Behavioural modelling of RF power amplifiers using modified Volterra series in the time domain

This paper presents a new method for the extraction of a behavioural model of an RF power amplifier using the complex envelope representations of the input and output signals in the time domain. A quasi-memory less model is extracted, and used to represent the nonlinearities of the device. The coefficient values are estimated for Volterra filters which are added in parallel to characterise the memory effects of the device. Compared with classic memoryless models, this model provides a more accurate prediction of the actual power amplifier output signal in both the time and frequency domains.