基于QR-RLS脉动阵列的非线性功放建模

Nonlinear Power Amplifier Modeling Based on QR-RLS Systolic Array

在宽带通信系统中,功率放大器表现出明显的记忆效应,无记忆模型无法准确描述其输入输出特性。采用Hamm erste in模型预测实际宽带发射机的记忆效应,并用脉动阵列的QR-RLS算法实现了快速高效的参数提取,不仅具有良好数值稳定性、减小了运算量和数据存储空间,而且易于在FPGA中实现。

For wideband communication systems, power amplifier exhibits strong memory effects, while memory-less models can not behavior their input and output characteristics correctly. This paper uses Hammerstein model to forecast memory effect of wideband RF transmitter. The model＇s parameter-identification uses QR-RLS systolic array implementation, which can improve numerical stability, reduce complexity and save memory space, and the new implementation can be easily implemented by FPGA.