The channel state information (CSI) at base station (BS)

obtained from users in the multiuser multiple input multiple output (MU-MIMO) system

leads to precoding performance degradation in frequency division duplexing (FDD) system.Existing robust beamforming precoding algorithms in MU-MIMO can reduce the performance loss caused by imperfect CSI.However

these algorithms only take one or two robustness channel errors into consideration

as a consequence

the system performance is limited to be improved.By establishing joint channel error model including channel estimation error

quantization error and delay error

we derive the closed-form solution of robust beamforming matrix based on minimum mean square error (MMSE).And then

we derive the closed-form solution of robust beamforming matrix based on MMSE of signal leakage with distributed characteristics while applying to distributed communication system.When compared to the traditional MMSE algorithm which only considers quantization error

numerical analysis shows that

the proposed centralized and distributed downlink MU-MIMO beamforming algorithms have better sum rate and bit error rate (BER)

and reduce the precoding performance loss caused by the above three channel errors.