On the Performance of Receive ZF MIMO Broadcast Systems with Channel Estimation Errors

While the zero-forcing (ZF) transmit beamforming is a widely used technique for realizing multi-input multi-output (MIMO) broadcast transmissions, the ZF receiver combined with multiuser scheduling is an effective alternative that yields improved robustness against CSI mismatch caused by feedback link errors. In this paper we consider the practical scenario that channels are imperfectly estimated at the receiver. Our goal is to characterize the sum rate performance of the receive ZF MIMO broadcast systems under channel estimation errors. We derive analytic sum-rate expressions for both the uniform transmit power and transmit water-filling cases. Our analytic results characterize the sum-rate floor incurred by channel estimation errors. Numerical simulations are used to confirmed the analytic study.