A Comparison of Surrogate Behavioral Models for Power Amplifier Linearization under High Sparse Data

• Published in: Sensors (Basel, Switzerland) Vol. 22; no. 19; p. 7461
• Main Authors: Galaviz-Aguilar, Jose Alejandro, Vargas-Rosales, Cesar, Cárdenas-Valdez, José Ricardo, Aguila-Torres, Daniel Santiago, Flores-Hernández, Leonardo
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 01.10.2022; MDPI
• Subjects: Accuracy; Algorithms; Amplifiers, Electronic; behavioral modeling; Comparative analysis; Complexity; Decision trees; Digital integrated circuits; Distribution functions; Equipment Design; Field programmable gate arrays; Linearization; linearization model; Machine learning; Methods; Model accuracy; Modelling; Neural networks; nonlinear modeling; Orthogonal Frequency Division Multiplexing; power amplifier; Power amplifiers; Quadrature amplitude modulation; Receivers & amplifiers; Mexico; power amplifier; nonlinear modeling; behavioral modeling; machine learning; linearization model

A good approximation to power amplifier (PA) behavioral modeling requires precise baseband models to mitigate nonlinearities. Since digital predistortion (DPD) is used to provide the PA linearization, a framework is necessary to validate the modeling figures of merit support under signal conditionin...