Interference mitigation by jointly splitting rates and beamforming for multi-cell multi-user networks

The paper considers the joint design for splitting bit rates in private and common message transmissions and coordinated beamforming in a multi-cell multi-user multiple input single output (MISO) system. Namely, the users are purposely paired to decode the common messages to subtract interferences. Both the private messages and common messages are further beamformed before the transmission to maximize the worst user's rate performance of the overall system. The common messages and private messages are decoded in sequence. Obviously, users pairing optimization alone already is a nontrivial combinatorics, while beamforming private and common messages also pose a difficult nonconvex optimization problem. Their jointly optimal design is thus a hard mixed combinatoric optimization problem. Nevertheless, we show that this joint design can be successfully addressed by means of d.c (difference of convex functions) optimization. Numerical simulation shows the advantage of our approach over the previous one.