RLS-DPD Algorithm for Hybrid Precoding Architecture in MIMO-OFDM systems

In emerging communication systems, such as 5G 6G, millimeter wave and terahertz communication, OFDM and MIMO are indispensable technologies. However, the cross-modulation among different subchannel would greatly degrade the performance of system due to the non-linearity of the large-scale amplifier array. Hybrid precoding technology is a general candidate for this problem due to its capability for linearization, and the predistortion matrix would be the key factor for its performance. In this paper, a memory polynomial model is adopted for modelling nonlinear power amplifier (PA) and Recursive Least Square based digital predistortion (RLS-DPD) algorithm is proposed for parameter training for hybrid predistortion module. The proposed RLS algorithm extracts the parameters of the digital predistortion module through adaptive learning in the training phase. Adaptive learning refers to the process of using the filter parameters obtained at the previous moment to automatically adjust the parameters at the current moment according to the estimated error, which minimizes the cost function. The simulation results show that the RLS-DPD algorithm can reduce the system bit error rate by 27.778
^%
, error vector magnitude (EVM) by 67.961 % at 8dB SNR, and enable the nonlinear OFDM signal to be demodulated correctly at the terminal.