Augmented Time-Delay Twin Support Vector Regression-Based Behavioral Modeling for Digital Predistortion of RF Power Amplifier

Abstract

In this paper, a dynamic behavioral model for digital predistortion (DPD) of RF power amplifier (PA) based on amplitude and phase augmented time-delay twin support vector regression (AP-TSVR) is proposed. Unlike other SVR-based methods, the TSVR model finds a pair of non-parallel planes by solving two related support vector machine (SVM) type problems, namely, the ε-insensitive up-and down-bound functions. Furthermore, in order to accelerate the training process, an effective linear regression algorithm was used to solve the paired quadratic programming problems (QPPs) of the TSVR model involved. The simulation results show that the proposed model is able to give improved modeling and distortion mitigation capability than the traditional memory polynomial-based model, and reduce CPU training time than the ordinary SVR model, even when the effects of both nonlinear characteristics and memory effects of PA are considered. To verify the effectiveness of the proposed method, experimental verification was performed using single-device gallium nitride (GaN) PA and GaN Doherty PA, respectively. The experimental results show that the new modeling approach can provide very efficient and extremely accurate linearization performance with improving generalization ability. INDEX TERMS Digital predistortion (DPD), radio frequency (RF) power amplifier (PA), twin support vector regression (TSVR), dynamic behavioral modeling.

A simplified adaptive sparse digital pre‐distorter for joint mitigation of frequency‐dependent transmitter impairments

“…As is well known, the Volterra series is a very general approach to model the nonlinear system with memory. In general baseband modeling system, only the odd‐order nonlinear terms of Volterra models are used in some published papers . However, some reported results reveal that the even‐order nonlinear terms of the Volterra series can enhance the pre‐distortion performance .…”

A new augmented support vector regression‐based behavioral model for multi‐device power amplifiers

“…However, when modeling the multi‐device Doherty PA, the accuracy remains insufficient. In Reference 4, an ASVR modeling technique is presented, and applied, as part of a single‐device Class F PA model, with an improvement in model accuracy reported. However, the accuracy is still limited when it is applied to multi‐device PAs, especially with strong memory effect and nonlinearity.…”

“…In order to address this issue, in this paper a new augmented modeling technique, distinct from Reference 4, is combined with the standard time‐delay SVR method.…”

“…I G U R E 4 AM/AM characteristics of the GaN Doherty PA from measurements and from the NASVR model [Color figure can be viewed at wileyonlinelibrary.com] F I G U R E 3 Output power spectra and spectra error of the basic SVR model,3 existing augmented SVR model,4 and the proposed NASVR model, compared with the measurement results on the multidevice GaN Doherty PA driven by a four-carrier long-term evolution (LTE) signal [Color figure can be viewed at wileyonlinelibrary.com]…”

A 0.8 mm2 Sub-GHz GaAs HBT Power Amplifier for 5G Application Achieving 57.5% PAE and 28.5 dBm Maximum Linear Output Power

“…In both methods, a signal pre-distorter is placed at the input of the PA to alter the non-linear transfer function's magnitude and phase of the PA. The difference between APD and DPD lies in the design of the pre-distorter where the latter generates the aforementioned non-linear responses with the aid of a DSP processor [35]- [38]. However, the complexity in integration, resulting in larger chip area and dual fabrication process serves to be the prime disadvantages.…”