Cooperative Beamforming for Multi-User Multi-Relay Networks Based on Second-Order Statistics of Channel State Information

This paper investigates linear beamforming techniques in relay networks with one source and multiple relays and destinations. The goal is to determine the beamforming matrix so that the signal to noise plus interference ratio at the destination is maximized while respecting power constraints. The considered power constraint here is a total (source and relays) maximum constant power. The key point of this paper is that we propose to use only the second order statistics of channel state information which is a more realistic scenario than the case where instantaneous channel coefficients are assumed to be perfectly known by the relays. We show that, despite some losses compared to the case of zero forcing using perfect channel information, second order statistics based beamforming algorithms are good candidates to reliably support multiple parallel data streams with signal to noise plus interference requirements in multiuser multi-relay systems.