Spectral weighted SAMP method for enhancing out-of-band DPD linearization

Abstract

The invention discloses a spectrum weighting SAMP method for enhancing out-of-band DPD linearization, and the method comprises the steps: constructing an offline operation module, driving a power amplifier (PA) through a test signal to obtain the input and output data of the PA, carrying out the in-band notch processing of the output signal of the PA in a frequency domain, improving a sparsity adaptive matching pursuit (SAMP) algorithm, employing an improved algorithm, namely a sparseness adaptation matching pursuit (SW-SAMP) algorithm of a spectrum weighting strategy, for selecting a group of most relevant basis functions, performing the process only once, inputting an obtained optimal basis set into a pre-distortion module of a power amplifier inverse model, estimating a DPD model coefficient by utilizing least square iteration, solving the inverse of a behavior model, and then assigning the final result to the predistorter. According to the method provided by the invention, the calculation complexity of the PA behavior model in the forward path can be reduced, and the stability of the whole pre-distortion system is improved.

Description

technical field [0001] The invention belongs to the field of digital signal processing, in particular to a spectrum weighted SAMP method for enhancing out-of-band DPD linearization. Background technique [0002] Today, the most commonly used method to solve the inherent linearity and low efficiency of the power amplifier (PA) is digital predistortion technology (Digital PreDitortion, DPD). Some technologies that emerged with 4G communication, such as carrier aggregation or MIMO, and then in 5G, the ever-increasing signal bandwidth and peak-to-average power ratio make it difficult for the predistorter to deal with the PA. However, the number of parameters required to compensate the static nonlinearity and dynamic memory effect in the DPD model is very large, which increases the computational complexity, which makes the DPD model extraction and adaptive process inefficient, resulting in overfitting when DPD estimates parameters . However, with the development and rise of Com...

AI Technical Summary

Problems solved by technology

With the rise of some technologies in 4G communication, such as carrier aggregation or MIMO, and then in 5G, the increasing signal bandwidth and peak-to-average power ratio make the predistorter encounter difficulties when dealing with PA

However, the number of parameters required to compensate the static nonlinearity and dynamic memory effect in the DPD model is very large, which increases the computational complexity, which makes the DPD model extraction and adaptive process inefficient, resulting in overfitting when DPD estimates parameters

Images