Support vector regression-based dynamic behavioral modeling for RF power amplifiers

Cai, Jialin; Yu, Chao; Sun, Lingling; Li, Jun; Su, Jiangtao; Wang, Xiang; Huang, Xiwei; Jie, Liang

doi:10.1109/ieee-iws.2018.8400885

Cited by 8 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then the loss can be calculated when the absolute value of the difference between f (x) i and y i is greater than ε. In the SVR model, the function f (x) is defined as the center, and there are ε isolation bands on the left and right [20]. Once the loss is minimized, the obtained function f (x) i can be close to y i .…”

Section: Theoretical Derivation Of the Tsvr Model A Basics Of Sumentioning

confidence: 99%

“…Unlike other SVR-based method, the TSVR model finds a pair of non-parallel planes by solving two related SVM-type problems, namely the ε -insensitive up-and down-bound functions. Meanwhile, in order to accelerate the training process, the TSVR model solve two smaller size quadratic programming problems (QPP) with only one constraint similar to TSVM [19], [20]. The advantage of this strategy is that two smaller size QPPs are solved rather than one larger QPP, which can reduce the computational complexity of the traditional SVR approach.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

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

Jiang

et al. 2019

IEEE Access

View full text Add to dashboard Cite

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.

show abstract

Section: Theoretical Derivation Of the Tsvr Model A Basics Of Sumentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

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

Jiang

et al. 2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…However, ANN based modeling technique is faced with the issue of overfitting which, in some cases, might attributes to unsatisfactory performance 32,33 . Thus, some other new modeling methods have been tried for RF and microwave device modelling, such as support vector regression (SVR) method 31,32,34–36 . In Reference 35, a new dynamic behavioral model for RF PAs based on SVR technique is provided.…”

Section: Introductionmentioning

confidence: 99%

“…32,33 Thus, some other new modeling methods have been tried for RF and microwave device modelling, such as support vector regression (SVR) method. 31,32,[34][35][36] In Reference 35, a new dynamic behavioral model for RF PAs based on SVR technique is provided. Later in Reference 36, the SVR is employed to build a behavioral model for GaN HEMT.…”

mentioning

confidence: 99%

Analysis and modeling of the kink effect in S22 based on support vector machine for GaN HEMTs

Zhu

Geng

Cai

et al. 2022

Int J Numerical Modelling

Self Cite

View full text Add to dashboard Cite

In this work, the kink effect (KE), typically visible in S22, is analyzed and modeled. Two different modeling techniques: equivalent circuit modeling (ECM) method and machine learning method which based on support vector regression (SVR) technique are presented and compared, when applied to the S22 behavior of a Gallium Nitride (GaN) high electron mobility transistor (HEMT). The device under test (DUT) has a width of 8 Â 125 μm, with a gate feature size of 0.25 μm. The proposed method identifies the effect that the bias voltage and extrinsic elements have on the S22 kink shape. Additionally, compared to ECM, the SVR model attains a superior fitting accuracy across the complete frequency band.

show abstract

“…Constructing an accurate behavioral model of the PA is one of the most effective ways to address its nonlinearity and memory effects when embedded within a system level design, and it is also one of the most important step of PA's digital pre-distortion (DPD). [11][12] Many behavioral models and modeling methodologies have been proposed that consider the PA as a black-box, [13][14][15][16][17] using only the input and output data to determine the model parameters, 18 that is, details surrounding the internal physical operation is not required.…”

Section: Introductionmentioning

confidence: 99%

Memory polynomial based support vector regression dynamic behavioral model for Doherty power amplifiers

Wang

King

Cai

2022

Int J Numerical Modelling

Self Cite

View full text Add to dashboard Cite

In this article, a novel behavior modeling methodology for radio frequency (RF) power amplifiers (PAs) is introduced. The model is targeted towards strongly nonlinear Doherty power amplifiers (DPAs), combining the memory polynomial (MP) topology with the existing support vector regression (SVR) algorithm. The resulting novel model, is termed the memory polynomial support vector regression (MP-SVR) model. Experimental validation proceeds by applying the proposed modeling method to both two standard gallium nitride (GaN) DPA with different nonlinearity, as well as a multi-transistor GaN DPA. Compared with traditional Volterra based models, the standard SVR model, the augmented SVR (ASVR) model and the new augmented SVR (NASVR) model, the proposed MP-SVR model gives superior prediction accuracy in both cases. This shows the efficacy of the proposed modeling method.

show abstract

Support vector regression-based dynamic behavioral modeling for RF power amplifiers

Cited by 8 publications

References 9 publications

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

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

Analysis and modeling of the kink effect in S22 based on support vector machine for GaN HEMTs

Memory polynomial based support vector regression dynamic behavioral model for Doherty power amplifiers

Contact Info

Product

Resources

About