Multi-Level Zero-Forcing Method for Multiuser Downlink System with Per-Antenna Power Constraint

Multiuser downlink beamforming methods have been studied recently. As known, the MIMO broadcast channel capacity can be achieved by dirty paper coding (DPC). However, the high complexity of DPC is hard to implement. Linear zero-forcing beamforming strategy can achieve the same asymptotic sum-rate as that of DPC, when the number of users goes to infinity. For the linear zero-forcing system, (D. Bartolome et al., 2004) gives several power allocation methods among users for different objectives. The above power allocation has an assumption of sum power constraint (SPC), however, recently some researcher study a more practical power constraint name per-antenna power constraint (PAPC). The optimum power allocation aims for maximize the sum-rate under the PAPC is given in (F. Boccardi et al., 2006). In this paper, we find that if the transmit antenna number M is larger than the user number K, a multi-level zero-forcing method can be adopted, and a multi-layer structure is proposed for realizing the method. A sub-optimal power allocation algorithm is also proposed to find the power allocation matrix. Simulation result shows that the proposed multi-level zero-forcing method can achieve a higher sum-capacity than that of (F. Boccardi et al., 2006) under PAPC.