Sum-Rate Maximization in Distributed-Antenna Heterogeneous MIMO Downlinks: Application to Measured Channels

In a distributed antenna multiple-input multiple- output (MIMO) multiuser channel, multiple nodes simultaneously access the wireless channel by cooperating to achieve virtual MIMO links, enabling improved exploitation of the channel spatial dimensions. However, practical considerations dictate that in such a system, the transmit power from each cooperating node must be constrained in a per-antenna power constraint, which differs from the sum-power constraint typical of multiuser MIMO signaling. Furthermore, existing multiuser MIMO communication algorithms are appropriate for specific assumptions regarding the signaling strategy allowed. This paper provides a general framework able to accommodate a variety of multiuser MIMO channel topologies, precoding techniques (linear or nonlinear), and power constraints including the per-antenna constraint appropriate for the cooperative MIMO channel involving heterogeneous nodes. In contrast to prior work incorporating per-antenna power constraints, the algorithm enables reduced implementation and computational complexity by avoiding the requirement of numerical optimization. Simulation results using both modeled and measured cooperative channel responses demonstrate that the sum rate of a cooperative broadcast channel with a per-antenna power constraint is very close to the optimal bound achieved using a sum-power constraint, even in practical scenarios. Additional results demonstrate the straightforward application of the algorithm to different practical multiuser topologies.