Low-Complexity Beam-Oriented Linearization Approaches for Massive MIMO Transmission

Digital beamforming is a crucial technology that enables fifth-generation (5G) devices to operate on millimeter wave (mmW) radio frequencies. However, linearizing massive arrays at the transmitter (TX) side remains a significant challenge as the complexity scales with the number of antennas. This work focuses on the orthogonal frequency division multiplexing (OFDM) modulation in 5G and beyond networks and presents low-complexity linearization approaches that can be used before the precoder. These approaches do not scale with the number of antennas and provide flexible control of the linearization performance across different parts of the spectrum. Simulation results for a 64-element uniform linear array (ULA) demonstrate that the proposed techniques can achieve performance comparable to conventional multi-digital pre-distortion (DPD) with only 8.91% and 3.52% of the complexity.