PROJECT TITLE :

Complexity-aware-normalised mean squared error 'CAN’ metric for dimension estimation of memory polynomial-based power amplifiers behavioural models

ABSTRACT:

The memory polynomial model is widely used for the behavioural modelling of radio-frequency non-linear power amplifiers having memory effects. One difficult task related to the current model is the choice of its dimension which is outlined by the non-linearity order and the memory depth. This study presents an approach appropriate for the choice of the model dimension in memory polynomial-based power amplifiers’ behavioural models. The proposed approach uses a hybrid criterion that takes into account the model accuracy and its complexity. The proposed technique is tested on 2 memory polynomial-based mostly behavioural models. Experimental validation allotted using experimental knowledge of two Doherty power amplifiers, engineered using different transistor technologies and tested with two completely different signals, illustrates consistent blessings of the proposed technique as it considerably reduces the model dimension by more than sixty% without compromising its accuracy.