Outage-Constrained Coordinated Beamforming With Opportunistic Interference Cancellation

In this paper, interference management is considered for the K-user multiple-input single-output (MISO) block-faded interference channel. It is assumed that perfect channel state information (CSI) can be obtained at the receivers, whereas only channel distribution information (CDI) is available to the transmitters. Furthermore, the receivers are assumed to be capable of implementing opportunistic interference cancellation (OIC). Based on these assumptions, the beamforming design problem for the MISO interference channel under consideration is formulated as a rate utility maximization problem under outage constraints and individual power constraints. By implementing OIC at the receivers, the outage probability is expected to be reduced, leading to a potential improvement on the utility value. However, this flexibility complicates the outage constraint, resulting in an optimization problem that is non-convex and difficult to handle. We propose a successive convex approximation (SCA) method that employs first-order approximation techniques and semi-definite relaxation (SDR) to efficiently obtain an approximate beamforming solution to this problem. Our proposed SCA algorithm has been shown to yield solutions that are stationary points of the original beamforming design problem. Furthermore, a decentralized version of the SCA algorithm is proposed, in which only local CDI and a limited amount of information exchange between transmitters are required. Our simulation results demonstrate that both of our proposed algorithms achieve near-optimal performance for the two-user case, and that enabling the receivers to implement OIC indeed improves the system performance by as large as 260% at high signal-to-noise ratios when user fairness is considered.