
Final Manuscript: 31 March 2023
Publication: May 2023

Guest Editors

Carlo Fischione (Lead Guest Editor)
KTH Royal Institute of Technology, Sweden

Marwa Chafii
New York University Abu Dhabi, UAE

Yansha Deng
King’s College London, UK

Melike Erol-Kantarci
University of Ottawa and Ericsson, Canada