Search

In this letter, a novel digital predistortion (DPD) model for broadband power amplifier (PA) linearization is proposed, namely Attention Mechanism based Bidirectional Long Short‐term Memory Network (AM‐BiLSTM) model. In order to verify the linearization performance of the AM‐BiLSTM model in digital predistortion process, a 100 MHz bandwidth 5G new radio (5G NR) signal is employed to test a sub‐6G PA operating at 2.6‐GHz. The experimental results show that the adjacent channel power ratio (ACPR) of the PA with AM‐BiLSTM model can be improved by 24 dB which is 6 dB and 3 dB better than the generalized memory polynomial (GMP) model and the Chebyshev polynomials LSTM (CP‐LSTM) model in ref [1], repspectively. Therefore, the proposed AM‐BiLSTM model is very effective for the DPD linearization of broadband PAs. [ABSTRACT FROM AUTHOR]

This letter presents a non‐linear dynamic behaviour model for characterising the broadband power amplifiers (PAs) by using deep convolution generative adversarial networks (DCGAN). The DCGAN structure is based on the convolution neural network model and combines the generative adversarial networks to improve its linearisation ability of the digital predistortion. The DCGAN contains a generation model and a discriminant model. It imports convolution with steps and deconvolution into the structure, respectively, which makes the accuracy of the power amplifier non‐linear model to improve further. For verification, a 5G NR test signal with 100 MHz bandwidth is employed for testing a Doherty RF‐PA that operates at 1800 MHz. The experimental results illustrate that the normalised mean square error value is at least 12 dB higher than the traditional models, and the out‐of‐band suppression of the DCGAN predistorter can be up to 15 dB better than other models. [ABSTRACT FROM AUTHOR]