How Long Can I Transmit? A Mobility Aware mmWave-based UAV Communication Framework

TL;DR

This paper develops an analytical framework to estimate the expected line-of-sight duration between UAVs and mobile ground users in urban environments, optimizing UAV-user association based on mobility-aware LoS predictions.

Contribution

It introduces a novel mobility-aware LoS duration model for UAV communications in urban settings using MPLP, and proposes an improved user association algorithm based on this model.

Findings

The derived LoS duration expression matches simulation results.

The mobility-aware association algorithm outperforms nearest-UAV schemes.

The framework effectively predicts LoS durations for mobile users in urban environments.

Abstract

One primary focus of next generation wireless communication networks is the millimeterwave (mmWave) spectrum, typically considered in the 30 GHz to 300 GHz frequency range. Despite their promise of high data rates, mmWaves suffer from severe attenuation while passing through obstacles. Unmanned aerial vehicles (UAVs) have been proposed to offset this limitation on account of their additional degrees of freedom, which can be leveraged to provide line of sight (LoS) transmission paths. While some prior works have proposed analytical frameworks to compute the LoS probability for static ground users and a UAV, the same is lacking for mobile users on the ground. In this paper, we consider the popular Manhattan point line process (MPLP) to model an urban environment, within which a ground user moves with a known velocity for a small time interval along the roads. We derive an expression for the expected duration of LoS between a static UAV in the air and a mobile ground user, and validate the same through simulations. To demonstrate the efficacy of the proposed analysis, we propose a simple user association algorithm that greedily assigns the UAVs to users with the highest expected LoS time, and show that it outperforms the existing benchmark schemes that assign the users to the nearest UAVs with LoS without considering the user mobility.