Multi-group multicast beamforming in cognitive radio networks via rank-two transmit beamformed Alamouti space-time coding

In this paper, we consider transmit design in multiple-input single-output (MISO) multi-group multicast (MM) cognitive radio (CR) systems. Previously, semidefinite relaxation (SDR)-based transmit beamforming has been very successful in transmit design. However, recent research shows that further performance gain is possible by suitably modifying the transmit structure. Here, we propose a transmit beamformed Alamouti space-time code scheme for MM-CR systems, whose corresponding transmit design problem can be reformulated as a rank-2 constrained fractional semidefinite program. We then develop an SDR framework for this scheme and study its signal-to-interference-and-noise ratio (SINR) performance via both theoretical analysis and simulations. Specifically, we show that the worst-case approximation accuracy of the proposed scheme scales on the order of √M_S log M_P, where M_P (resp. M_S) is the number of primary (resp. secondary) users in the CR network. This unifies and generalizes a number of results in the literature and is, to the best of our knowledge, the first provable bound on the performance of a beamforming scheme in a general MM-CR system. Finally, simulation results show that our proposed scheme indeed has a better performance in both MM and MM-CR scenarios than the traditional beamforming scheme.