Scheduling and power control in statistical beamforming networks using B bits of feedback

In this work, we develop limited feedback techniques that utilize both Channel State Information (CSI) and Channel Distribution Information (CDI) for communication in multiuser MIMO beamforming networks with SINR constraints on the links. We minimize power in the network using a CDI-based algorithm, and then use limited CSI feedback to improve on this scheme by reducing power consumption further and transmit opportunistically. We study three cases of the CSI feedback channel-with one bit of bandwidth, with infinite bits, and with B bits. We develop feedback techniques for the one-bit and infinite-bit cases, and then derive the optimal quantizer for the B-bit case. Our results show that significant power reduction can be achieved using a small number of bits.