Power amplifier behavioral modeling based on LaguerrePolar‐Volterraseries with independent truncations

This article proposes an approach that modifies the Polar-Volterra series expanded by a set of Laguerre Orthonormal Basis Functions and the Polar-Volterra series using independent truncation techniques to significantly reduce the amount of parameters, where the resulting models reproduce the distortions of the nonlinearities of a radio frequency power amplifier (RF PA). An RF PA GaN HEMT class AB, stimulated by a wave centered at 900 MHz modulated by two 3GPP WCDMA envelope signals, each having a bandwidth of 3.84 MHz and shifted by 5 MHz, was used. A comparison of several models showed that the two proposed models, both containing a larger number of independent truncation factors, can provide a reduction over 50% in the number of generated coefficients, reaching a maximum of 84%. In another comparison with the other models, the modified Polar-Volterra series expanded by a set of Laguerre functions significantly improved the modeling accuracy, quantified by improvements of normalized mean square error of up to 6.51 dB.