Beam-Dependent Active Array Linearization by Global Feature-Based Machine Learning

An approach based on machine learning is proposed for the global linearization of microwave active beamforming arrays. The method allows for a low-complexity real-time update of the digital predistortion (DPD) coefficients by exploiting order-reduced model features, hence avoiding the need of repeated local DPD identification steps across the various operating conditions of the beamformer array (e.g., different beam angles or RF power levels). The validation is performed by over-the-air (OTA) measurements of a 1x4 beamforming array operating at 28 GHz across 100-MHz modulation bandwidth.