PaperlayerView exact revision on arXiv

Source-linked AI summary

The Sky Is Not the Limit: LTE for Unmanned Aerial Vehicles

Xingqin Lin, Vijaya Yajnanarayana, Siva D. Muruganathan, Shiwei Gao, Henrik Asplund, Helka-Liina Maattanen, Mattias Bergström A, Sebastian Euler, Y. -P. Eric Wang

arXiv:1707.07534v2cs.NI

TL;DR

Beyond-visual-LOS UAV operations need reliable wide-area communications, while terrestrial LTE faces distinctive airborne coverage and interference conditions. The paper characterizes these conditions, models them through measurements and ray tracing, and evaluates LTE feasibility with simulations, finding that existing LTE networks should support initial small-UAV deployments while requiring performance enhancements to protect ground users.

  • Problem

    Small UAVs need beyond-visual-LOS connectivity, but Wi-Fi may not provide the wide-area service required and terrestrial LTE has distinct airborne coverage and interference challenges.

  • Method

    The paper identifies airborne requirements, characterizes LOS/NLOS propagation and pathloss, uses ray tracing and channel models, and evaluates LTE through system-level simulations.

  • Results

    Existing mobile LTE networks should be able to provide wide-area connectivity for initial small-UAV deployments, although airborne SINRs are lower and interference management is needed.

  • Takeaways & Limitations

    LTE can support early small-UAV connectivity, while performance-enhancing solutions are needed to improve airborne links and protect ground mobile devices.

Abstract

from arXiv · show

Many use cases of unmanned aerial vehicles (UAVs) require beyond visual line-of-sight (LOS) communications. Mobile networks offer wide area, high speed, and secure wireless connectivity, which can enhance control and safety of UAV operations and enable beyond visual LOS use cases. In this article, we share some of our experience in Long-Term Evolution (LTE) connectivity for low altitude small UAVs. We first identify the typical airborne connectivity requirements and characteristics, highlight the different propagation conditions for UAVs and mobiles on the ground with measurement and ray tracing results, and present simulation results to shed light on the feasibility of providing LTE connectivity for UAVs. We also present several ideas on potential enhancements for improving LTE connectivity performance and identify fruitful avenues for future research.

I. INTRODUCTION

Low-altitude small UAVs are expanding into commercial applications, but Wi-Fi may not meet beyond-visual-LOS needs. The article examines whether LTE can provide suitable wide-area connectivity and how to improve it.

  • I. INTRODUCTION: Low-altitude small UAVs are defined here as weighing no more than 55 pounds, flying up to 100 miles per hour, and operating up to 400 feet AGL.The definition also permits operation within 400 feet of a structure if higher than 400 feet AGL.
  • I. INTRODUCTION: Commercial UAV applications include delivery, communications and media, infrastructure inspection, surveillance, search and rescue, agriculture, and wildlife conservation.The paper cites drone-delivery trials by Amazon in the U.K. and U.S.
  • I. INTRODUCTION: Wi-Fi may be insufficient for beyond-visual-LOS communications, particularly when UAVs require wide-area connectivity.A U.S. FAA-NASA initiative focuses on communications and navigation solutions for safe control of beyond-visual-LOS unmanned aircraft systems.
  • I. INTRODUCTION: Mobile networks provide wide-area, high-speed, and secure connectivity that can enhance UAV control and safety and support beyond-visual-LOS use cases.The paper notes increasing field trials using terrestrial LTE networks for UAV connectivity.
  • I. INTRODUCTION: The article identifies UAV connectivity requirements, compares airborne and ground propagation, evaluates LTE feasibility through simulations, and discusses potential performance enhancements.It also points to future research directions for LTE connectivity to small UAVs.

II. CONNECTIVITY REQUIREMENTS AND CHARACTERISTICS

UAV links primarily support command and control or data communication, but terrestrial LTE designs create distinct coverage and interference challenges in the air.

  • II. CONNECTIVITY REQUIREMENTS AND CHARACTERISTICS: Command and control requires data rates up to 100 kbps, packet error rates below 0.1%, and latency within 50 ms.These requirements are specified by 3GPP for proper UAV operational control.
  • II. CONNECTIVITY REQUIREMENTS AND CHARACTERISTICS: Data communication for applications such as aerial cameras and surveillance may require data rates up to 50 Mbps.The traffic can include telemetry data, pictures, or videos sent from UAVs.
  • II. CONNECTIVITY REQUIREMENTS AND CHARACTERISTICS: Terrestrial LTE coverage is challenging because base-station antennas are down-tilted to reduce inter-cell interference, leaving UAVs potentially served by antenna sidelobes.The terrestrial network is optimized for ground broadband communication rather than airborne users.
  • II. CONNECTIVITY REQUIREMENTS AND CHARACTERISTICS: Airborne propagation is more favorable than terrestrial propagation, raising whether improved propagation can compensate for reduced antenna gain.The paper frames this as a central coverage question for aerial LTE links.
  • II. CONNECTIVITY REQUIREMENTS AND CHARACTERISTICS: Favorable airborne propagation can increase uplink interference to neighboring cells and downlink interference from them, potentially degrading ground-UE performance.Managing this interference is identified as a key objective of the 3GPP aerial-vehicle study.
  • II. CONNECTIVITY REQUIREMENTS AND CHARACTERISTICS: LTE coverage and interference for UAVs differ substantially from terrestrial connectivity, motivating enhancements that protect ground UEs.The paper seeks more effective and efficient airborne connectivity without negatively impacting terrestrial users.

III. AERIAL CHANNEL CHARACTERISTICS

Aerial LTE connectivity is shaped by wireless channels that differ from those experienced by ground users. The section focuses on LOS/NLOS conditions and large-scale pathloss.

  • III. AERIAL CHANNEL CHARACTERISTICS: Distinct UAV connectivity phenomena are rooted in wireless channels different from terrestrial links.A complete characterization of the ground-to-air channel is outside the article’s scope.
  • III. AERIAL CHANNEL CHARACTERISTICS: The article highlights LOS versus NLOS propagation and large-scale pathloss as two key aspects of aerial wireless channels.These aspects provide the basis for its channel analysis.

A. LOS versus NLOS propagation

The paper defines LOS and NLOS by whether the direct ray is obstructed and uses ray tracing to derive altitude-relevant LOS probability models for LTE simulations.

  • A. LOS versus NLOS propagation: LOS means the direct ray between two points is clear of obstacles, whereas NLOS means that ray is obstructed.A radio link is treated as either LOS or NLOS at a given instant.
  • A. LOS versus NLOS propagation: Figure 2 maps rural terrain and building heights on the left and plots the resulting LOS probabilities against 2D distance on the right.The map uses x-y coordinates and height-density values in meters.
  • A. LOS versus NLOS propagation: System-level simulations determine LOS or NLOS using a probability function dependent on transceiver distance and heights, then generate corresponding pathloss and small-scale fading.The study includes Fresnel-zone effects in pathloss modeling but not in LOS modeling.
  • A. LOS versus NLOS propagation: Existing 3GPP LOS probability models are not applicable to the small-UAV altitudes studied because their maximum UE height is 23 m.Accurate altitude-relevant LOS/NLOS modeling is therefore important for system-wide LTE evaluation.
  • A. LOS versus NLOS propagation: Ray tracing derives LOS probability models from a 5 m horizontal- and 0.15 m vertical-resolution 3D terrain map near Stockholm, Sweden.The procedure randomly places base stations and user equipment, then checks building and terrain blockage across UE heights.

B. Large-scale pathloss

Existing terrestrial pathloss models do not adequately cover small-UAV altitudes, while airborne links approach free-space behavior at sufficient height but incur diffraction loss over long ranges.

  • Statistical pathloss models adopted by 3GPP are important for estimating received signal power in system analysis and simulation.
  • 3GPP pathloss models are not applicable to many small-UAV altitudes; the rural macrocell model covers UE heights below 10 m and 2D distances up to 10 km.
  • At higher UAV altitudes, propagation may approach free space because LOS probability is close to one and the sky environment is clear.
  • At shorter distances, LOS pathloss is close to or lower bounded by free-space pathloss, whereas Earth curvature increases loss through diffraction at longer ranges.
  • Using the ground-UE 3GPP rural macrocell LOS model above 10 m overestimates pathloss, especially at higher altitudes and large 2D distances.

IV. TECHNICAL FEASIBILITY AND CHALLEGNES

System-level simulations indicate that LTE can provide useful airborne coupling gains, but stronger inter-cell interference lowers aerial SINR and can reduce coverage and terrestrial-network efficiency.

  • The rural LTE simulation uses 37 sites with 3 cells per site, 10 MHz bandwidth at 700 MHz, and BS antennas 35 m high with 6° downtilt.
  • The airborne simulation models BS sidelobes with a synthesized antenna array specified by (M, N, P) = (8, 1, 2) and 0.8λ vertical element spacing.
  • At 40 m and 120 m, fifth-percentile downlink coupling gains exceed the fifth-percentile downlink path gain at the 1.5 m ground-level benchmark.
  • At 20% resource utilization, median SINRs at 40 m and 120 m are 10.9 dB and 11.3 dB lower than at 1.5 m, respectively.
  • Aerial UEs may fall out of Release-12 coverage because downlink coverage normally requires a minimum SINR of -6 dB; Release 13 supports SINR as low as -10 dB.
  • With equal scheduling and power-control treatment, aerial UEs at 40 m or 120 m use more uplink resources and achieve lower throughput than the 1.5 m benchmark.

V. POTENTIAL ENHANCEMENTS FOR EFFICIENT UAV CONNECTIVITY

The paper frames LTE enhancement as an optimization problem: improve small-UAV connectivity while protecting the performance of ground mobile devices.

  • Potential enhancements target improved LTE performance for small UAVs without negatively impacting ground mobile devices.

A. Interference mitigation

The paper surveys interference-mitigation options for aerial LTE, emphasizing that airborne links expose more cells and create coordination and resource-allocation challenges.

  • Interference mitigation is a key objective of the 3GPP study on enhanced LTE support for aerial vehicles.
  • CoMP must scale because aerial UEs receive interference from more downlink cells and their uplink signals are visible to more cells under LOS propagation.
  • For CoMP, optimal cooperation grouping and the tradeoff among overhead, coordination complexity, and interference-mitigation gain remain open problems.
  • Multiple antennas can support interference cancellation, suppression, and beamforming for small UAVs, which can more feasibly carry extra antennas than smartphones.
  • Orthogonal resource partitioning is simpler, but static reservations may be inefficient when aerial resources are underutilized.
  • Dedicated sky-pointing cells may be particularly useful in UAV hotspots with frequent and dense takeoffs and landings.

B. Mobility enhancement

Aerial UEs may associate with farther base stations because downtilted antennas favor terrestrial coverage, producing fragmented cell associations. This can worsen SINR and complicate prediction of handover performance.

  • B. Mobility enhancement: Aerial UEs may select farther base stations than the closest site, fragmenting the association pattern into disconnected areas.Downtilted antenna mainlobes optimize terrestrial coverage, while aerial UEs may be served by sidelobes.
  • B. Mobility enhancement: Lower aerial-UE SINR may increase radio-link failures and failed handovers by disrupting commands, target-cell access, or measurement reports.Measurement reports may be lost or delayed, which can delay handover decisions.
  • B. Mobility enhancement: Handover performance is difficult to predict because it depends on the scenario, base-station antenna pattern, and aerial-UE trajectories and speeds.Cell selection, handover efficiency, and robustness are identified as study objectives.

C. Aerial UE identification

LTE optimization assumes that the network can recognize airborne-capable UEs and determine when they are flying. Detecting unauthorized or terrestrial UEs in flight is challenging, so interference-triggered identification is proposed.

  • C. Aerial UE identification: An airborne-capable UE is a UE certified or specially subscribed to connect to an LTE network while airborne.The network is assumed to identify both airborne capability and whether the UE is flying.
  • C. Aerial UE identification: Airborne capability can be signaled during connection setup through radio resource control signaling.The remaining challenge is determining whether the capable UE is currently flying.
  • C. Aerial UE identification: A terrestrial LTE UE attached to a UAV may create excessive interference and may be restricted by regulations.The network could apply performance-enhancing measures, limit service, or drop the connection.
  • C. Aerial UE identification: The network should begin identifying potential flying terrestrial UEs when it detects increased interference or receives another trigger such as emergency detection.Scanning every terrestrial UE that might be flying would be excessive.
  • C. Aerial UE identification: Identification can combine multi-base-station received-power patterns, position estimates, and Doppler-based speed estimates, but hovering or high-rise locations remain difficult to distinguish.The paper concludes that identifying flying terrestrial UEs deserves further study.

VI. CONCLUSIONS AND RESEARCH DIRECTIONS

The paper concludes that existing LTE networks should support initial small-UAV deployments, while identifying enhancements and research directions for scaling aerial connectivity. Proposed directions include UAV-to-UAV communication, higher-altitude aircraft, and 5G support.

  • VI. CONCLUSIONS AND RESEARCH DIRECTIONS: Existing terrestrial LTE networks should be able to provide wide-area connectivity for the initial deployment of small UAVs.The paper also identifies enhancements intended to improve aerial connectivity while protecting ground-UE performance as UAV connections increase.
  • VI. CONCLUSIONS AND RESEARCH DIRECTIONS: UAV communication remains an emerging and underexplored field with fruitful avenues for future research.The conclusion frames the research directions as an ongoing area of investigation.
  • VI. CONCLUSIONS AND RESEARCH DIRECTIONS: LTE device-to-device and vehicle-to-everything features could support UAV sensing, collision avoidance, coordination, and out-of-coverage UAV communications.The paper highlights reuse of onboard LTE modules for UAV-to-UAV detection and communication.
  • VI. CONCLUSIONS AND RESEARCH DIRECTIONS: Higher-altitude aircraft require higher link budgets, and beamforming is identified as an attractive technique for wide-area connectivity.This extends the paper’s low-altitude focus toward aircraft such as airliners.
  • VI. CONCLUSIONS AND RESEARCH DIRECTIONS: 5G is expected to provide higher capacity for terrestrial and aerial devices, supporting the goal of ubiquitous mobile broadband coverage on the ground and in the sky.The 5G study item is presented as a step toward this direction.
Loading 1707.07534v2…