A decentralized method for joint admission control and beamforming in coordinated multicell downlink

In cellular networks, admission control and beamforming optimization are intertwined problems. While beamforming optimization aims at satisfying users' quality-of-service (QoS) requirements or improving the QoS levels, admission control looks at how a subset of users should be selected so that the beamforming optimization problem can yield a reasonable solution in terms of the QoS levels provided. However, in order to simplify the design, the two problems are usually seen as separate problems. This paper considers joint admission control and beamforming (JACoB) under a coordinated multicell MISO downlink scenario. We formulate JACoB as a user number maximization problem, where selected users are guaranteed to receive the QoS levels they requested. The formulated problem is combinatorial and hard, and we derive a convex approximation to the problem. A merit of our convex approximation formulation is that it can be easily decomposed for per-base-station decentralized optimization, namely, via block coordinate decent. The efficacy of the proposed decentralized method is demonstrated by simulation results.