A Low Complexity LUT-Based Digital Predistortion Block With New Pruning Method

Abstract

Generalized memory polynomial (GMP) model is widely used to build and compensate for the nonlinearity of power amplifier (PA). To reduce the complexity of the GMP model, a pruning method represented by the doubly orthogonal matching pursuit (OMP) algorithm was proposed. Lookup table (LUT) based predistortion blocks for compensating PAs are widely used. The minimum complexity of the GMP model does not represent the minimum complexity of LUT. To address this issue, this letter proposes one new pruning method to reduce the complexity of digital predistortion (DPD) block based on LUT. This method first classifies and selects the basis functions according to the memory effect and then optimizes the structure of LUT. Experimental results show that the proposed method can obtain the low complexity of DPD block based on LUT and ensure high precision compensation for the nonlinearity of PA.