Analysis of UAV communications in cell-free massive MIMO systems

We study support for unmanned aerial vehicle (UAV) communications through a cell-freemassive MIMO architecture, wherein a large number of access points (APs) is deployed in place of largeco-located massive MIMO arrays. We consider also a variation of the pure cell-free architecture by applyinga user...

ver descrição completa

Detalhes bibliográficos
Autores: D’Andrea, Carmen, García Rodríguez, Adrián, Geraci, Giovanni, Galati Giordano, Lorenzo, Buzzi, Stefano
Formato: artículo
Estado:Versión publicada
Fecha de publicación:2020
País:España
Recursos:Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)
Repositorio:Recercat. Dipósit de la Recerca de Catalunya
OAI Identifier:oai:recercat.cat:10230/43476
Acesso em linha:http://hdl.handle.net/10230/43476
http://dx.doi.org/10.1109/OJCOMS.2020.2964983
Access Level:acceso abierto
Palavra-chave:Cell-free massive MIMO
Ricean fading channel
Spectral efficiency
Power allocation
UAV communications
User-centric
Descrição
Resumo:We study support for unmanned aerial vehicle (UAV) communications through a cell-freemassive MIMO architecture, wherein a large number of access points (APs) is deployed in place of largeco-located massive MIMO arrays. We consider also a variation of the pure cell-free architecture by applyinga user-centric association approach, where each user is served only from a subset of APs in the network.Under the general assumption that the propagation channel between the mobile stations, either UAVs orground users (GUEs), and the APs follows a Ricean distribution, we derive closed form spectral efficiencylower bounds for uplink and downlink with linear minimum mean square error channel estimation. Weconsider several power allocation and user scheduling strategies for such a system, and, among these,also minimum-rate maximizing power allocation strategies to improve the system fairness. Our numericalresults reveal that cell-free massive MIMO architecture and its low-complexity user-centric alternativemay provide better performance than a traditional multi-cell massive MIMO network deployment.