Joint Channel Bandwidth Assignment and Relay Positioning for Predictive Flying Networks

Ruben Queiros 1 , Megumi Kaneko 2 , Helder Fontes 1 , Rui Campos 1 · 2025

Read Joint Channel Bandwidth Assignment and Relay Positioning for Predictive Flying Networks by Ruben Queiros 1 , Megumi Kaneko 2 , Helder Fontes 1 , Rui Campos 1 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-14248.txt.

Introduction

Ruben Queiros, Megumi Kaneko, Helder Fontes, and Rui Campos present a method for flying networks that relay data by drone, in a research paper from 2025. Flying networks can bring wireless links where coverage is weak or gear is damaged, as in disaster zones. Keeping relay links strong while drones move is still a hard problem.

Their method, SAFnet, assigns channel width and moves the relay drone as needs shift. It uses known paths of the first drone tier to plan ahead. A search method with penalty rules guides the choices. Results come close to full search at far lower cost. The work suits planned, moving drone fleets.

Catalog record

CREATORRuben Queiros 1 , Megumi Kaneko 2 , Helder Fontes 1 , Rui Campos 1
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE31 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-14248

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Joint Channel Bandwidth Assignment and Relay Positioning for Predictive Flying Networks Thanks: This work is co-financed by Component 5 - Capitalization and Business Innovation of core funding for Technology and Innovation Centres (CTI), integrated in the Resilience Dimension of the Recovery and Resilience Plan within the scope of the Recovery and Resilience Mechanism (MRR) of the European Union (EU), framed in the Next Generation EU, for the period 2021 - 2026, by National Funds through the Portuguese funding agency, FCT - Fundação para a Ciência e Tecnologia, under the PhD grant 2022.10093.BD, by the NII MoU grant between NII and INESC TEC, and by the Grants-in-Aid for Scientific Research no. 20H00592 and 22KK0156 from the Ministry of Education, Science, Sports, and Culture of Japan. Ruben Queiros 1 , Megumi Kaneko 2 , Helder Fontes 1 , Rui Campos 1 Abstract Flying Networks (FNs) have emerged as a promising solution to provide on-demand wireless connectivity when network coverage is insufficient or the communications infrastructure is compromised, such as in disaster management scenarios. Despite extensive research on Unmanned Aerial Vehicle (UAV) positioning and radio resource allocation, the challenge of ensuring reliable traffic relay through backhaul links in predictive FNs remains unexplored. This work proposes Simulated Annealing for predictive FNs (SAFnet), an innovative algorithm that optimizes network performance under positioning constraints, limited bandwidth and minimum rate requirements. Our algorithm uniquely leverages prior knowledge of the first-tier node trajectories to assign bandwidth and dynamically adjust the position of the second-tier flying relay. Building upon Simulated Annealing, our approach enhances this well-known AI algorithm with penalty functions, achieving performance levels comparable to exhaustive search while significantly reducing computational complexity. I Introduction Natural and man-made disasters can rapidly bring down existing communications infrastructure, creating urgent demand for adaptable, resilient connectivity solutions. Flying networks (FNs), composed of Unmanned Aerial Vehicles equipped with communications hardware payloads, offer a scalable, cost-effective solution to extend and reinforce network coverage and capacity in such critical scenarios, enabling the deployment of intelligent emergency communications. Unlike fixed base stations, UAVs can be redeployed multiple times, and their inherent mobility enables rapid, on-the-fly adjustments to the network topology and positioning – capabilities far beyond those of conventional Cell on Wheels solutions [ 1 ] . Despite the potential of FNs, key challenges persist, particularly in the optimal positioning and real-time allocation of radio resources. Existing research has extensively explored these aspects, yet it often overlooks the dynamics of predictive FNs, where UAVs follow predefined trajectories with specific Quality of Service (QoS) requirements [ 2 ] . We address this scenario, where the Flying Edge Nodes (FEN) have previously defined missions [ 3 ] with known QoS requirements and predefined trajectories assigned by the mission commander. Figure 1 illustrates a predictive FN, where the FENs have their mission assigned, and the High Altitude Platform (HAP) has the role of ensuring the relay of communications to a backhaul network. This network is not directly communicating with FENs due to differences in wireless technology or communications range. In addition to HAP positioning, the management of limited wireless channels is crucial to ensure resource allocation that maximizes network performance while minimizing radio interference. Furthermore, low-complexity algorithms are needed to meet decision time requirements and reduce energy consumption, which are key aspects in FNs. Fig. 1: Two-tier predictive Flying Network. In recent years, several solutions have been proposed to address the UAV optimal positioning. In [ 4 , 5 ] the authors address the joint optimization of the UAV placement and routing. However, they consider static default values for the channel bandwidth, and overlook the link to the backhaul network. In [ 6 ] , the authors maximize the number of connections in an FN that satisfy minimum QoS requirements, by optimizing radio resources, node trajectory and computing resources but the computational complexity of the proposed method does not scale with the number of nodes. Reference [ 7 ] proposes a low-complexity algorithm that maximizes the minimum throughput for the ground users served by the FN, considering the optimization of the UAV trajectories and transmission power. However, similarly to [ 6 ] , the authors do not consider the link to the backhaul network nor predefined FEN trajectories. A backhaul-aware solution has been proposed in [ 8 ] , but using branch-and-bound and discrete grid search to determine the optimal UAV position is computationally expensive, potentially preventing a timely computation of the solution. In [ 9 ] , the authors optimize the number and position of UAVs in an FN considering the backhaul link and using orthogonal frequency channels. Yet, they neglect the limited backhaul link capacity and channel bandwidth. In [ 10 ] , the authors optimize the UAV positioning and bandwidth allocation in single-tier FNs focusing on the ground users’ links. They consider the backhaul link but assume the bandwidth can be arbitrarily fractioned, which is not valid for real communication systems. In this paper, we propose an innovative algorithm designed to optimize network performance under predefined FEN trajectories, limited bandwidth of the wireless links, minimum rate requirements for the FENs, and limited backhaul link capacity. Our algorithm uniquely leverages prior knowledge of the FENs’ trajectories to dynamically adjust the position of the HAP and assign bandwidth to the relay and backhaul links (cf. Figure 1 ). By utilizing Simulated Annealing (SA), our approach incorporates penalty functions to adapt this well-known AI algorithm to the problem, achieving performance levels comparable to exhaustive search while significantly reducing complexity. The contributions of this paper are three-fold: 1. We mathematically formulate the joint problem of wireless channel assignment and HAP positioning in predictive FNs, considering constraints such as channel bandwidth, backhaul capacity and minimum rate requirements of the FENs; 2. We design an approach based on SA to solve this intricate problem, by incorporating constraint-specific penalty functions to effectively manage these constraints while maintaining low computational complexity; 3. We provide numerical results that demonstrate the benefits of the proposed approach in terms of weighted sum-rate and outage probabilities, outperforming state-of-the-art benchmarks. II System Model The two-tier FN considered in this work consists of a HAP h assigned to serve a group of E FENs, relaying their traffic to a Backhaul Network node b , as shown in Fig. 1 . Let j\in\{1,...,E\} denote the generic FEN. The zone to be covered by the FN is defined as a cuboid with dimensions: Z_{x}\times Z_{y}\times Z_{z} . We discretize time T into N time periods evenly spaced with duration \Delta T , so that N=\frac{T}{\Delta T} . Let T_{n} denote the n -th time period, where n\in\{1,...,N\} . We adopt a 3D coordinate system, and the positions of the flying nodes are assumed to be fixed during each time period T_{n} . Therefore, the location of FEN j is given by \ell_{j}(T_{n})=(x_{j}(T_{n}),y_{j}(T_{n}),z_{j}(T_{n})) . Similarly, the HAP location is given by \ell_{h}=(x_{h},y_{h},z_{h}) and the location of the Backhaul Network node by \ell_{b}=(x_{b},y_{b},z_{b}) . The FENs have their mission predefined at the beginning of T , and their positions over T are known and given by the set \mathcal{L} . Let \mathcal{C} denote the set of supported channel bandwidth values by the IEEE 802.11 standard, and let B_{F} be the available bandwidth in the

Read the complete text