The Threshold Optimization of the Canonical Piecewise Linear Function-Based Model for RF PA Linearization

Various canonical piecewise linear (CPWL) function-based models have been widely used for modeling radio frequency (RF) power amplifiers (PAs). In most literatures, the thresholds of CPWL are often selected according to the uniform distribution, and how to select the optimal thresholds is rarely discussed. This letter presents a new iterative method, which updates the thresholds after calculating the model coefficients, and finally searches the optimal thresholds. The experimental results confirmed that the method can keep good linearization performance and significantly reduce the number of thresholds and the number of model coefficients.