【作者】 刘一； 鲍景富； 李明玉；

【Author】 Liu Yi~1,Bao Jingfu~2,Li Mingyu~3 (School of electronic engineer,University of Electronic Science and Technology of China,Chengdu 610054) 123

【机构】 电子科技大学电子工程学院；

【摘要】 Volterra级数模型是射频功率放大器建模的常用方法,但是随着系统记忆长度和阶次的增加,其中参数数量会迅速增长,限制了该模型在高阶系统建模中的应用。本文通过引入阀值,提出一种新的简化的Volterra级数模型。同时通过使用以复值Givens旋转为基础的QR分解递归最小二乘算法(QRD-RLS)代替传统最小二乘法更新模型参数,在降低了运算复杂度的同时提高了模型精确性。更多还原

【Abstract】 General Volterra series based models have been successfully applied for radio frequency(RF) power amplifiers(PA) behavioral modeling,but the requirement for an estimation of a large number of parameters is a major limitation in using the Volterra series to model nonlinear RF power amplifiers.In this letter,we propose a new simplified Volterra series based model for RF power amplifiers by employing a"valve-value"pruning algorithm to remove the coefficients which are very small,or else not sensitive to the output error.At the same time,we propose utilizing the complex QR-decomposition based recursive least squares(QRD-RLS) algorithm,which is implemented using the complex Givens rotations,to instead of the standard least squares algorithms to updating the coefficient vector of the memory polynomial modeling.Therefore dramatically reducing the complexity and enhancing the fidelity of the behavioral model.更多还原