Recurrent Neural Networks for Handover Management in Next-Generation Self-Organized Networks

Multi-Task Learning for Efficient Management of Beyond 5G Radio Access Network Architectures

“…As a result, the third approach to HO management, which is also the one considered in this paper and our previous work [16], is a data-driven approach. It aims at using experience extracted from network data to include the vision of long-term optimization in the HO management decision.…”

“…A preliminary and partial version of this work was presented in [16], and [17]. In [16], we focused only on the single task HO management use case, and used an LSTM RNN to solve the regression problem and estimate the QoE of the users. The obtained results proved that the learning approach outperforms traditional HO solutions.…”

“…• We present a holistic solution to handover based on download time that is not limited to adjusting typical HO parameters but considers previous experience to select a target next-Generation Node B (gNB) to improve users' QoE. • Extending our previous work in [16], we introduce an additional RAN use case, i.e., initial MCS selection.…”

Machine Learning: A Catalyst for THz Wireless Networks

“…This is where the conventional heuristic based exploration of state space needs to be extended to support UE mobility in an online manner. Aspired by this, several contributions presented ML-based mobility management solutions that aim at accurately tracking the UE and proactively steering the AP and UE beams (Burghal et al, 2019;Guo et al, 2019) as well as performing handovers between beams and APs/BSs (Yan et al, 2019;Ali et al, 2020). In particular, in (Burghal et al, 2019), a RNN with a modified cost function that takes as input the observed received signal as well as the previous AoA estimation, was employed to track the AoA in a mmW network.…”

Distributed and Multi-Task Learning at the Edge for Energy Efficient Radio Access Networks

“…Regarding the simulation scenario, we consider 7 threesectorial BSs, which corresponds to 21 sectors, and 10 UEs per sector, which results in a total of 210 UEs in the whole simulation field. For an exhaustive description of the simulation scenario and parameters, the reader can refer to [49]. During the simulations, it may happen that some of the data are not available, because UEs might experience a Radio Link Failure (RLF) when forced to handover to a BS with poor channel conditions.…”