Robust Bayesian Precoding for Mitigation of TDD Hardware Calibration Errors.

Samer Bazzi,Wen Xu
DOI: https://doi.org/10.1109/lsp.2016.2569199
2016-01-01
IEEE Signal Processing Letters
Abstract:The predominantly used channel reciprocity condition in time-division-duplex mode only holds in practical systems if the transmit and receive hardware chains at the base station (BS) and at the user equipments are perfectly calibrated. Otherwise, there is a mismatch between the uplink (UL) and downlink channel state information (CSI) that leads to performance degradations. In this paper, we consider a multi-userMIMO system with hardware calibration errors (CaEs) at the BS and derive robust Bayesian precoders that minimize inter-user interference using the UL CSI and CaEs statistics as side information. The proposed robust precoding outperforms conventional non-robust precoding, and its gains compared to non-robust precoding increase with increasing numbers of users and CaEs. Furthermore, robust precoding can be used to relax calibration requirements, which is a result of practical interest. Our approach is applicable to any number of BS antennas but is especially important for massive MIMO systems where larger CaEs might be expected.
What problem does this paper attempt to address?