Behavioral modeling for concurrent dual-band power amplifiers using 2D hammerstein/wiener models

Abstract

Behavioral modeling for the concurrent dual‐band power amplifier (PA) is a critical problem in practical applications. The nonlinear distortion in the concurrent dual‐band PA is quite different from that in the conventional single‐band PA. This article analyzes the nonlinearities in the concurrent dual‐band PA and reveals that both input signals in the dual bands are important for the behavioral modeling. The 2D Hammerstein model and 2D Wiener model are proposed for the first time for the concurrent dual‐band PA. They are extended versions of conventional Hammerstein and Wiener structures used in the single‐band PA by including the cross‐band intermodulation in the static nonlinearity block. The proposed 2D models require much less coefficients than the original work of the 2D‐DPD model. Experiments were carried out for an 880 MHz/1960 MHz concurrent dual‐band Doherty PA to demonstrate the effectiveness of the proposed models. The results clearly show that less than −40 dB normalized mean square errors (NMSEs) are obtained in the dual bands in the behavioral modeling. © 2012 Wiley Periodicals, Inc. Int J RF and Microwave CAE 23: 646–654, 2013.

Dynamic deviation reduction-based concurrent dual-band digital predistortion

“…The 2D-DPD model, the 2D-MMP model, the DB-DDR model, and the SDB-DDR model were tested for comparison. And the normalized mean square error (NMSE) and adjacent channel power ratio (ACPR) were considered as the evaluation criteria to evaluate the model accuracy and DPD performance [7,12,18]. Specifically, in concurrent mode, NMSE between measured and estimated signals in each band is defined as NMSE i 510 log 10 where i 2 1; 2 f g is the index used to indicate variables in LB (i51) or UB (i52) again, y i meas n ð Þ and y i est n ð Þ are the measured and estimated signals in the ith band, and N is the sampling length of signals.…”

“…The authors in [11] substituted two summations for three ones, and derived the low-complexity 2D-DPD model, which can achieve comparable modeling accuracy as the 2D-DPD model with fewer coefficients. They also proposed a 2D Hammerstein/Wiener model [12], a 2D augmented Hammerstein (2D-AH) model [13], and a 2D modified memory polynomial (2D-MMP) model [14]. The 2D Hammerstein and 2D Wiener models are special cases of the 2D-DPD model, by dividing the static nonlinearity and the memory effects into multiple subsystems in cascade and including the crossband intermodulation in the static nonlinearity block.…”

Orthonormal basis functions for memory predistorter in oversampled systems

“…Once we have the oversampling rate S, the closedform expression of E½W H W can be obtained. Thus, the new upper triangular matrix U new can be solved by substituting (22) and (23) into (21).…”

Volterra series-based model for concurrent dual-band power amplifier using dynamic memory depth

“…In these methods, the concurrent dual‐band signal is regarded as a broadband signal with an extremely large bandwidth, then substitute the dual‐band input into the single‐band model directly. Finally the dual‐band model can be obtained by analyzing the components of output expressions . The effectiveness of these models has been demonstrated by some specific experiments.…”

“…Finally the dual-band model can be obtained by analyzing the components of output expressions. [2][3][4][5][6] The effectiveness of these models has been demonstrated by some specific experiments. However these prove only that these models can describe the main nonlinear characteristics of the concurrent dual-band transmitter.…”