Fiche du document
13 septembre 2021
- handle: 10670/1.cfls9v
- isbn: 978-1-7281-7586-7
- cea-04549323
- ARXIV: 2106.02540
- doi: 10.1109/PIMRC50174.2021.9569681
Ce document est lié à :
info:eu-repo/semantics/altIdentifier/arxiv/2106.02540
Ce document est lié à :
info:eu-repo/semantics/altIdentifier/doi/10.1109/PIMRC50174.2021.9569681
Ce document est lié à :
info:eu-repo/grantAgreement//826276/EU/H2020 Project CPS4EU, Nr. 826276/CPS4EU
Ce document est lié à :
info:eu-repo/grantAgreement//101017011/EU/H2020 RISE-6G Project, Nr 101017011./RISE-6G
info:eu-repo/semantics/OpenAccess
Mots-clés
Radio resource management Base stations 5G mobile communication Reinforcement learning Network architecture Complexity theorySujets proches
Group dynamics Association Groups, SocialCiter ce document
Mohamed Sana et al., « Transferable and Distributed User Association Policies for 5G and Beyond Networks », HAL-SHS : sciences de l'information, de la communication et des bibliothèques, ID : 10.1109/PIMRC50174.2021.9569681
Métriques
Partage / Export
Résumé
We study the problem of user association, namely finding the optimal assignment of user equipment to base stations to achieve a targeted network performance. In this paper, we focus on the knowledge transferability of association policies. Indeed, traditional non-trivial user association schemes are often scenario-specific or deployment-specific and require a policy re-design or re-learning when the number or the position of the users change. In contrast, transferability allows to apply a single user association policy, devised for a specific scenario, to other distinct user deployments, without needing a substantial re-learning or re-design phase and considerably reducing its computational and management complexity. To achieve transferability, we first cast user association as a multi-agent reinforcement learning problem. Then, based on a neural attention mechanism that we specifically conceived for this context, we propose a novel distributed policy network architecture, which is transferable among users with zero-shot generalization capability i.e., without requiring additional training. Numerical results show the effectiveness of our solution in terms of overall network communication rate, outperforming centralized benchmarks even when the number of users doubles with respect to the initial training point.