A Memoryless Semi-Physical Power Amplifier Behavioral Model Based on the Correlation Between AM–AM and AM–PM Distortions

A new semi-physical memoryless computationally effective behavioral model (BM) capable of predicting amplitude-modulation-to-amplitude-modulation (AM-AM) and amplitude-modulation-to-phase-modulation (AM-PM) distortions is proposed. In recent years, AM-AM and AM-PM distortions have been separately studied in literature. Here, we investigate the correlation between AM-AM and AM-PM distortions first. Based on the observed correlation, a novel AM-PM model is derived from the well-known Rapp AM-AM model. On the basis of the close relationship between the AM-AM and AM-PM model, a highly accurate and computationally effective large-signal BM is obtained. The newly developed model addresses the current needs of the mobile industry that requires BMs of the power amplifier (PA), which can be interpreted by the designers. In addition, the models must be computationally effective due to the limited computation power in mobile handset. Therefore, only memoryless BMs can be taken into account and larger errors are accepted for the sake of computational effectiveness. In this paper, it is shown that the newly introduced model achieves excellent results, when it is applied to actual industrial applications of two GaAs-based class-AB PAs in 65-nm technology and one CMOS-based class AB PA in 28-nm technology, which are designed for mobile communications.