Extended Hammerstein Behavioral Model Using Artificial Neural Networks

In this paper, a novel extended Hammerstein model is presented to accurately mimic the dynamic nonlinearity of wide-band RF power amplifiers (RFPAs). Starting with a conventional Hammerstein model scheme, which fails to predict the behavior of the RFPA with short-term memory effects, two areas of improvements were sought and found to allow for substantial improvement. First, a polar feed-forward neural network (FFNN) was carefully chosen to construct the memoryless part of the model. The error signal between the output and the input signal of the memoryless sub-model was then filtered and then post-injected at the model output. This extra branch, when compared to the conventional Hammerstein scheme, allowed for an extra mechanism to account for the memory effects due to dispersive biasing network that was present otherwise. The excellent estimation capability of the polar FFNN together with the additional filtered error signa post-injection led to remarkable accuracy when modeling two different RFPAs both driven with four-carrier wideband code division multiple access signals. Despite its simple topology and identification procedure, the extended Hammerstein model demonstrated is capable in accurately predicting the dynamic AM/AM and AM/PM characteristics and the output signal spectrum of the RFPA under test.

本文提出了一种新颖的扩展哈默斯坦模型，用于精确模拟宽带射频功率放大器（RFPA）的动态非线性。从传统的哈默斯坦模型方案出发，该方案无法预测具有短期记忆效应的RFPA的行为，我们寻求并找到了两个改进方向，从而实现了显著的改进。首先，精心选择了一个极坐标前馈神经网络（FFNN）来构建模型的无记忆部分。然后，对无记忆子模型的输出和输入信号之间的误差信号进行滤波，并在模型输出端后注入。与传统的哈默斯坦方案相比，这个额外的分支提供了一种额外的机制，用于解释由于存在的色散偏置网络而产生的记忆效应。极坐标FFNN出色的估计能力以及额外的滤波误差信号后注入，在对两个由四载波宽带码分多址信号驱动的不同RFPA进行建模时，实现了显著的精度。尽管其拓扑结构和识别过程简单，但所展示的扩展哈默斯坦模型能够准确预测被测RFPA的动态幅度/幅度（AM/AM）和幅度/相位（AM/PM）特性以及输出信号频谱。