Limited Feedback Schemes for Multiuser MIMO Downlink Transmission

Abstract

The major problem of multiuser MIMO precoding and scheduling is the required Channel State Information (CSI) at the base station which needs to be provided to the base station via a feedback channel in case of frequency division duplex. Each user quantizes the channel direction information and computes a Channel Quality Indicator (CQI), e.g., based on the Signal-to-Interference-and-Noise Ratio (SINR), both fed back as CSI to the base station. Since the precoder is not known when the feedback information is computed, the SINR needs to be approximated where the quality of the approximation depends strongly on the number of finally scheduled data streams. In this paper, we propose limited feedback schemes which improve the performance of the scheduler by exploiting this dependency and adapting the feedback information accordingly. Three types of feedback schemes are presented: first, feeding back more than one approximation such that the base station can choose the more accurate one depending on the number of finally scheduled data streams, second, a time multiplexing of these approximations in order to decrease the feedback overhead, and third, feeding back the channel magnitude and estimating the approximations via additional statistical measures. Simulation results show that compared to state-of-the-art techniques, the proposed feedback schemes perform well for a wide range of correlation scenarios.