Enhanced memory polynomial with reduced complexity in digital pre-distortion for wireless power amplifier

Power Amplifier (PA) is one of the prominent devices in a communications system. Ideally, the PA linearly amplifies signals, but exhibits non-linearity when operates in the actual world, where PA output power deviates away from the ideal linear region. The non-linearity of the PA has result in va...

Full description

Saved in:
Bibliographic Details
Main Author: Choo, Hong Ning
Format: Thesis
Language:English
Published: 2017
Subjects:
Online Access:http://psasir.upm.edu.my/id/eprint/71221/1/FK%202017%2079%20-%20IR.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Power Amplifier (PA) is one of the prominent devices in a communications system. Ideally, the PA linearly amplifies signals, but exhibits non-linearity when operates in the actual world, where PA output power deviates away from the ideal linear region. The non-linearity of the PA has result in various undesired effects include amplitude and phase distortion which contributes to Adjacent Channel Interference (ACI) that degrades the signal quality at the receiver side. Inevitable increasing bandwidth and transmission speed causes memory effects in the PA. Memory effects causes scattering of the PA output signal and increases overhead processing requirements at the receiver side to decode/rectify deteriorated signal quality. PA linearization is therefore required to neutralize the non-linearity effects on the system. Among various linearization methods, Digital Pre-distortion (DPD) stands out due to its balanced advantages and trade-offs in terms of implementation simplicity, supported bandwidth, efficiency, flexibility and cost. DPD models the PA, pre-distorts the input signal with an inversed function of the PA, and further feeds the pre-distorted input signal into the PA. The Memory Polynomial method (MP) by (Ding, 2004), a simplified derivative of the Volterra Series is capable of modeling the PA with Memory Effects with reduced complexity. This project presents the MP with Binomial Reduction method (MPB) which is an optimized MP with reduced addition and multiplication operations. Referring to Computational Complexity Reduction Ratio (CCRR) by (Hou, 2011), Multiplication Operations Reduction Ratio (MORR) and Addition Operations Reduction Ratio (AORR) are derived to showcase the reduction percentage of addition/multiplication operations in MPB against the method to be compared. Comparing to MP, MPB is capable of achieving similar Adjacent Channel Power Reduction (ACPR) Ratio performance, amplitude and phase distortion reduction, memory effects elimination, improvements in Normalized Mean Square Error (NMSE) of 36.5dB, 86.43% AORR and 50% MORR. MPB is also compared with one of the recent derivatives of MP, the Augmented Complexity Reduced General MP (ACRGMP) by (Liu, 2014) with 56.76% MORR, 84.38% AORR and 92.36dB of NMSE improvement. The method is simulated in MATLAB by Mathworks using a modeled ZVE-8G PA and fed with sampled 4G (LTE) signals.