A robust and broadband digital predistortion utilizing negative feedback iteration

In this paper, a robust and broadband digital predistortion (DPD) based on negative feedback iteration technique is proposed for linearization of power amplifier with strong memory effects. Instead of estimating the DPD parameters directly in conventional direct learning architecture, the proposed DPD identifies the optimal DPD signal which eliminates the modeling error at first and then calculate the corresponding DPD parameters. The proposed DPD can realize superior linear performance for broadband signal than conventional direct learning DPD while having a relative low computational complexity. A 2-carrier 200MHz bandwidth LTE signal at 2.1GHz DPD test is carried out for validation, which achieves better than 11.8dB ACPR correction and outperforms conventional direct leaning method.