PaperlayerView exact revision on arXiv

Source-linked AI summary

A Tutorial on UAVs for Wireless Networks: Applications, Challenges, and Open Problems

Mohammad Mozaffari, Walid Saad, Mehdi Bennis, Young-Han Nam, Merouane Debbah

arXiv:1803.00680v2cs.IT

TL;DR

UAV wireless-network research must address communication challenges that conventional UAV-centric work often neglects, while existing surveys leave broader cellular-network and analytical questions insufficiently covered. This tutorial synthesizes UAV wireless applications, challenges, analytical tools, and representative results to guide system analysis and design, including congestion-aware association and resource-allocation approaches that improve network performance.

  • Problem

    Communication challenges in UAV systems have often been neglected, while prior surveys commonly focus on relay-based or isolated topics rather than fully fledged UAV cellular networks and analytical frameworks.

    Source: §I · p. 1, D. Relevant Surveys on UAVs and Our… · p. 3

  • Method

    The paper develops a comprehensive tutorial covering UAV wireless applications, deployment and channel challenges, energy constraints, analytical frameworks, and representative optimization and game-theoretic approaches.

    Source: 1) Challenges · pp. 9, 15, 3) Representative Result · p. 15, E. Game Theory · p. 22

  • Results

    Representative results show that congestion-aware cell association yields up to a 72% lower average delay, optimal bandwidth allocation produces a 51% shorter flight time, and increasing UAV count reduces hovering energy consumption.

    Source: 3) Representative Result · pp. 15–16

  • Takeaways & Limitations

    UAVs can support high-capacity wireless services through adaptable aerial deployment, including line-of-sight links and combinations with millimeter-wave and massive MIMO technologies.

    Source: 1) Coverage and Capacity Enhancement… · p. 4

    UAV communication designs remain constrained by limited onboard energy and flight duration, and aerial base stations require wireless backhaul connectivity to the core network.

    Source: 1) Challenges · pp. 14–15

Abstract

from arXiv · show

The use of flying platforms such as unmanned aerial vehicles (UAVs), popularly known as drones, is rapidly growing. In particular, with their inherent attributes such as mobility, flexibility, and adaptive altitude, UAVs admit several key potential applications in wireless systems. On the one hand, UAVs can be used as aerial base stations to enhance coverage, capacity, reliability, and energy efficiency of wireless networks. On the other hand, UAVs can operate as flying mobile terminals within a cellular network. Such cellular-connected UAVs can enable several applications ranging from real-time video streaming to item delivery. In this paper, a comprehensive tutorial on the potential benefits and applications of UAVs in wireless communications is presented. Moreover, the important challenges and the fundamental tradeoffs in UAV-enabled wireless networks are thoroughly investigated. In particular, the key UAV challenges such as three-dimensional deployment, performance analysis, channel modeling, and energy efficiency are explored along with representative results. Then, open problems and potential research directions pertaining to UAV communications are introduced. Finally, various analytical frameworks and mathematical tools such as optimization theory, machine learning, stochastic geometry, transport theory, and game theory are described. The use of such tools for addressing unique UAV problems is also presented. In a nutshell, this tutorial provides key guidelines on how to analyze, optimize, and design UAV-based wireless communication systems.

I. INTRODUCTION AND OVERVIEW ON UAVS

UAVs are emerging as flexible aerial base stations and cellular-connected aerial user equipments for wireless networking applications. Their deployment offers opportunities but also requires addressing technical, energy, and regulatory constraints.

UAV roles

  • UAVs can serve as aerial base stations or cellular-connected aerial user equipments for wireless communication applications.Aerial base stations can provide on-demand connectivity, while aerial user equipments include applications such as delivery or surveillance drones.Source: A. Motivation · p. 1
  • Flying base stations can adjust altitude, avoid obstacles, and improve the likelihood of line-of-sight links compared with terrestrial base stations.Source: A. Motivation · p. 1

Technical challenges

  • Key drone-base-station challenges include 3D deployment, resource allocation, flight-time and trajectory optimization, and network planning.Drone-UE scenarios additionally involve handover management, channel modeling, low-latency control, 3D localization, and interference management.Source: A. Motivation · p. 2

UAV classification

  • UAVs are categorized by altitude into HAPs above 17 km and LAPs, with LAPs offering faster deployment for time-sensitive applications and HAPs offering longer endurance.LAPs can typically fly without a permit up to 400 feet under the cited US Federal Aviation Administration regulation.Source: B. UAV Classification · p. 2

Regulation

  • UAV deployment is constrained by privacy, public safety, security, collision avoidance, data protection, and country-specific regulatory requirements.Regulatory criteria include applicability, operational limitations, administrative procedures, technical requirements, and ethical constraints.Source: C. UAV Regulations · p. 2

D. Relevant Surveys on UAVs and Our Contributions

Existing UAV communication surveys are fragmented, often focusing on relay-based ad hoc networks or isolated topics rather than comprehensive cellular and networking systems. This article responds with a holistic tutorial covering applications, challenges, open problems, representative results, and analytical frameworks.

Relevant surveys

  • Earlier surveys mainly address UAV relay stations in ad hoc networks rather than flying base stations or drone-UEs supporting complex networks such as 5G cellular systems.Source: D. Relevant Surveys on UAVs and Our… · p. 3
  • The prior literature is also fragmented across isolated UAV topics and generally does not introduce analytical frameworks for designing and analyzing UAV-based communication systems.Source: D. Relevant Surveys on UAVs and Our… · p. 3

Contributions

Analytical frameworks

II. WIRELESS NETWORKING WITH UAVS: MOTIVATING APPLICATION USE CASES

UAVs are presented as flexible airborne infrastructure for beyond-5G networks, supporting coverage, capacity, high-capacity mmW links, terrestrial-network assistance, and emergency communications. The section also highlights UAVs’ potential to complement existing technologies while requiring careful planning and deployment.

Coverage and Capacity Enhancement

  • UAV-carried flying base stations can complement heterogeneous 5G networks by addressing limitations in D2D, ultra-dense small cells, and mmW systems.LAP-UAVs are described as cost-effective for areas with limited cellular infrastructure and temporary events.Source: 1) Coverage and Capacity Enhancement… · p. 4
  • UAV-enabled mmW communications can establish LoS links and, when combined with massive MIMO, support high-capacity wireless transmission.The paper frames this as a potential dynamic flying cellular network when properly planned and operated.Source: 1) Coverage and Capacity Enhancement… · p. 4

Terrestrial Network Assistance

  • Drones can assist D2D and vehicular networks by disseminating information rapidly, improving link reliability, exploiting transmit diversity, and reducing ground-network transmissions.Their mobility and LoS communications support broadcasting common information to ground devices.Source: 1) Coverage and Capacity Enhancement… · p. 4

Public Safety Scenarios

  • UAVs are proposed for public-safety communications because terrestrial infrastructure can be damaged or overloaded during natural disasters.Their mobility, flexible deployment, and rapid reconfiguration enable on-demand emergency connectivity.Source: 2) UAVs as Flying Base Stations for… · p. 5

3) UAV-assisted Terrestrial Networks for Information Dissemination :

UAVs can enhance information dissemination and connectivity in terrestrial, IoT, and three-dimensional wireless networks through mobility, LoS opportunities, clustering, aerial beamforming, and adaptive deployment. These benefits extend to reliability, energy-efficient uplinks, and mmW communications, while UAV antenna arrays add flexible MIMO capabilities.

Information Dissemination

  • UAVs can assist D2D, ad-hoc, and vehicular networks by disseminating information, extending connectivity, mitigating interference, and improving link reliability.Mobility and LoS opportunities allow drones to serve as flying base stations or broadcast common information to ground devices.Source: 3) UAV-assisted Terrestrial Networks… · p. 5
  • Clustering lets a UAV communicate directly with cluster heads while ground devices use multi-hop communication within clusters.Efficient clustering and UAV mobility can significantly improve terrestrial-network connectivity.Source: 3) UAV-assisted Terrestrial Networks… · p. 6

3D MIMO and mmW Communications

  • Aerial positions and on-demand deployment allow UAVs to support massive MIMO, 3D network MIMO, and mmW communications through three-dimensional beamforming.3D beamforming creates separate beams in three-dimensional space simultaneously.Source: 4) 3D MIMO and Millimeter Wave… · p. 6
  • Drone-based antenna arrays provide airborne beamforming with unconstrained element count, adjustable spacing, and mechanical steering in any 3D direction.Arrays of small UAVs can form arbitrary shapes and perform beamforming for ground users.Source: 4) 3D MIMO and Millimeter Wave… · p. 6
  • UAVs equipped for mmW can combine LoS links, small antennas, massive MIMO, and reconfigurable airborne arrays.These capabilities support high-frequency operation and flexible antenna configurations.Source: 4) 3D MIMO and Millimeter Wave… · p. 6

IoT Communications

  • UAVs can provide reliable and energy-efficient uplink IoT communications by reducing shadowing and blockage through elevated deployment.IoT networks particularly require energy efficiency, ultra-low latency, reliability, and high-speed uplinks because devices are battery limited.Source: 5) UAVs for IoT Communications · pp. 6–7

6) Cache-Enabled UAVs:

UAVs can address mobility-related caching inefficiencies by dynamically storing popular content, tracking users, and moving to serve them. The broader wireless-network role of UAVs also includes drone users, FANETs, backhaul, and smart-city services.

Caching motivation

  • Static SBS caching can fail for mobile users after handovers, requiring replicated content that increases signaling overhead and storage usage.Source: 6) Cache-Enabled UAVs · p. 7

Cache-enabled UAVs

  • Cache-enabled UAVs can dynamically cache popular content, track user mobility, and move to deliver requested services.Source: 6) Cache-Enabled UAVs · p. 7
  • Cloud processors can use mobility patterns and content-request distributions to determine UAV locations and mobility paths, reducing cache-update overhead.Source: 6) Cache-Enabled UAVs · p. 7

Drone users

FANETs and backhaul

E. Summary of Lessons Learned

The paper identifies diverse UAV roles in wireless networks, including aerial base stations, cellular users, FANET nodes, backhaul platforms, and smart-city infrastructure. It summarizes their supported benefits while emphasizing that practical deployment still requires substantial technical research, especially in channel modeling.

Roles and applications

  • UAVs can operate as aerial base stations, cellular user equipments, mobile relays in FANETs, wireless-backhaul platforms, and smart-city systems.Source: E. Summary of Lessons Learned · p. 8

UAV base stations

  • UAV base stations can improve network coverage and capacity, support public-safety dissemination, facilitate millimeter-wave communications, and offload traffic through caching.Source: E. Summary of Lessons Learned · p. 8

Cellular-connected drones

FANETs

Open challenges

Channel modeling

  • Accurate A2G channel models are needed because UAV channel characteristics differ from ground channels and affect coverage and capacity.Source: 1) Challenges · p. 9

3) Representative Result:

UAV air-to-ground communication depends on probabilistic line-of-sight conditions shaped by environment and geometry. Deployment must therefore jointly account for altitude, channel characteristics, user locations, and interference.

Channel modeling

  • The adopted air-to-ground path-loss model makes link conditions depend on UAV and ground-device locations and propagation environment.Links may be line-of-sight or non-line-of-sight across rural, suburban, urban, and high-rise urban settings.Source: 3) Representative Result · p. 9
  • Building-height randomness motivates probabilistic line-of-sight modeling using environment-dependent statistical parameters.Building heights are modeled probabilistically, and the resulting LoS expression uses parameters derived from ITU-R statistics.Source: 3) Representative Result · p. 10
  • The LoS probability increases with elevation angle, because greater elevation reduces blockage and makes the link more likely to be LoS.The model uses environment-dependent constants and the elevation angle θ.Source: 3) Representative Result · p. 10

3D deployment

  • Three-dimensional UAV deployment is challenging because optimal placement depends on geography, ground-user locations, channel characteristics, and altitude.The challenge applies to coverage, capacity, public safety, smart-city, caching, and IoT applications.Source: 1) Challenges · p. 10
  • Prior studies examine UAV deployment for energy-efficient IoT data collection, maximum coverage, sum-rate, user coverage, and relay-rate optimization.The literature includes single- and multi-UAV placement, directional antennas, interference, and cellular-infrastructure supplementation.Source: 2) State of the Art · pp. 10–11

3) Representative Results:

Representative deployment and trajectory studies show that UAV flexibility can improve IoT energy efficiency, while multi-UAV coverage requires altitude reductions to manage interference. Trajectory design must also incorporate mobility, energy, channel, and physical constraints.

3D deployment

  • Optimization theory and facility-location methods determine 3D placements for four UAVs collecting uplink data from IoT devices over a 1 km × 1 km area.The framework targets dynamic deployment and mobility for reliable, energy-efficient IoT communications.Source: 3) Representative Results · p. 11
  • 78% lower average device transmit power is achieved by optimally deploying UAVs rather than using pre-deployed aerial base-station locations.Uplink transmit power also decreases as the number of UAVs increases.Source: 3) Representative Results · p. 11
  • Increasing the number of UAVs from 3 to 6 reduces optimal altitude from 2000 m to 1300 m to limit overlapping coverage and interference.Higher antenna beamwidths also require lower UAV altitudes in the considered deployment.Source: 3) Representative Results · p. 11

Trajectory optimization

  • Trajectory optimization is important for smart cities, drone-UE, and caching applications, but must account for flight time, energy, user demand, and collision avoidance.Continuous trajectory optimization also involves channel variation, UAV dynamics, and flight constraints.Source: C. Trajectory Optimization · pp. 11–12
  • Wireless-communication trajectory studies optimize objectives including minimum average rate, energy consumption, rate, reliability, target detection, and ad-hoc connectivity.Prior work spans control and navigation, localization, and wireless-communication perspectives.Source: 1) State of the Art · p. 12

2) Representative Result:

UAV trajectory and altitude decisions create measurable energy and rate tradeoffs in wireless networks. Representative results show substantial energy savings from optimized paths and altitude-dependent sum-rate maxima shaped by interference.

Trajectory optimization

  • Optimal path planning reduces average total drone energy consumption by 74% compared with predefined, non-optimal paths.The experiment uses five drones collecting data from 500 IoT devices in a time-varying 1 km × 1 km IoT network.Source: 2) Representative Result · p. 12
  • Trajectory design must jointly consider wireless metrics such as throughput and coverage together with UAV energy constraints.Joint trajectory-communication optimization is described as challenging but potentially performance-improving.Source: 2) Representative Result · p. 12

Performance analysis

  • Performance analysis of UAV systems must address coverage probability, throughput, delay, reliability, altitude, mobility, and distinct channel characteristics.UAV energy limitations further distinguish this analysis from conventional ground networks.Source: 1) Challenges · p. 13
  • The UAV-D2D model derives tractable coverage and rate expressions for static and mobile UAVs using a spatially distributed user network.Average sum-rate is evaluated against UAV altitude and D2D transmitter-receiver separation.Source: 3) Representative Result · p. 13
  • Average sum-rate is maximized at about 300 m for d0 = 20 m and 400 m for d0 = 30 m; between 300 m and 1300 m, LoS interference reduces performance.The optimal altitude therefore depends on the D2D-pair distance.Source: 3) Representative Result · pp. 13–14

E. Cellular Network Planning and Provisioning with UAVs

UAV cellular network planning must handle dynamic positioning, association, signaling, interference, resource constraints, and wireless backhaul. Congestion-aware cell association can substantially reduce delay, while limited energy and backhaul remain important boundaries.

Network planning

  • UAV network planning covers base-station positioning, traffic estimation, frequency allocation, cell association, backhaul, signaling, and interference mitigation.Unlike static terrestrial networks, UAV systems require dynamic signaling to track changing UAV locations and numbers.Source: 1) Challenges · p. 14

Backhaul connectivity

  • Wireless backhaul remains a challenging design problem because aerial base stations require connections to the core network.WiFi offers lower cost and latency than satellite, while millimeter-wave and FSO links can provide high-capacity backhaul when LoS is available.Source: 1) Challenges · p. 14

Cell association

  • Cell planning is formulated to minimize average transmission delay for four UAV base stations and two ground macro base stations in a 4 km × 4 km area.The study uses truncated-Gaussian hotspot users and optimizes cell association.Source: 3) Representative Result · pp. 14–15
  • Up to 72% lower average delay is obtained with congestion-aware cell association than with classical SNR-based association.The proposed approach avoids highly loaded cells and is more robust against network congestion.Source: 3) Representative Result · p. 15

Resource management

  • Network planning affects throughput, delay, operational cost, and energy consumption in UAV-enabled wireless networks.Resource management must also accommodate time-varying interference, traffic, mobility, heterogeneous spectrum, and UAV energy constraints.Source: 3) Representative Result · p. 15, 1) Challenges · p. 15

Energy efficiency

  • Limited onboard energy restricts transmission, mobility, control, processing, and payload operations, making drone flight duration typically short.These constraints limit flight and hover time and hinder long-term continuous wireless coverage.Source: 1) Challenges · p. 15

2) State of the Art:

UAV wireless systems require joint management of propulsion and communication energy, bandwidth, and flight time. Representative results show that allocation and deployment choices create measurable efficiency gains but also tradeoffs.

Energy models

  • UAV energy consumption comprises communication-related energy and propulsion energy, with propulsion typically consuming significantly more.Communication energy covers transmission, computation, and processing; propulsion energy covers movement and hovering.Source: 2) State of the Art · pp. 15–16
  • Fixed-wing and rotary-wing UAVs use different propulsion energy models whose parameters depend on vehicle and aerodynamic characteristics.The fixed-wing model uses constants tied to weight, wing area, and air density, while rotary-wing parameters include rotor and fuselage properties.Source: 2) State of the Art · p. 16

Bandwidth allocation

  • 51% shorter flight time is achieved by optimal bandwidth allocation compared with equal allocation for serving 50 users requiring 100 Mb each.The scenario uses five UAV base stations over a 1 km × 1 km area, and optimal allocation accounts for user load and location.Source: 3) Representative Result · p. 16

Energy-bandwidth tradeoff

  • 53% lower total hovering energy consumption results when the number of UAVs increases from 2 to 6 in an interference-free scenario.More UAVs partition cells into smaller regions, reducing user distances and hover time, but require proportionally more bandwidth.Source: 3) Representative Result · p. 16, Figure 14 · p. 16

Resource management

  • Resource allocation strategies and UAV energy constraints significantly affect UAV communication-system performance.Available energy, bandwidth, and time must therefore be managed jointly for efficient wireless-network operation.Source: 3) Representative Result · p. 17

G. Drone-UEs in Wireless Networks

Drone-UEs extend UAV applications into cellular networks but introduce coexistence, mobility, handover, interference, and low-latency challenges. Representative results show that increasing drone-UE density can reduce ground-user connectivity through LoS interference.

Applications

  • Drone-UEs support delivery, surveillance, and virtual-reality applications as flying users within cellular networks.They capture information and transmit it to remote users, but cellular networks were primarily designed for ground users.Source: 1) Challenges · p. 17

Design challenges

  • Drone-UE operation must address mobility, LoS interference, handover, energy constraints, and low-latency control.Downtilted ground-base-station antennas designed for terrestrial users can limit UAV-UE performance and motivate adaptive beamforming or UAV-aware designs.Source: 1) Challenges · p. 17, H. Summary of Lessons Learned · p. 18

Interference

  • Ground-user connectivity decreases as the number of drone-UEs increases because of dominant line-of-sight interference.In the modeled scenario, drone-UEs are uniformly deployed at 100 m altitude over a disk of radius 1000 m.Source: 3) Representative Result · p. 17
  • 18% lower connectivity probability at 150 m occurs when drone-UEs increase from 5 to 15.The result motivates effective interference-management techniques for drone-UE scenarios.Source: 3) Representative Result · p. 17

Design challenges

  • UAV-enabled network design must account for channel models, 3D placement, trajectories, performance metrics, network planning, and energy efficiency.Relevant metrics include coverage probability, area spectral efficiency, reliability, and latency, while deployment and trajectory depend on users, obstacles, QoS, and energy.Source: H. Summary of Lessons Learned · p. 18

IV. OPEN PROBLEMS AND FUTURE OPPORTUNITIES FOR UAV-BASED WIRELESS COMMUNICATION AND NETWORKING

The paper identifies open problems spanning channel modeling, deployment, trajectory optimization, and performance analysis. These problems arise because UAV networks must represent mobility, altitude, interference, heterogeneous aerial-terrestrial systems, and environmental conditions.

Channel modeling

  • More realistic air-to-ground channel models require broader real-world measurements across urban, rural, weather, and operational environments.Existing efforts are often limited to a single UAV or specific environments, while air-to-air models must capture time variation, Doppler, and multipath fading.Source: A. UAV Channel Modeling · pp. 18–19

Deployment

  • Optimal 3D UAV placement remains open when aerial systems coexist with terrestrial networks and mutual interference.Deployment solutions must account for UAV-specific features and interactions between aerial and terrestrial systems.Source: B. UAV Deployment · p. 19
  • Open deployment problems include jointly optimizing 3D locations with bandwidth allocation, cell association, and obstacle-aware coverage.The objectives include minimizing maximum latency, minimizing total flight time, and maximizing the coverage area or covered users.Source: B. UAV Deployment · p. 19

Trajectory optimization

  • UAV trajectory optimization must address throughput, energy efficiency, spectral efficiency, delay, dynamic conditions, and UAV type.The paper identifies trajectory design as an ongoing technical problem despite substantial prior work.Source: C. UAV Trajectory Optimization · p. 19

Performance analysis

  • Tractable coverage and spectral-efficiency expressions are needed for heterogeneous networks containing aerial and terrestrial users and base stations.The paper calls for more complete performance characterization in terms of coverage and capacity.Source: D. Performance Analysis · p. 20

E. Planning Cellular Networks with UAVs

Cellular network planning with UAVs must coordinate coverage, backhaul, spectrum, provisioning, mobility, interference, and signaling overhead. The paper connects these design problems to centralized optimization and analytical tools including optimal transport theory.

Network planning

  • UAV network planning must determine coverage, backhaul-aware deployment, frequency use, dynamic provisioning, mobility adaptation, and signaling overhead.The objectives include coverage and QoS while accounting for irregular areas, aerial-terrestrial coexistence, and highly mobile drone-UEs.Source: E. Planning Cellular Networks with UAVs · p. 20

Resource management

  • Resource management must dynamically coordinate bandwidth, energy, transmit power, flight time, and the number of UAVs.The paper also identifies interference mitigation, dynamic spectrum sharing, and suitable frequency-band selection as open problems.Source: F. Resource Management in UAV Networks · p. 20

Drone-UE planning

  • Drone-UE planning must address mobility, line-of-sight interference, handover, energy constraints, and low-latency control.These requirements create open problems in interference mitigation and dynamic handover for cellular-connected UAVs.Source: G. Drone-UEs Scenarios · p. 20, H. Lessons Learned · p. 20

Centralized optimization

  • Early UAV-base-station deployments are envisioned to rely on centralized control, motivating centralized optimization at a cloud or macrocell base station.This approach is especially relevant to cellular-capacity enhancement while operators retain network control during early trials.Source: A. Centralized Optimization Theory for… · pp. 20–21

Analytical tools

  • Optimal transport theory can provide tractable solutions for user association, resource allocation, and flight-time optimization.In semi-discrete settings, its transport map partitions a continuous device distribution and assigns regions to discrete communication points.Source: B. Optimal Transport Theory for UAV… · p. 21

C. Performance Analysis using Stochastic Geometry

The paper surveys analytical and decision-making tools for UAV-enabled wireless networks, including stochastic geometry, machine learning, and game theory. These tools target performance analysis, adaptive control, resource management, trajectory optimization, cooperation, and network-scale efficiency.

Stochastic geometry

  • Stochastic geometry models users and base stations as point processes to evaluate coverage, rate, throughput, and delay in three-dimensional UAV networks.The paper presents it as an extension of techniques used for two-dimensional heterogeneous cellular networks.Source: C. Performance Analysis using… · p. 21

Machine learning

  • Machine learning can adapt UAV positions, flight directions, and motion control while autonomously optimizing trajectories in dynamic environments.Neural networks and data analytics can also predict user mobility and load distribution for deployment and operation.Source: D. Machine Learning · pp. 21–22

Game theory

  • Game theory supports distributed resource management and trajectory optimization when each UAV makes decisions according to its own objective function.The paper further motivates stochastic differential and mean-field games for jointly optimizing communication and control and analyzing scaling with user numbers.Source: E. Game Theory · p. 22

Cooperation

  • Coalitional game theory is proposed for dynamically forming and coordinating UAV swarms, while contract theory can design incentive mechanisms and matching.These tools address cooperative behavior as an open problem in UAV communications.Source: E. Game Theory · p. 22

Analytical framework

  • The tutorial organizes challenges, open problems, references, and analytical tools for analyzing, optimizing, and designing UAV-enabled wireless networks.Its concluding synthesis also identifies optimization theory, stochastic geometry, optimal transport, machine learning, and game theory as relevant tools for deployment, performance, association, load balancing, and related problems.Source: F. Lessons Learned and Summary · p. 22, §VI · p. 22

Related papers

Loading 1803.00680v2…