Iterative pre-distortion of the non-linear satellite channel

TL;DR

This paper introduces an iterative pre-distortion method for non-linear satellite channels, improving signal quality for high-order modulations like DVB-S2, with reduced complexity and enhanced performance over existing techniques.

Contribution

It proposes a novel iterative pre-distortion algorithm for long-memory non-linear satellite channels, with approximations to reduce complexity while maintaining effectiveness.

Findings

Improved pre-distortion performance for high-order modulations

Reduced computational complexity through approximations

Enhanced robustness against non-linear satellite channel impairments

Abstract

Digital Video Broadcasting - Satellite - Second Generation (DVB-S2) is the current European standard for satellite broadcast and broadband communications. It relies on high order modulations up to 32-amplitude/phase-shift-keying (APSK) in order to increase the system spectral efficiency. Unfortunately, as the modulation order increases, the receiver becomes more sensitive to physical layer impairments, and notably to the distortions induced by the power amplifier and the channelizing filters aboard the satellite. Pre-distortion of the non-linear satellite channel has been studied for many years. However, the performance of existing pre-distortion algorithms generally becomes poor when high-order modulations are used on a non-linear channel with a long memory. In this paper, we investigate a new iterative method that pre-distorts blocks of transmitted symbols so as to minimize the Euclidian distance between the transmitted and received symbols. We also propose approximations to relax the pre-distorter complexity while keeping its performance acceptable.