功率放大器及预失真器模型的设计

传统多项式常用在功率放大器模型和预失真器模型的设计里。然而，随着多项式阶数逐次增加时，传统多项式模型表现出数据的不稳定性。详细介绍了一种用于功率放大器以及预失真器模型的正交多项式。从理论上讲，传统多项式与正交多项式模型是等效的，但实践表明，在有限精度范围内，这两种方法对数据进行了不同的处理。MATLAB仿真结果表明，正交多项式可以减轻数值不稳定的问题，建立高精度的功率放大器和预失真器模型，进而改善系统的线性化指标。

Designs of Power Amplifier and Predistorter Models

Traditional polynomial model is commonly used in the design of power amplifier and predistorter models. However, with the gradual increase of orders, the conventional polynomial model exhibits certain numerical instability. This paper gives a novel set of orthogonal polynomials for power amplifier as well as predistorter modeling. In theory, the conventional and orthogonal polynomial models are equivalent and, thus, should behave similarly, but the experiment indicates that these two approaches process data in quite different ways within the finite precision range. The simulation with MATLAB shows that the orthogonal polynomial can reduce the numerical instability, establish high-precision power-amplifier and predistorter models and further improve linearization indexes.