An Order-Reduced Memory Polynomial Behavioral Model for RF Power Amplifier

This paper presents a behavioral model of order-reduced memory polynomial (ORMP) for digital predistortion (DPD) in RF power amplifiers. After iterative training with Greedy-Search method and Model-Growth technique, the proposed ORMP model with additional crossing terms achieves better predistortion performance and lower complexity compared with other DPD techniques. The effectiveness of the proposed model is verified by the measurement results of a switched capacitor power amplifier (SCPA). For LTE 10MHz64 QAM PAPR S.4dB signal, the EVM is improved by 6. 144dB and the ACPR is improved by 6. 189dB. For LTE 20MHz64 QAM PAPR 8. 4dB signal, the EVM and ACPR are improved by 4. 723dB and 2. 542dB respectively.