Real-Time Optimization of Dynamic Predistortion Model for 5G Massive MIMO Broadband RF Power Amplifier

This letter proposed a real-time single-channel digital predistortion (DPD) architecture for multiple-input multiple-output (MIMO) transmitters. A two-stage beam search algorithm is used to linearize millimeter-wave (mm-Wave) beamforming emitters with dynamic configurations. Instead of using the same model structure for all power levels, the model structure can be adjusted dynamically and flexibly. Determine the maximum amount of energy in the group, and then search for the best combination of models with in the best beams. The improved beam search method is used to further reduce the model complexity within the combination. In order to verify the effectiveness of the proposed method, a mm-Wave beamforming transmitter with beamable angle experiment and eight-channel input power is configured. Simulation and experimental results show that the proposed algorithm can quickly find the optimal model structure when conducting beam search and model matching. It also maintains robust linearization performance with relatively low complexity.