Cooperative TDD Reciprocity Calibration with Diversity Combining for Distributed Large-Scale MIMO Systems

Thanks to channel reciprocity, the time-division duplexing operation is typically used for distributed large-scale multiple-input multiple-output systems. With the knowledge of uplink channel state, APs (Access Points) perform the downlink precoding to cooperatively communicate with multiple UEs (User Equipments). Unfortunately, the channel reciprocity is often jeopardized by the non-symmetric transceiver RF (Radio Frequency) circuits at both sides of the link. By excluding UEs from the calibration procedure, the TLS (Total Least Squares) method is extended to the case of partial calibration, where only the APs are involved to exchange calibration signals to compensate for RF mismatches. Nevertheless, channel fluctuations between APs significantly degrade system performance. Therefore, we propose a scheme to obtain the diversity gain, by performing a calibration with the combination of the calibration signals of different channel coherent times. Simulation results confirm the effective performance and robustness of our scheme.