Digital predistortion challenges in the context of software defined transmitters

This paper discusses the challenges associated with the application of the digital predistortion (DPD) technique in the context of software defined transmitters. Given the importance of the DPD technique for compensating for the sources of distortions, and consequently improving the tradeoff between the linearity and the power efficiency, this paper examines the sensitivity of two DPD schemes, namely Memory Polynomial and the Two Hidden Layer Artificial Neural Networks, to the frequent changes in the characteristics (such as the modulation bandwidth, probability density function, peak-to-average power ratio) of the software-defined transmitters´ input signals.