Mehdi Bennis

TL;DR: This deliverable provides an overview of the WINNER+ system concept which encompasses the following innovations areas for IMT-Advanced technologies: advanced RRM, spectrum technologies, advanced co-located antennas and coordinated multipoint systems.

TL;DR: In this paper, the authors analyzed the main factors that influence the environmental footprint of federated learning policies compared with traditional centralized learning algorithms running in data centers and proposed an analytical framework taking into account both learning and communication energy costs, as well as the carbon equivalent emissions.

TL;DR: This paper presents a time-triggered FL algorithm (TT-Fed) over wireless networks, which is a generalization of classic synchronous and asynchronous FL, and jointly considers the user selection and bandwidth optimization problem to minimize the FL training loss.

TL;DR: A novel offline DAC scheme is put forward, which estimates the optimal control policy from a previously collected dataset without any further interactions with the system, demonstrating strong robustness to dataset quality.

TL;DR: In this paper , a graph neural network (GNN) based learning algorithm is proposed to directly map channel coefficients to the optimized network parameters, and the convergence analysis illustrates the detrimental impact of the accumulated aggregation error over all rounds.