New beamforming method for two-way relay networks

Multiple Accesses Broadcasting (MABC) is a spectral efficient protocol in two-way relay channel (TWRC). The drawback of this protocol in decode-and-forward (DF) mode is decoding complexity at the relay´s side. Erroneous decoding at the relay can bring about error propagation that leads to performance degradation in all of the relay networks operating in the DF mode. In this paper, we propose ML detection at the multiple antenna relay. With analytical investigation, we prove that this decoding rule can achieve full diversity order at the MAC phase of the MABC protocol. Then, we propose a new beamforming algorithm named as “Channel Alignment” and show that the proposed algorithm will result in better performance compared with the well-known MAX-MIN antenna selection method while it has full diversity order at the BC phase of the MABC protocol. The analytical results are verified through simulations and the performance of our proposed method is also compared with another method.