This paper investigates the problem of joint multiuser admission control and beamforming optimization for multiple-input single-output (MISO) heterogeneous networks (HetNet). Considered is a HetNet where multiple newly deployed femtocell base-stations (FBS) have the coverage overlapped with that of an existing macrocell base-station (MBS). The design objective is to serve as many femto-users (FUE) as possible at their quality-of-service (QoS) requirements, while maintaining the QoS requirements at the macro-user (MUE). This paper then proposes three algorithmic schemes to perform multiuser admission control and beamforming optimization based on the levels of coordination between the MBS and FBSs. In Scheme I with full MBS-FBS coordination, a joint optimization framework is presented, and two solution approaches are proposed to determine the admission control and beamforming design for the FUEs in a centralized manner.
In Scheme II with limited MBSFBS coordination, a distributed algorithm that allows each FBS to unilaterally determine its admission control and beamforming strategy is proposed.This algorithm requires certain coordination from the MBS by setting a limit on the amount of crosstier intercell interference (ICI) that can be generated by each FBS. The convergence of the proposed distributed algorithm to a fixed-point, where the QoS at the MUEs and the admitted FUEs is guaranteed, is then proved. Finally, in Scheme III without any MBS-FBS coordination, a joint multiuser admission control and zero-forcing beamforming design is then proposed. In particular, each FBS distributively admits its own FUEs while suppressing its cross-tier ICI to the MUEs, which effectively maintains the QoS at these MUEs. Simulation results show comparable performances between the distributed algorithms and the centralized ones in terms of number of FUEs served and the network power usage.