Joint Air-Ground Distributed Federated Learning for Intelligent Transportation Systems

Implementation and Evaluation of a Federated Learning Framework on Raspberry PI Platforms for IoT 6G Applications

“…ML technology has gained a lot of attention for enabling intelligent solutions in wireless networks including mobile communication networks [9], wireless sensor networks [10], transportation systems [6], non-terrestrial networks [11]. Complex wireless communication problems such as resource management [12], data offloading toward edge networks [13], spectrum management [14], routing [15], user-server allocation [7], etc., can effectively solve through different ML techniques. Among others, FL is widely considered a distributed learning approach to provide efficient learning solutions.…”

“…Several forms of distributed learning, such as Federated Learning (FL), collaborative learning, split learning, and multi-agent reinforcement learning, are widely considered in wireless networks for enabling intelligence-at-the-edge. Among others, FL has achieved great success in building high-quality ML models based on dispersed wireless data [6,7]. FL is also a candidate for next-generation 6G communication standard allowing for setting up an intelligence-at-the-edge framework [8].…”

Integration of Terrestrial and Non-Terrestrial Networks for Automotive: challenges and perspectives within the S11 RESTART project

“…When a terrestrial ITS is also extended toward air and space layers additional solutions should be taken into account trying to leverage on such a complex environment composed of a multitude of distributed nodes. With this in mind, some recent solutions consider the possibility of implementing distributed learning frameworks in T/NT environment where the several nodes around vehicular users can be efficiently exploited for implementing machine learning algorithms [17].…”