PaperlayerView exact revision on arXiv

Source-linked AI summary

UAV Communications for 5G and Beyond: Recent Advances and Future Trends

Bin Li, Zesong Fei, Yan Zhang

arXiv:1901.06637v1cs.NI

TL;DR

The paper addresses UAV communications for upcoming 5G/B5G networks and surveys their roles as aerial platforms. It reviews space-air-ground integration and UAV-enabled 5G techniques, reporting potential energy savings and proactive caching benefits while identifying deployment challenges.

  • Problem

    UAV communications are examined as a component of upcoming 5G/B5G wireless networks, including the challenge of safely managing high densities of low-altitude UAVs.

  • Method

    The paper surveys UAVs as aerial communication platforms, including base stations and mobile relays, and categorizes research across physical-layer, network-layer, and joint communication, computing, and caching techniques.

  • Results

    Simulations indicate that energy consumption can be reduced by up to 50%, while proactive caching shows potential for addressing limited UAV endurance.

  • Takeaways & Limitations

    Space-air-ground integrated networks and UAV communications are presented as a platform for future network standardization that is expected to be reliable and real-time.

Abstract

from arXiv · show

Providing ubiquitous connectivity to diverse device types is the key challenge for 5G and beyond 5G (B5G). Unmanned aerial vehicles (UAVs) are expected to be an important component of the upcoming wireless networks that can potentially facilitate wireless broadcast and support high rate transmissions. Compared to the communications with fixed infrastructure, UAV has salient attributes, such as flexible deployment, strong line-of-sight (LoS) connection links, and additional design degrees of freedom with the controlled mobility. In this paper, a comprehensive survey on UAV communication towards 5G/B5G wireless networks is presented. We first briefly introduce essential background and the space-air-ground integrated networks, as well as discuss related research challenges faced by the emerging integrated network architecture. We then provide an exhaustive review of various 5G techniques based on UAV platforms, which we categorize by different domains including physical layer, network layer, and joint communication, computing and caching. In addition, a great number of open research problems are outlined and identified as possible future research directions.

I. INTRODUCTION

5G/B5G networks face rising traffic, heterogeneous connectivity needs, and difficult infrastructure deployment in emergencies. UAVs offer mobile communication, networking, computing, and caching capabilities, but their operation involves altitude, energy, and coverage trade-offs.

  • Emergency deployment of terrestrial infrastructure can be economically infeasible and challenging in sophisticated, volatile environments.
  • UAVs can act as aerial base stations or mobile relays that provide or enhance services during high traffic demand and overloaded conditions.
  • Multi-UAV networks can rapidly recover or expand communications, while UAVs can also offload IoT tasks and cache popular content at the network edge.
  • UAV communications benefit from probable LoS links and dynamic positioning, while altitude creates a trade-off between LoS probability, path loss, and cell coverage.
  • UAV design is constrained by size, weight, and power, which affect altitude, communication, coverage, computation, and endurance capabilities.

A. Existing Surveys and Tutorials

The paper reviews prior UAV-communication surveys and motivates a broader 5G/B5G-focused synthesis. It organizes this survey around integrated space-air-ground networks, 5G techniques, and open research challenges.

  • Existing Surveys and Tutorials: Existing surveys cover UAV applications, communication issues, cybersecurity, routing, channel modeling, and wireless charging.The surveyed literature includes both broad UAV-network perspectives and focused treatments of specific technical areas.
  • Existing Surveys and Tutorials: Prior work also examines airborne communication networks across LAP, HAP, and integrated network paradigms.These studies address primary mechanisms and protocols for different airborne communication configurations.
  • Paper Contributions and Organization: The survey addresses the need to reflect on current UAV-communication achievements and identify future 5G/B5G research trends.The authors motivate an overview of emerging studies integrating 5G technologies with UAV communication networks.
  • Paper Contributions and Organization: It introduces a space-air-ground integrated network architecture and highlights open research challenges.The architecture spans space-based, air-based, and ground-based segments intended to support flexible end-to-end services.
  • Paper Contributions and Organization: The review categorizes UAV integration with 5G technologies into physical-layer, network-layer, and joint communication, computing, and caching domains.The paper then identifies open problems for future research across these areas.
  • Paper Contributions and Organization: The proposed B5G architecture incorporates mmWave, energy harvesting, NOMA, D2D communication, MEC, and caching to describe interconnectivity among emerging technologies.The architecture is presented as a platform for future network standardization with expected reliability and real-time operation.

B. Potential Challenges

The survey identifies technical challenges spanning channel modeling, deployment, mobility, interference, energy, backhaul, security, and real-time information exchange in space-air-ground integrated networks.

  • Channel Modeling: UAV-to-ground channels depend on altitude, elevation angle, UAV type, and propagation environment, making generic modeling difficult.Accurate channel models require measurements and simulations across diverse environments.
  • Deployment: UAV and satellite mobility complicates integrated-network deployment, including 3D positioning, handover reduction, collision avoidance, and constrained satellite links.Satellite systems also face limited power and bandwidth, transmission delay, and severe high-frequency fading.
  • Path Planning: Swarm path planning must balance proximity to ground users for low delay against maintaining inter-UAV connectivity and area coverage.Dynamic trajectory control is identified as necessary for increasing end-to-end link connections while preserving target-area coverage.
  • Interference Dynamics and Limited Energy: Multi-tier UAV networks face co-channel interference and mobility-induced Doppler shifts, while battery dependence limits UAV operation time.Interference management and advanced charging technologies are highlighted as open needs.
  • Backhaul Cellular Communication and Network Security: Limited wireless backhaul can become a QoS bottleneck, and heterogeneous multi-tier networks with wireless LoS propagation are vulnerable to malicious attacks.Security concerns include protecting SDN controllers and detecting unauthorized airspace intrusion.
  • Survey Scope: The survey reviews UAV communications in 5G/B5G across physical-layer, network-layer, and joint communication, computing, and caching perspectives.These reviews are intended to provide guidelines for understanding the referenced literature.

III. PHYSICAL LAYER TECHNIQUES

UAV communications are reviewed as a means to complement terrestrial networks, especially for temporary capacity and coverage needs. The physical-layer survey covers mmWave, NOMA, cognitive radio, and energy harvesting technologies.

  • UAV-Assisted Cellular Networks: UAV-BSs can complement terrestrial cellular systems by providing additional capacity to hotspot areas during temporary events.Their portable transceivers and signal-processing capabilities support dynamic connections and coverage.
  • Applications: UAV communications are considered for emergency and public-safety situations when terrestrial networks are damaged or not fully operational.
  • Physical-Layer Technologies: The physical-layer review focuses on five candidate technologies: mmWave communication, NOMA transmission, cognitive radio, and energy harvesting.The passage introduces these technologies as key physical-layer candidates, while the detailed list names four categories explicitly.

A. mmWave UAV-Assisted Cellular Networks

The survey examines mmWave UAV cellular networks and NOMA transmission as physical-layer approaches for high-bandwidth and multi-user UAV communications. It emphasizes propagation, mobility, beamforming, resource allocation, and user fairness challenges.

  • mmWave UAV-Assisted Cellular Networks: mmWave offers abundant bandwidth and short wavelengths that allow compact UAV antenna arrays, but propagation loss and blockage remain major challenges.Directional beamforming can counter path loss and atmospheric absorption or scattering losses.
  • mmWave UAV-Assisted Cellular Networks: UAV movement intensifies mmWave challenges by requiring efficient beamforming training, beam tracking, Doppler handling, and joint position and user discovery.
  • mmWave UAV-Assisted Cellular Networks: Reported mmWave studies investigate hierarchical beamforming, spatial-division access, secrecy, relay placement, channel tracking, resource allocation, deployment, and propagation modeling.A UAV relay was reported to provide more accurate and efficient solutions than an existing relay method.
  • UAV NOMA Transmission: The section identifies channel-condition estimation and interlayer-interference elimination as continuing requirements for UAV NOMA implementation.
  • UAV NOMA Transmission: NOMA multiplexes multiple users in the power domain over the same time-frequency carrier, using power allocation and decoding order with successive interference cancellation.UAV NOMA studies optimize altitude, power, coverage, throughput, outage performance, and fairness across two-user and multi-user settings.
  • UAV NOMA Transmission: NOMA-based UAV studies address fairness, outage probability, throughput, operation range, and max-min rate through altitude, power, beamwidth, bandwidth, and trajectory optimization.Limited-feedback schemes use user distance or angle information for user ordering, with angle-based feedback reported as superior to distance-based feedback.

C. Cognitive UAV Networks

Cognitive UAV communications address spectrum scarcity by enabling UAVs to share terrestrial spectrum, while managing strong-LoS interference and protecting primary receivers. Existing studies optimize spectrum sharing, UAV trajectories, power, density, and energy efficiency.

  • Motivation: Spectrum scarcity arises from rapidly growing mobile-device usage and coexistence among Bluetooth, WiFi, LTE, and cellular networks.Dynamic utilization of existing frequency bands is therefore identified as necessary for UAV communications.
  • Cognitive UAV communications: Cognitive UAV communications enable UAVs and terrestrial mobile devices to coexist in the same frequency band.Strong LoS UAV-to-ground links can cause severe interference to existing terrestrial devices.
  • Existing approaches: Prior work jointly optimized UAV trajectory and transmit power to maximize cognitive-UAV throughput while keeping interference at primary receivers below a tolerable level.This formulation combines throughput maximization with interference protection.
  • Existing approaches: Stochastic-geometry analysis derived drone-cell coverage probability and the optimal UAV-BS density for maximizing throughput under spectrum sharing.The work considered single-tier drone-cell sharing and sharing between drone-cells and traditional two-dimensional cellular networks.
  • Existing approaches: An underlay cognitive-radio UAV system maximized energy efficiency by opportunistically sharing primary spectrum for UAV-to-ground transmission.The stated objective was to support effective and long-time UAV operations.

D. Energy Harvesting UAV Networks

Energy harvesting and energy-aware control target UAV endurance, which is constrained by battery capacity and propulsion, communication, and operational demands. Research spans scheduling, deployment, reinforcement learning, solar harvesting, wireless charging, and joint throughput optimization, while practical harvesting uncertainty remains important.

  • Energy constraints: Battery life is usually less than 30 minutes, restricting UAV flight or hovering time and challenging stable, sustainable communication services.UAV energy is consumed by flight control, sensing, data transmission, and applications.
  • Energy-aware operation: Energy-aware studies minimize UAV energy consumption or extend network lifetime through transmission scheduling, trajectory design, deployment, and coverage-power optimization.One cooperative-relaying scheduling study saved up to 50% energy in simulations.
  • Energy-aware operation: Deep reinforcement learning has been used to study UAV movement energy consumption while maintaining fair communication coverage and network connectivity.Other formulations jointly minimize flying and communication energy or decompose coverage maximization and power minimization.
  • Solar harvesting: Solar-powered communication studies jointly consider harvesting, aerodynamic consumption, onboard storage, QoS, position, power, subcarriers, and mobility to optimize throughput.One simulation found that a UAV could harvest more solar energy when flying right above the clouds.
  • Practical limitations: Solar harvesting is weather-dependent and unpredictable, and most current works did not consider the resulting uncertainty from random energy arrivals.Alternative approaches include rotational energy harvesting and RF wireless power transfer from recharging stations.

IV. NETWORK LAYER TECHNIQUES

Network-layer research integrates UAVs into multi-tier architectures containing drone cells, ground small cells, and device-to-device communications. The surveyed studies address QoS coordination, coverage, interference, demand-aware deployment, traffic offloading, mobility, and multi-tier aerial-network design.

  • Network architecture: Future networks combine drone-cell, ground-small-cell, and user-device tiers, creating new network-layer coordination issues.Specific strategies are necessary to coordinate the QoS of nodes across integrated tiers.
  • UAV-assisted HetNets: UAV-assisted HetNets use flexible drone-cell deployment and mobility to bring users closer to short-range LoS links and serve highly mobile users with high data-rate demands.The architecture includes one macro base station and multiple drone cells.
  • Aerial HetNets: Multi-tier UAV networks introduce challenges including drone energy consumption and interference management, with performance examined across high-rise, suburban, and dense urban environments.Traffic offloading from congested ground base stations is another motivation for aerial HetNets.
  • UAV-assisted HetNets: Existing HetNet studies optimize UAV placement, bandwidth, interference coordination, traffic prediction, association, latency, and demand-based assignment.Objectives include dynamic coverage, network capacity, spectral efficiency, power minimization, and reduced delay.
  • UAV-assisted HetNets: UAV-BSs can provide coverage and broadband connectivity in disaster-affected regions, while machine-learning and neural frameworks target demand matching, traffic prediction, and delay reduction.These studies also consider energy limitations and mobility in deployment and association decisions.
  • Aerial HetNets: Aerial HetNets comprise multiple UAV types and tiers whose construction depends on user and service density, analogous to terrestrial heterogeneous networks.Research also investigates mapping UAVs to demand areas and the feasibility of multi-tier architectures across urban environments.

B. Combined UAVs and D2D Communications

Combining UAVs with D2D communications creates flexible aerial-terrestrial networks that can reuse spectrum and support relaying, content delivery, emergency connectivity, and social networking. The main unresolved issue is coordinating interference and mobility across shared-spectrum links.

  • Network model: D2D communications commonly reuse licensed spectrum through underlay links, while UAVs add an aerial dimension for rapidly constructing D2D-enabled networks.This combination introduces shared-spectrum interference-management challenges.
  • Coexistence and resource allocation: UAV-D2D research analyzes coexistence under static and mobile UAV scenarios and assigns radio channels while accounting for high mobility of UAVs and D2D nodes.UAVs may serve as local content servers and aerial D2D nodes.
  • Coexistence and resource allocation: A distributed anti-coordination game algorithm was designed for channel assignment in multi-UAV wireless networks with D2D communications.Other work analyzes downlink and D2D coverage probability and optimizes UAV altitude for ground-network capacity.
  • Applications: Using UAVs to discover potential D2D devices for emergency transmissions reduced device energy consumption and increased network capacity in simulations.The UAV thereby supports establishment of D2D transmissions as an emergency communication network.
  • Applications: Full-duplex UAV relaying jointly designs transmit power and UAV trajectory to enable spectrum sharing between aerial UAV and terrestrial D2D communications.Other work applies multiple D2D peers to UAV-supported social networking while considering physical interference and social connections.

C. Software Defined UAV Networks

Software-defined networking gives UAV networks centralized visibility and flexible control, supporting reconfiguration, routing, handover, and resource allocation. The section also reviews UAV-based MEC, where computation offloading can reduce device energy use and network traffic.

  • C. Software Defined UAV Networks: SDN improves controllability and visibility of network components, enabling flexible infrastructure management and resource allocation.Its architecture separates control and data planes and uses controllers to coordinate UAV network functions.
  • C. Software Defined UAV Networks: UAVs can act as SDN data-plane switches while controllers collect context and direct network reconfiguration across UAVs.This supports changing protocols, creating paths, and integrating or disintegrating drone-cells.
  • C. Software Defined UAV Networks: SDN-based UAV research addresses handover latency, controller placement, resilient multipath routing, overload relief, and QoS-aware resource allocation.Existing studies also consider trade-offs between control-information overhead and end-to-end delay.
  • UAV-based MEC: Computation offloading through flying UAVs can save mobile-device energy, reduce fixed-cloud traffic, and lower application execution latency.UAVs provide flexible connectivity and edge computing capacity, while remote execution can improve battery performance.
  • UAV-based MEC: UAV-mounted MEC lets mobile devices offload intensive computation to nearby aerial edge servers instead of executing tasks locally.Local execution consumes device resources and energy, while offloading transfers execution to the UAV MEC server.
  • UAV-based MEC: UAV MEC remains constrained by limited onboard computation resources, restricting efficient execution of complex applications.Remote edge or cloud offloading is presented as a way to enhance effective UAV computation capability.

B. Caching in the Sky

UAV caching addresses rapidly growing mobile traffic and backhaul pressure by placing popular content near moving users. The surveyed schemes use dynamic placement, prediction, cooperation, and optimization to reduce latency, traffic, or UAV energy costs.

  • Caching in the Sky: Explosive mobile traffic burdens backhaul links, motivating intelligent caching of popular content at UAVs, relays, or D2D devices.Repeated requests can then be served closer to users without duplicate backhaul transmissions.
  • Caching in the Sky: Flying base stations can dynamically cache popular content, track user mobility, and serve users directly during peak-load periods.This reduces transmission latency and backhaul traffic while supporting flexible caching for mobile users.
  • Caching in the Sky: UAV-assisted edge caching can improve user QoE and reduce required backhaul capacity through UAV- and D2D-based content distribution.Content may be cached at UAV base stations or cooperatively stored and shared among nearby users.
  • Caching in the Sky: Learning-based caching schemes predict content demand and mobility to determine UAV locations, cached contents, and licensed or unlicensed resource allocation.The reviewed approaches include dynamic resource allocation and conceptor-based echo-state prediction.
  • Caching in the Sky: Proactive caching showed great potential for overcoming limited UAV endurance by pre-delivering files to cooperatively caching ground nodes.Users retrieve files locally or from neighboring devices through D2D communications.
  • Caching in the Sky: UAV caching research also formulates joint content-placement and service-location optimization to trade off user service probability against transmission overhead.One multi-UAV formulation models the problem as a caching game with an obtainable optimal solution.

VI. FUTURE RESEARCH DIRECTIONS

The survey identifies energy, channel modeling, mobility, integration, IoT coordination, and security as open challenges for practical UAV-assisted networks. It emphasizes that UAV endurance, rapidly changing links, heterogeneous segments, and cyber-attacks constrain deployment.

  • Energy and charging: Energy limitation is the bottleneck in UAV communications, while harvesting efficiency is reduced by long distance and random energy arrivals.Energy beamforming and distributed wireless power transfer are identified as promising charging directions.
  • Mobility and routing: High-speed UAV mobility frequently disconnects neighboring links, so traditional routing protocols cannot work well in FANETs.Flight control must maintain service quality while multi-UAV operation also requires collision avoidance.
  • Channel modeling: Realistic air-to-ground and satellite-to-UAV channel models remain needed because obstacles, foliage, temperature, wind, mobility, and Doppler affect propagation.Existing satellite-to-UAV propagation models are described as immature and lacking detailed effects.
  • Integrated networks: Space-air-ground integration requires cooperative incentives, cross-layer protocols, and scalable interfaces to coordinate heterogeneous networks and ensure link reliability.Control must jointly address spectrum allocation, link scheduling, and protocol design for low-latency, reliable delivery.
  • UAV-IoT synergy: IoUAV endurance and reliability are fundamentally limited by small practical battery capacity, while mobility adds further energy consumption.The section calls for energy-aware coordination between UAVs and IoT systems.
  • Security and privacy: Open security challenges arise because UAVs are unattended, easily captured or attacked, and expose data and control signals over radio links.The survey calls for lightweight protection against eavesdropping, man-in-the-middle attacks, and related threats.

F. Space-Air-Ground Integrated Vehicular Networks

Space-air-ground integrated vehicular networks combine ground, satellite, and UAV connectivity to support urban, rural, remote, congested, and poorly covered areas. Their development requires coordinated control, realistic channels, and joint networking, computing, and caching.

  • F. Space-Air-Ground Integrated Vehicular Networks: Ground, satellite, and UAV segments can respectively support urban or suburban data rates, rural or remote connectivity, and coverage expansion or information collection.This division of roles motivates integrating space-air-ground communications with vehicular networks.
  • F. Space-Air-Ground Integrated Vehicular Networks: High satellite and UAV mobility continually changes propagation conditions through free-space path loss and Doppler effects.These dynamics complicate interworking between space-air-ground and vehicular networks.
  • F. Space-Air-Ground Integrated Vehicular Networks: Reliable low-latency delivery requires coordinated spectrum allocation, link scheduling, protocol design, and control across space-air-ground segments.The architecture must provide mechanisms for seamless integration, information exchange, and cooperation among heterogeneous networks.
  • F. Space-Air-Ground Integrated Vehicular Networks: Joint networking, computing, and caching should balance operating costs such as energy consumption against benefits such as decreased latency.The survey notes that separate treatment of these techniques is insufficient for next-generation smart IoT requirements.
  • F. Space-Air-Ground Integrated Vehicular Networks: Big-data prediction can support proactive actions that avoid network faults or service failures when network events are predicted accurately.The surveyed ecosystem places services, content, and functions at network edges and uses data for caching and prediction.
  • F. Space-Air-Ground Integrated Vehicular Networks: The survey concludes that UAV communications in 5G/B5G span physical-layer, network-layer, and joint communication, computing, and caching techniques.It also identifies open research issues and presents the space-air-ground integrated network as a B5G architecture.
Loading 1901.06637v1…