The article addresses the problem of jointly optimizing the trajectory planning (TP), user association (UA), resource allocation (RA), and power control (PC) to maximize the proportional fairness (PF) in an aerial IoT network. The key highlights are:
The authors point out a prevalent mistake in coordinate optimization approaches adopted by the majority of the literature, where the methods often stagnate at a non-stationary point, significantly degrading the network utility in mixed-integer problems such as joint TP and radio resource optimization.
To address this issue, the authors convert the formulated problem into a Markov decision process (MDP) and design a non-iterative framework that cooperatively optimizes the trajectory and radio resources without an initial trajectory choice.
The authors propose a novel temporal decoupling method to separate the PF problem into unit-time sub-problems, allowing the application of well-known RRM techniques specialized for optimizing unit-time network snapshots.
The authors introduce a generalized water-filling algorithm that can find the optimal UA and RA combination when both the minimum resource requirements and resource budget co-exist.
The authors evaluate the proposed framework using various decision-making schemes, including genetic algorithm, depth-first search, and deep Q-learning, and show that it significantly outperforms the state-of-the-art methods, nearly achieving the global optimum.
إلى لغة أخرى
من محتوى المصدر
arxiv.org
الرؤى الأساسية المستخلصة من
by Hyeonsu Lyu,... في arxiv.org 05-03-2024
https://arxiv.org/pdf/2405.01314.pdfاستفسارات أعمق