Nokia Corporation

Peter Fertl

Ari T Hottinen

In accordance with aspects of the present invention, a method, apparatus and system for learning antenna weighting factors in a communication system including relay nodes. In one embodiment, an apparatus (e.g., a relay node ()) for use with a communication system includes a first antenna () configured to receive a first signal including a pilot training sequence from a source node () and a second signal including a power-based feedback signal or a signal-to-noise based feedback signal from a destination node (). The apparatus also includes a second antenna () configured to transmit at least a portion of the first signal with an antenna weighting factor (e.g., a perturbed antenna weighting factor). The apparatus further includes an antenna weighting factor module () coupled to the first antenna () and configured to iteratively adjust the antenna weighting factor in response to the second signal.

System for distributed beamforming for a communication system employing relay nodes

Apparatus, methods and computer program products implement a channel-aware approach for controlling a relay network operative in a wireless communications system wherein weighting matrices are used at a MIMO-capable source electronic device and at the relays where at least one of the weighting matrices is selected or derived using knowledge of channel components of at least one channel link, where the channel link may be from the relays to the destination antennas of the destination device; or from source antennas of the MIMO-capable source device to the relays, or from the source antennas of the MIMO-capable source device to the destination antennas of the destination device. In one approach weighting matrices applied at the source and relay nodes are selected from a predetermined codebook of weighting matrices in dependence on monitored channel conditions. Alternatively, the weighting matrices may be determined using an eigenvalue decomposition and power allocation (e.g., water-filling) approach by using the effective channel correlation matrix possibly combined with the use of the noise correlation matrix.

Growing community of inventors

Vienna, Austria

Peter Fertl