Max-min Rate Deployment Optimization for Backhaul-limited Robotic Aerial 6G Small Cells

TL;DR

This paper proposes an optimization framework for deploying robotic aerial base stations in 6G small cell networks, enhancing user data rates through joint deployment, user association, and resource allocation, using semidefinite relaxation techniques.

Contribution

It introduces a novel joint optimization approach for RABS deployment and resource management, formulated as a nonconvex QCQP and solved with a semidefinite relaxation heuristic.

Findings

Minimum user data rate improved by up to 95.43%.

SDR heuristic outperforms linear relaxation baseline.

Deployment of RABS significantly enhances network performance.

Abstract

To overcome the limited on-board battery issue of nominal airborne base stations (ABSs), we are exploring the use of robotic airborne base station (RABS) with energy neutral grasping end-effectors that are able to autonomously perch at tall urban landforms. Specifically, this paper studies a heterogeneous network (HetNet) assisted by a movable RABS as a small cell which connects to a macro base station (MBS) through a limited-capacity wireless backhaul link, which can be deemed as another major challenge. To exploit the potential gains that the mobility of RABS can bring in the system, the minimum rate among all users is maximized by jointly optimizing the RABS deployment, user association and subcarrier allocation. This problem is initially formulated as a binary polynomial optimization (BPO) problem. After reformulating it as a nonconvex quadratically constrained quadratic programming (QCQP), we propose a semidefinite relaxation (SDR) based heuristic method to capture a high-quality solution in polynomial time. Numerical results reveal that deploying a RABS as the small cell can improve the minimum data rate by 95.43% at most and 33.97% on average, and the developed SDR heuristic algorithm significantly outperforms the linear relaxation (LR) baseline method.