par Lebrun, Yann;Nsenga, Jimmy;Ramon, Valéry;Bourdoux, André 
;Horlin, François ;Lauwereins, Rudy
Référence (24-28 August, 2009: Glasgow, Scotland), Proc. 17th European Signal Processing Conference, EUSIPCO
Publication Publié, 2009-08
;Horlin, François ;Lauwereins, RudyRéférence (24-28 August, 2009: Glasgow, Scotland), Proc. 17th European Signal Processing Conference, EUSIPCO
Publication Publié, 2009-08
Publication dans des actes
| Résumé : | Multi-antenna wireless mesh networks are a promising means to achieve high data rate in next generation of wireless networks. However, because co-channel interference (CCI) is a major source of performance degradation in such networks, cooperative techniques have been proposed to cope with CCI. Nevertheless, these techniques require the exchange of information between the transmit nodes (e.g. channel knowledge acquisition, synchronization, etc...) and are thus very complex to implement. Conversely, non-cooperative techniques have limited complexity and are therefore an interesting alternative. Two non-cooperative methods have been proposed in the literature: the zero-forcing and the maximal signal-to-noise ratio beamformers. However, those two schemes do not maximize the aggregate capacity of the network nor minimize the bit error rate (BER) of each communication channel. In this paper, we consider a different optimization metric based on maximizing the SINR at the transmitters. We show that the proposed scheme achieves maximal capacity and outperforms the two existing non-cooperative schemes in terms of BER. |
