Aprendizagens Direta e Indireta para a Identificação de Pré-distorsores de Largura de Banda Limitada

This work addresses the linearization of power amplifiers for wireless communication systems through band- limited digital baseband predistortion (DPD). This work contribution is to present a comparative study between the direct and indirect learning architectures for the parameter identification of band-limited DPDs. Simulation results from a case study show that the indirect learning offers a linearization capability very close to the one offered by the direct learning, but demanding for a significantly lower computational cost.