Saved in:
Bibliographic Details
Main Authors: Di, Haoyang, Zhu, Xiaodong, Shao, Yulin
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2411.08680
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915281450827776
author Di, Haoyang
Zhu, Xiaodong
Shao, Yulin
author_facet Di, Haoyang
Zhu, Xiaodong
Shao, Yulin
contents Unmanned aerial vehicles (UAVs) have become key enablers in relay-assisted wireless communications thanks to their flexibility and line-of-sight channel advantage. However, most existing trajectory optimization frameworks assume ideal Gaussian inputs, overlooking the fact that practical wireless systems rely on structured, finite-alphabet constellations. This mismatch can lead to suboptimal, and sometimes misleading, design choices. In this paper, we challenge that convention by introducing a finite-alphabet-aware framework for joint trajectory and precoder optimization in UAV-assisted relay systems. We formulate a non-convex design problem that directly accounts for discrete signal structures and propose an efficient solution based on alternating optimization and successive convex approximation. Simulation results reveal that strategies optimized under Gaussian assumptions can waste energy and degrade throughput in real deployments. In contrast, our approach adapts both the UAV's trajectory and transmission strategy to the underlying modulation format, delivering consistent performance gains under practical system constraints. This work takes a key step toward aligning UAV communication design with the realities of modern wireless systems: discrete signals, power limits, and intelligent mobility.
format Preprint
id arxiv_https___arxiv_org_abs_2411_08680
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Finite-Alphabet-Aware Trajectory and Precoder Optimization for UAV Relaying
Di, Haoyang
Zhu, Xiaodong
Shao, Yulin
Information Theory
Signal Processing
Unmanned aerial vehicles (UAVs) have become key enablers in relay-assisted wireless communications thanks to their flexibility and line-of-sight channel advantage. However, most existing trajectory optimization frameworks assume ideal Gaussian inputs, overlooking the fact that practical wireless systems rely on structured, finite-alphabet constellations. This mismatch can lead to suboptimal, and sometimes misleading, design choices. In this paper, we challenge that convention by introducing a finite-alphabet-aware framework for joint trajectory and precoder optimization in UAV-assisted relay systems. We formulate a non-convex design problem that directly accounts for discrete signal structures and propose an efficient solution based on alternating optimization and successive convex approximation. Simulation results reveal that strategies optimized under Gaussian assumptions can waste energy and degrade throughput in real deployments. In contrast, our approach adapts both the UAV's trajectory and transmission strategy to the underlying modulation format, delivering consistent performance gains under practical system constraints. This work takes a key step toward aligning UAV communication design with the realities of modern wireless systems: discrete signals, power limits, and intelligent mobility.
title Finite-Alphabet-Aware Trajectory and Precoder Optimization for UAV Relaying
topic Information Theory
Signal Processing
url https://arxiv.org/abs/2411.08680