PaperlayerView exact revision on arXiv

Source-linked AI summary

Survey on UAV Cellular Communications: Practical Aspects, Standardization Advancements, Regulation, and Security Challenges

Azade Fotouhi, Haoran Qiang, Ming Ding, Mahbub Hassan, Lorenzo Galati Giordano, Adrian Garcia-Rodriguez, Jinhong Yuan

arXiv:1809.01752v2cs.NI

TL;DR

UAV cellular communications raise practical, interference, regulatory, and security questions as consumer drone use expands. The paper surveys UAV types and cellular integration across standardization, flying infrastructure, prototypes, regulation, and cyber-physical security, finding substantial activity but an early-stage field with continuing research and development.

  • Problem

    Integrating UAVs into cellular systems requires addressing aerial-user interference, practical flying infrastructure, regulation, and cyber-physical security across emerging use cases.

  • Method

    The paper comprehensively surveys practical UAV characteristics and complementary developments from academia, industry, standardization, and regulation.

  • Results

    The survey finds that 3GPP identified major interference issues, vendors demonstrated UAV-mounted flying base stations, and regulation and security activities are growing.

  • Takeaways & Limitations

    UAV cellular communication is at an early development stage, with ongoing progress across standards, prototypes, research, regulation, and security.

Abstract

from arXiv · show

The rapid growth of consumer Unmanned Aerial Vehicles (UAVs) is creating promising new business opportunities for cellular operators. On the one hand, UAVs can be connected to cellular networks as new types of user equipment, therefore generating significant revenues for the operators that can guarantee their stringent service requirements. On the other hand, UAVs offer the unprecedented opportunity to realize UAV-mounted flying base stations that can dynamically reposition themselves to boost coverage, spectral efficiency, and user quality of experience. Indeed, the standardization bodies are currently exploring possibilities for serving commercial UAVs with cellular networks. Industries are beginning to trial early prototypes of flying base stations or user equipments, while academia is in full swing researching mathematical and algorithmic solutions to address interesting new problems arising from flying nodes in cellular networks. In this article, we provide a comprehensive survey of all of these developments promoting smooth integration of UAVs into cellular networks. Specifically, we survey (i) the types of consumer UAVs currently available off-the-shelf, (ii) the interference issues and potential solutions addressed by standardization bodies for serving aerial users with the existing terrestrial base stations, (iii) the challenges and opportunities for assisting cellular communications with UAV-based flying relays and base stations, (iv) the ongoing prototyping and test bed activities, (v) the new regulations being developed to manage the commercial use of UAVs, and (vi) the cyber-physical security of UAV-assisted cellular communications.

I. INTRODUCTION

The survey examines how rapidly growing consumer UAV use can connect to cellular networks and support UAV-mounted flying relays or base stations. It synthesizes practical UAV characteristics, standardization, research, prototypes, regulation, and security developments.

  • Standardization: 3GPP studies aerial UAVs as a new UE type, but their enhanced line-of-sight links can significantly increase interference.New strategies are therefore needed to accommodate aerial and ground UEs in the same system.
  • Flying cellular infrastructure: UAV-mounted flying relays and base stations can dynamically reposition to boost coverage, spectral efficiency, and user quality of experience.Major vendors have field-trialled prototypes demonstrating this concept.
  • Survey scope: The survey addresses a practical gap by integrating consumer UAV types, cellular standardization, flying infrastructure, prototypes, regulation, and cyber-physical security.Earlier surveys focused on UAV ad hoc networks, application communications, multi-tier drone networks, or wireless networks generally.
  • UAV characteristics: Drone payload determines which cellular equipment can be carried, ranging from consumer UEs below 1 kilogram to BSs or RRHs requiring at least a few kilograms.Greater payload generally requires a larger drone, higher battery capacity, and shorter airborne duration.
  • UAV types: Multi-rotor drones provide precise deployment and hovering for cellular assistance, whereas fixed-wing drones offer greater speed, energy efficiency, and payload capacity.Multi-rotors consume significant power and have limited mobility; fixed-wing drones require runways and cannot hover.
  • UAV types: Hybrid fixed/rotary-wing drones combine vertical takeoff, gliding travel, and hovering through a compromise between the two main flying mechanisms.The Parrot Swing illustrates this transition between flight modes.

C. Range and Altitude

Range, altitude, speed, endurance, and power supply constrain UAV deployment for cellular communications. Low-altitude platforms are commonly favored, while high-altitude platforms offer broader coverage but can create severe interference and deployment challenges.

  • Range and altitude: Drone control range spans tens of meters for small drones to hundreds of kilometers for large ones, while altitude is the maximum achievable height.Altitude remains subject to country-specific regulations.
  • Range and altitude: UAV base stations may need altitude variation to maximize ground coverage and satisfy different quality-of-service requirements.Maximum flying altitude is therefore a critical parameter for UAV-aided cellular communications.
  • Platform classes: Low-altitude platforms are cost-effective, deploy quickly, and provide short-range line-of-sight links that can enhance communication performance.They are usually employed to assist cellular communications.
  • Platform classes: High-altitude platforms provide wider coverage and longer airborne duration, but their deployment is more complex and can cause total network outage through extreme inter-cell interference.Consequently, they are rarely considered in UAV-aided cellular-network literature.
  • Speed and flight time: Small drones typically fly below 15 m/s, large drones can reach 100 m/s, and small commercial drones usually endure 20–30 minutes.The survey identifies limited endurance of off-the-shelf UAVs as a major restriction on full-scale cellular deployment.
  • Power supply: Drone power systems must support both flight and onboard cellular equipment, including antenna arrays, amplifiers, and circuits.A typical aerial BS requires 5 W maximum transmit power from its onboard energy source.

III. STANDARDIZATION: ENABLING UAV CELLULAR COMMUNICATIONS

3GPP investigated how existing cellular networks can support aerial UAV users, focusing on traffic requirements, channel modeling, performance, and interference mitigation. The study found that UAVs’ elevated line of sight creates distinctive interference and connectivity challenges.

  • 3GPP’s study item addressed UAV traffic requirements, air-to-ground channel modeling, LTE reuse, and enhancements needed to serve aerial devices.
  • UAV traffic between ground level and 300 meters comprises synchronization and radio control, command and control, and application data.
  • Higher UAV altitudes increase line-of-sight probability; in urban-macro scenarios, UAVs above 100 meters are considered line-of-sight with all deployed cellular BSs.
  • The channel model captures decreasing path-loss exponents and shadowing variability with height for line-of-sight UAV–BS links, including a free-space-like exponent of α = 2.2.
  • UAVs experience more downlink and uplink interference than ground users because they can maintain line of sight with many base stations.Downlink transmissions suffer poorer SINRs, while UAV uplinks can strongly interfere with multiple ground BSs.
  • 3GPP examined complementary mitigation techniques, including beamforming, directional antennas, CoMP, and differentiated uplink power control.The study’s channel-modeling work included experimental and simulation studies alongside performance evaluation.

1) Association and handover:

Aerial UAV association and handover differ from ground-user behavior because antenna sidelobes and broad interference exposure shape connectivity. Proposed enhancements use flight information, reporting changes, beamforming, cooperation, and power control to improve coexistence.

  • Association and handover: UAV location and flight-plan knowledge can help anticipate candidate BSs and facilitate handovers.
  • Association and handover: UAV-specific handover triggers and optimized reporting-load control were proposed, with their evaluation assigned to a subsequent work-item study.
  • Interference mitigation: FD-MIMO reduces interference through beamformed transmissions and improves resource use through spatial multiplexing.
  • Interference mitigation: Directional UAV antennas and beamforming can reduce downlink interferers and increase useful serving-BS power, but increase transceiver hardware complexity.
  • Interference mitigation: CoMP can turn line-of-sight interferers into useful contributors, although practical gains are limited by many interfering BSs and increased inter-BS signaling.
  • Interference mitigation: Differentiated fractional path-loss compensation and P0 offsets for UAVs and GUEs can mitigate UAV-generated uplink interference.

B. 3GPP Work Item: Main Outcomes and Way Forward

The 3GPP work item translated the identified UAV interference and mobility challenges into LTE enhancements, while related bodies addressed architecture, spectrum, privacy, and security. Aerial base-station research additionally faces placement, mobility, energy, and end-to-end connectivity constraints.

  • B. 3GPP Work Item: Main Outcomes and Way Forward: The 3GPP work item concluded with LTE enhancements for UAV uplink power control, airborne-status signaling, interference detection, subscription access, and mobility.
  • B. 3GPP Work Item: Main Outcomes and Way Forward: New RRC signaling allows UAVs to communicate flight plans to serving BSs, supporting handover assistance and motivating future 5G-based work.
  • Other standardization activities: ITU-T’s Y.UAV.arch work item targets a UAV architecture, application capabilities, service support, and security measures for IMT-2020 networks.
  • Other standardization activities: ETSI examined UAV use cases, spectrum-rule needs, and implications for future IP-suite architecture, while IEEE focused on consumer-drone taxonomy, privacy, and security requirements.
  • IV. AERIAL BASE STATIONS: CHALLENGES AND OPPORTUNITIES: Flying base stations can be repositioned in three-dimensional space to potentially improve coverage, load balancing, spectral efficiency, and user experience.
  • IV. AERIAL BASE STATIONS: CHALLENGES AND OPPORTUNITIES: Aerial base-station deployment must address placement and mobility optimization, power consumption and recharging, and end-to-end access-to-backbone links.

A. Placement Optimization for Aerial BSs

Aerial-BS placement is more complex than terrestrial placement because altitude changes coverage and both uplink and downlink channels. Existing studies optimize location, height, performance, power, or the number of aerial BSs, while mobility remains largely unmodeled.

  • Placement optimization: Aerial-BS placement is harder than terrestrial placement because coverage and uplink/downlink channels vary with altitude.Studies therefore formulate either 2D placement problems or 3D problems that also optimize height.
  • Placement optimization: Brute-force placement of helper aerial BSs improves 5th percentile spectral efficiency in disaster-affected areas, with optimal locations along cell edges.Lowering aerial-BS height markedly improves 5th percentile throughput, while height has little effect on overall throughput.
  • Placement optimization: 3D placement formulations can be mixed-integer nonlinear problems, addressed using interior-point optimization and bisection search or transmit-power minimization.Other studies use genetic algorithms, K-means clustering, or evolutionary computing for placement objectives.
  • Placement optimization: Most surveyed placement studies optimize only the access link and assume infinite-bandwidth backhaul, although later work jointly optimizes access and backhaul spectral efficiency.This distinction matters because an aerial BS connects users through both access and backhaul links.
  • Mobility optimization: Mobility is an intrinsic aerial-BS capability that can dynamically improve placement as ground users move, but surveyed placement formulations generally omit it.Mobility studies distinguish BSs that stop service while relocating from BSs that continue serving users in motion; cruising systems also require heading control and collision avoidance.

C. Power-Efficient Aerial BSs

Power-efficient aerial-BS operation must address short battery lifetimes caused by both communication and mechanical energy consumption. The literature reduces transmission energy, schedules transmissions, models flight energy, and trades altitude or motion against coverage and endurance.

  • Energy challenge: Aerial BSs have short lifetimes because batteries power both communication electronics and mechanical mobility.Power-efficient operation is therefore required to extend battery lifetime.
  • Communication energy: Transmission-power minimization, cooperative deployment, relay selection, and transmission scheduling reduce communication energy under data-rate or trajectory constraints.Examples include TDMA sensor-data collection and game-theoretic beacon scheduling for competing UAVs.
  • Communication energy: Communication-energy reductions may have limited overall effect because communication energy is generally negligible relative to total UAV energy consumption.This motivates focusing on mechanical energy consumption.
  • Mechanical energy: The surveyed mechanical-energy models include hover, motor, lifting, mass, gravity, altitude, speed, and operating-time terms.One model uses β, α, Pmax, h, v, and t to represent power and motion-related consumption.
  • Mechanical energy: Higher altitude increases observation radius but also energy consumption, creating a tradeoff between coverage radius and energy efficiency.Researchers also optimize flight radius and speed and constrain path planning by battery availability or flight time.

D. Recharging of Aerial BSs

Aerial-BS deployment requires practical solutions for battery depletion and reliable fronthaul to the terrestrial network. The survey also links these engineering issues to relay design, end-to-end validation, interference scope, and potential site-cost savings.

  • Recharging and replacement: Battery depletion creates a short aerial-BS lifetime, motivating charging locations and replacement by fully charged UAVs.These approaches require urban charging or replacement infrastructure and regular battery monitoring, making them more costly and complex than energy reduction.
  • Fronthauling and access links: A reliable fronthaul link is crucial because it connects the UAV-mounted BS to the terrestrial network and supports sufficient access-link bandwidth and reliability.Low-complexity aerial BSs are expected to operate as relay nodes with reduced protocol stacks, potentially using amplify-and-forward.
  • Fronthauling and access links: End-to-end validation is fundamental for assessing aerial-BS adoption, including relay schemes and massive-MIMO wireless fronthaul.Studies evaluate decode-and-forward and amplify-and-forward schemes and multi-tier 5G integrated access and backhaul scenarios.
  • Fronthauling and access links: Some aerial-BS performance studies omit interference from and toward neighboring terrestrial macro BSs, despite high aerial-BS line-of-sight probability potentially affecting large areas.This limits how directly their conclusions transfer to broader cellular deployments.
  • Cost model: Drone-mounted BSs could reduce operator site costs because site acquisition and construction account for 52% of CAPEX and land rent reaches 42% of OPEX in developed countries.Flying BSs may remove or reduce site-related costs, although drone and network-testing costs may increase.
  • Cost model: Replacing terrestrial coverage extensions with flying BSs or relays could avoid costly cell reorganization and support time-varying deployment and opportunistic communication services.The survey describes potential models involving public transportation and drone fleets that deliver goods and data together.

V. PROTOTYPING AND FIELD TESTS

The survey reviews prototypes that use high- and low-altitude platforms to extend coverage or flexibly deploy cellular base stations. Examples include Aquila, Loon, Nokia F-Cell, Perfume, and Huawei Digital Sky.

  • High-altitude platforms: Aquila targets remote-area Internet coverage from 18–20 km altitude across an approximately 100 km communication region.The solar-powered aircraft follows a defined trajectory and uses free-space optical links to ground access points serving users through Wi-Fi or LTE.
  • High-altitude platforms: Google Loon uses stratospheric balloons to relay links from ground stations to LTE phones outside traditional cellular coverage.Unlike Aquila, Loon can relay communication directly to end users and controls coverage through balloon positioning rather than propellers.
  • Low-altitude platforms: Nokia F-Cell addresses small-cell deployment cost with solar-powered, self-configured drone cells connected by a 64-antenna massive-MIMO wireless backhaul.Its design removes wired power and backhaul requirements while enabling flexible relocation.
  • Low-altitude platforms: F-Cell combines a digital massive-MIMO hub with analog repeaters to create a multi-stage wireless fronthaul with sparse remote radio heads.The architecture is transparent to baseband units and user equipment and supports non-LOS FDD networking.
  • Low-altitude platforms: Perfume develops autonomous aerial cellular relay robots whose machine-learning algorithms update the optimal three-dimensional relay position.The relays are intended to enhance connectivity and throughput for commercial terminals.
  • Field trials: Huawei Digital Sky established an end-to-end Shanghai ecosystem and authorized flying zones for trials involving connected-drone use cases.The zones have a 6 km diameter and a maximum height of 200 m.

F. Other relevant testbeds

Additional testbeds examine UAV networking, navigation, communication reliability, battery consumption, and weather effects, while the survey also frames UAV regulation around operational and societal concerns.

  • Other relevant testbeds: Small testbeds investigate communication-link reliability, UAV battery consumption, and weather-condition impacts.The survey summarizes these practical experiments in Table VI.
  • Other relevant testbeds: An autonomous-helicopter experiment compares a planned trajectory with the actual flight path and reports a maximum error of 3 meters in a one-square-kilometre field.The navigation system was tested for a single UAV.
  • Regulatory context: UAV regulation is motivated by privacy, data-protection, and public-safety concerns arising from increasingly capable drone operations.The survey notes that existing privacy legislation may be outdated relative to emerging technologies.

B. Criteria

UAV regulations are structured around six criteria and are evolving alongside spectrum, safety, privacy, and economic considerations. Cross-country frameworks differ in their operational and data-protection provisions.

  • Criteria: Current UAV regulations are framed around applicability, technical requirements, operational limitations, administrative procedures, human-resource requirements, and ethical constraints.
  • Criteria: Operational limitations include maximum altitude, airport and individual separation distances, and prohibited areas.
  • Criteria: Administrative procedures can require registration, operational certificates, and insurance, while some operations require qualified pilots.
  • Criteria: Ethical constraints address data and privacy protection during drone operations.
  • Communications regulation: European and U.S. communications regulators are addressing UAV spectrum needs, including European frequency harmonization and a proposed 5030–5091 MHz band for U.S. C&C and non-payload links.
  • Current frameworks: Across eight countries, every framework includes operational limitations, most require privacy protection, and only the UK is identified as protecting collected UAV data.
  • Security considerations: UAV-assisted cellular communications raise additional security stakes because compromising a flying cellular base station can cause loss of service and potentially a crash.

A. Cyber Security

The survey analyzes cyber and physical threats across UAV communication links and identifies countermeasures for attacks affecting confidentiality, integrity, availability, control, and onboard data.

  • Threat scope: The security analysis covers UAV use as cellular UEs or flying BSs/relays, including cellular, Wi-Fi, satellite, GCS, and GPS-related links.
  • Threat scope: Threat identification assesses radio-link risks and pairs identified threats with countermeasures and likelihood-impact evaluations.The analysis is summarized in the risk and threat tables.
  • Radio-link attacks: Jamming disrupts reception through same-band interference, while increasing SNR is a defense limited by transmitter power and receiver-noise reduction.
  • Radio-link attacks: Eavesdropping threatens confidentiality on wireless cellular and Wi-Fi channels, motivating encryption and physical-layer security.
  • Radio-link attacks: Hijacking can disconnect a drone from its GCS through deauthentication frames and enable remote control over 802.11 protocols.Detection algorithms and frame encryption are described as countermeasures.
  • Radio-link attacks: Spoofing can use false GPS signals to take over a drone, with jamming-to-noise sensing and multi-antenna defenses proposed against GPS spoofing.
  • Physical attacks: Physical attacks range from extracting telemetry through interfaces to advanced side-channel, fault-injection, and software attacks.Proposed defenses include self-destruction, encryption, cryptography, and secure key management, with self-destruction carrying public-safety and data-loss side effects.

VIII. LESSONS LEARNED

The survey finds that UAV cellular communications span aircraft capabilities, aerial access, flying base stations, regulation, security, and simulation realism, with substantial technical and administrative gaps remaining.

  • Sec. I: UAV types and characteristics: UAV capabilities span low-cost recreational drones to commercial and military platforms, with a persistent trade-off among flying time, payload, and cost.The survey identifies joint optimization of these metrics as a continuing research need.
  • Sec. II: Standardization advancements: Existing cellular networks face interference from aerial users and may struggle to provide high-quality service to ground and aerial users because they were designed mainly for 2D coverage.Future systems may need three-dimensional spatial flexibility, with massive MIMO identified as a candidate technology.
  • Sec. III: UAV-assisted cellular communications: Aerial base stations could provide coverage where and when needed, but placement, mobility, power efficiency, recharging, and security remain major concerns before cost-effective deployment.These challenges limit replacing conventional ground base stations, even partially.
  • Sec. IV: Prototyping and field tests: Prototypes including solar-powered aircraft, autonomous balloons, and self-powered aerial small cells have been tested, but their effectiveness in mixed aerial-ground cellular architectures remains uncertain.Realistic verification focuses on drawbacks and constraints associated with aerial base stations.
  • Sec. V: Regulation: UAV communications also face privacy, public-safety, administrative, licensing, and regulatory challenges, with no unified worldwide basic rule set.Existing national regulations commonly address flying height, weight, and safety distance from people.
  • Sec. VI: Security: Security must protect people from falling drones and protect captured or transferred data from hijacking, while stronger hardware and software security can increase costs.The survey treats security as both a physical-safety and data-protection requirement.
  • Future research directions: Simulation-based evaluations can be unrealistic because they often allow unconstrained UAV movement, motivating simulators that model obstacles, vehicle types, and telemetry.Proposed extensions would bring UAV mobility simulation closer to the support available for vehicular communications.

B. Advanced UAV Mobility Control Based on Image Processing and Deep Learning

The surveyed directions combine image processing, deep learning, radio measurements, and mobility-aware system design to improve UAV positioning, aerial-user identification, and cellular support, while exposing reliability and resource constraints.

  • Advanced UAV Mobility Control Based on Image Processing and Deep Learning: Cameras and deep learning can help UAVs identify hovering locations or flight directions that improve signal propagation around real-world obstacles.The target link differs depending on whether the UAV acts as an aerial UE or aerial base station.
  • Advanced UAV Mobility Control Based on Image Processing and Deep Learning: Mobility control should also account for UAV energy status and battery charging operations.Energy-aware control is presented as part of advanced UAV mobility management.
  • Advanced UAV Mobility Control Based on Image Processing and Deep Learning: Tracking antennas can use gyroscope, accelerometer, and GPS information to follow ground stations and tilt airborne antennas for higher data-rate transmission.Limited onboard space remains an additional antenna-installation concern.
  • D. Aerial UE Identification: Aerial UEs can be detected with UE- and network-based indicators, and machine-learning methods using LTE radio measurements achieve up to 99% accuracy.The discussed measurements include RSRP and RSSI, while future algorithms may use additional characteristics.
  • Advanced UAV Mobility Control Based on Image Processing and Deep Learning: UAV base stations introduce physical reliability risks such as battery depletion, wind-driven loss of control, and collisions, which can affect wireless service and user experience.These risks must be addressed before integrating UAV base stations into cellular networks.
  • Advanced UAV Mobility Control Based on Image Processing and Deep Learning: Supporting edge computing on UAV base stations increases payload and energy demands because computing platforms such as GPUs must be carried onboard.Computing-session continuity is another issue for UAV-supported edge services.
  • Advanced UAV Mobility Control Based on Image Processing and Deep Learning: Aerial caching uses UAV mobility and local memory to deliver frequently requested multimedia content more efficiently.The motivation is the high volume and repeated demand of video and image data.
  • X. CONCLUSION: The survey connects standardization, industry prototypes, academic research, regulation, and cyber-physical security as complementary parts of UAV cellular integration.Its conclusion reports interference challenges, demonstrated UAV-mounted base stations, expanding research activity, regulatory development, and emerging security and cost concerns.
Loading 1809.01752v2…