Dataset Generation for Drone Optimal Placement Using Machine Learning

TL;DR

This paper details the creation of a dataset for optimizing drone placement in vehicular networks using machine learning, aiming to enhance communication and energy efficiency.

Contribution

It introduces a dataset specifically designed for training algorithms to optimize UAV placement in drone-assisted vehicular networks.

Findings

Delay analysis of vehicle requests using queuing theory

Modeling of drone energy consumption

Simulation scenario and dataset features described

Abstract

Unmanned aerial vehicle (UAV), or drone is increasingly becoming a promising tool in communication system. This report explains the generation details of a dataset which will be used to designing an algorithm for the optimal placement of UAVs in the drone-assisted vehicular network (DAVN). The goal is to improve the drones' communication and energy efficiency after our previous work. The report is organized as followed: the first section is devoted to the delay analysis of the vehicle requests in the DAVN using queuing theory; the second part of the report models the energy consumption of the drones while the third section explains the simulation scenario and dataset features. The notations and terminologies used in this report are summarized in the last section.