PaperlayerView exact revision on arXiv

Source-linked AI summary

Fundamental Tradeoffs in Communication and Trajectory Design for UAV-Enabled Wireless Network

Qingqing Wu, Liang Liu, Rui Zhang

arXiv:1805.07038v1cs.ITmath.DSmath.OC

TL;DR

The article addresses UAV-enabled wireless-network design under interference, delay requirements, and communication-resource constraints. It shows that UAV mobility and trajectory design can be jointly designed with communication resource allocation to balance throughput and delay, while IUIC is proposed for multi-UAV networks.

  • Problem

    Multi-UAV networks can experience severe interference from other UAVs due to line-of-sight channels, while delay requirements require joint trajectory and communication-resource design.

  • Method

    The paper jointly designs UAV trajectory and communication resource allocation, and proposes IUIC for multi-UAV-enabled networks.

  • Results

    The study shows that controllable UAV mobility with trajectory design can balance users' throughput and delay requirements through joint communication-resource allocation.

  • Takeaways & Limitations

    UAV trajectory design provides a way to trade communication performance requirements against movement-related energy costs, including energy savings at IoT devices at the cost of UAV movement.

Abstract

from arXiv · show

The use of unmanned aerial vehicles (UAVs) as aerial communication platforms is of high practical value for future wireless systems such as 5G, especially for swift and on-demand deployment in temporary events and emergency situations. Compared to traditional terrestrial base stations (BSs) in cellular network, UAV-mounted aerial BSs possess stronger line-of-sight (LoS) links with the ground users due to their high altitude as well as high and flexible mobility in three-dimensional (3D) space, which can be exploited to enhance the communication performance. On the other hand, unlike terrestrial BSs that have reliable power supply, aerial BSs in practice have limited on-board energy, but require significant propulsion energy to stay airborne and support high mobility. Motivated by the above new considerations, this article aims to revisit some fundamental tradeoffs in UAV-enabled communication and trajectory design. Specifically, it is shown that communication throughput, delay, and (propulsion) energy consumption can be traded off among each other by adopting different UAV trajectory designs, which sheds new light on their traditional tradeoffs in terrestrial communication. Promising directions for future research are also discussed.

I. INTRODUCTION

UAVs offer flexible aerial communication platforms with high mobility, favorable LoS links, and broad deployment possibilities, but their operation introduces throughput, delay, and energy tradeoffs. The article surveys these tradeoffs and examines trajectory and communication designs for managing them.

  • UAVs support aerial BS, relay, backhaul, data-collection, wireless-power-transfer, and aerial-user applications in 5G and beyond-5G networks.
  • High altitude increases LoS probability and reduces channel impairments, while controllable 3D mobility enables swift deployment and trajectory adaptation.
  • LoS channels can provide coverage over more ground users or BSs, requiring UAVs to maintain favorable links while controlling interference.
  • Moving closer to users or BSs can improve throughput but increases communication delay and propulsion-energy consumption.
  • The throughput-delay, throughput-energy, and delay-energy tradeoffs arise because trajectory choices jointly affect proximity, movement speed, and energy use.
  • The article overviews state-of-the-art results on UAVs as communication platforms and compares these tradeoffs with terrestrial communication.

II. FUNDAMENTAL TRADEOFFS IN UAV-ENABLED COMMUNICATION

The paper frames UAV-enabled communication around throughput, delay, and energy tradeoffs. It reviews terrestrial results and highlights differences caused by LoS channels, trajectory design, and high propulsion energy.

  • Throughput, delay, and energy have fundamental tradeoffs in wireless communication.
  • The paper reviews classic terrestrial tradeoff results before examining their UAV-to-ground counterparts.
  • UAV-specific differences arise from LoS channels, UAV trajectory design, and high propulsion-energy consumption.

A. Throughput-Delay Tradeoff

In terrestrial systems, throughput-delay tradeoffs arise from fading, scheduling, and user movement. In UAV-enabled communication, controllable mobility and trajectory scheduling determine delay over a longer flying-time scale.

  • Terrestrial communication: Terrestrial scheduling can improve throughput through multi-user diversity but increases user delay as the number of users grows.
  • Terrestrial communication: A general throughput-delay tradeoff exists for communication over fading channels.
  • Terrestrial communication: MANET throughput and average delay also trade off because users may wait until movement brings them sufficiently close.
  • UAV-enabled communication: In UAV-enabled communication, LoS channels reduce fading’s role, while UAV mobility and location determine channel conditions through distance.
  • UAV-enabled communication: Joint trajectory and communication scheduling can properly control UAV-enabled communication delay, unlike random MANET movement.
  • UAV-enabled communication: Terrestrial delay is measured on channel-coherence timescales such as milliseconds, whereas UAV delay follows flying time and may be measured in seconds.
  • UAV-enabled communication: Exploiting trajectory-based throughput-delay tradeoffs requires applications to tolerate more delay than terrestrial communication.

B. Throughput-Energy Tradeoff

UAV throughput-energy tradeoffs differ from terrestrial ones because propulsion energy dominates trajectory decisions. Reaching and serving users more closely can improve throughput but requires additional movement and maneuvering energy.

  • Traditional communication: Traditional throughput-energy tradeoffs follow transmit-power effects on achievable rate and energy efficiency.
  • Traditional communication: With circuit power included, energy efficiency first increases and then decreases with transmit rate.
  • UAV-enabled communication: Propulsion energy, generally in the kilowatt range, is several orders of magnitude higher than transmit and circuit energy.
  • UAV-enabled communication: Propulsion energy therefore dominates the trajectory decisions governing UAV throughput-energy tradeoffs.
  • UAV-enabled communication: Higher throughput may require longer and faster flights, close user proximity, hovering, altitude changes, and sharp turns to avoid blockages.
  • UAV-enabled communication: These trajectory requirements can substantially increase propulsion energy consumption for throughput enhancement.
  • Delay-energy implications: UAV throughput-delay and throughput-energy tradeoffs exhibit new aspects that produce substantially different delay-energy tradeoffs from terrestrial communication.
  • Delay-energy implications: Reducing delay requires maximum-speed travel between users and minimum-speed service near them, with both choices consuming more propulsion energy.

III. THROUGHPUT-DELAY TRADEOFF

The paper characterizes throughput-delay tradeoffs by jointly optimizing a UAV’s trajectory and communication for single- and multi-UAV systems. In the single-UAV case, longer flight periods improve common throughput but increase user waiting delay.

  • The study jointly optimizes UAV trajectory and communication to characterize throughput-delay tradeoffs, first for one UAV serving two GUs and then for multiple UAVs.
  • As the flight period T increases, the UAV flies closer to the two GUs and, for sufficiently large T, uses maximum speed between them to reserve more hovering time.At T = 100 s, the UAV flies between the GUs at maximum speed and hovers above them for communication.
  • The closer GU is scheduled at each instant, so each GU periodically waits T/2 s for the UAV to approach again.A larger T therefore produces a longer waiting time for each GU.
  • Compared with a static UAV, a mobile UAV significantly improves common throughput as T increases, at the cost of greater user delay.The static-UAV baseline fixes the UAV midway between the two GUs.

B. Multi-UAV Enabled Wireless Network

The multi-UAV design uses inter-UAV interference coordination through joint trajectory and power optimization. Cooperation improves the throughput-delay tradeoff by reducing communication distances and managing strong LoS interference.

  • Multiple UAVs can improve throughput by cooperatively serving GUs, but spectrum sharing creates severe LoS interference that requires inter-UAV interference coordination.The design jointly considers UAV trajectories, transmit power, and user associations.
  • The IUIC problem is non-convex with infinitely many trajectory variables, so time discretization, block coordinate descent, and successive convex optimization produce a suboptimal solution.A circular trajectory is used to initialize the algorithm.
  • Without power control, optimized trajectories shorten UAV–GU distances and enlarge UAV separation to alleviate co-channel interference, sometimes sacrificing direct-link gains.This tradeoff is illustrated for a two-UAV, six-GU setup with T = 120 s.
  • With power control, optimized trajectories avoid sacrificing direct-link gains because power control can mitigate strong interference even when UAVs are close.
  • The multi-UAV network significantly improves user throughput over the single-UAV network at the same user delay, verifying an improved throughput-delay tradeoff through IUIC.

C. Further Discussion and Future Work

The paper identifies future work involving multiple-access design, delay-aware resource allocation, channel-model-dependent trajectories, interference mitigation, and mobility adaptation. These directions reflect the sensitivity of UAV trajectory design to channel, interference, and user-movement conditions.

  • Whether hover-fly-hover with superposition coding achieves capacity for UAV broadcast channels with more than two users or other multiuser models remains open.
  • Future work should model delay requirements ranging from milliseconds to several seconds and jointly design UAV trajectories and communication resources to meet them.
  • Probabilistic LoS or Rician fading models may be more suitable than the LoS model in urban environments and can substantially affect optimal trajectory design.Under probabilistic LoS, lowering altitude generally decreases LoS probability, unlike under the always-beneficial LoS model.
  • Multi-UAV systems require further investigation of severe air-to-ground interference and Doppler effects induced by three-dimensional mobility.
  • The paper proposes IUIC to mitigate strong LoS interference through trajectory design, while CoMP may instead favor UAV fleets following a common trajectory for cooperative beamforming.A common trajectory is undesirable for IUIC because of inter-UAV interference.
  • Future designs should dynamically adjust UAV trajectories according to GU movement to improve throughput and delay performance.

IV. THROUGHPUT-ENERGY TRADEOFF

The paper characterizes throughput-energy tradeoffs using propulsion-energy models for fixed-wing and rotary-wing UAVs. Propulsion power is minimized at intermediate speeds, making both very low and very high speeds inefficient.

  • The section jointly designs UAV trajectories and communication while accounting for propulsion energy consumption.
  • Fixed-wing and rotary-wing UAVs have distinct advantages and limitations, motivating separate analytical propulsion-energy models for each type.
  • Propulsion energy depends on UAV velocity, including flying speed and direction, as well as acceleration.
  • For both UAV types, propulsion power first decreases and then increases as flying speed rises, so very low and very high speeds are not energy-efficient.
  • Very low-speed flight is extremely energy-consuming and may be impossible for fixed-wing UAVs, making hovering over small areas difficult.Rotary-wing UAVs do not face this hovering limitation to the same extent.
  • Rotary-wing UAVs consume excessive propulsion power at very high speeds, reducing their efficiency for missions over wide geographical areas.

B. Energy-constrained Trajectory Optimization

The section studies common-throughput maximization for a two-ground-user UAV system under propulsion-energy and mobility constraints. Increasing available propulsion energy enables trajectories and speed profiles that approach the unconstrained-throughput design, but practical speed and acceleration limits eventually become decisive.

  • Problem formulation: A battery-powered UAV serves two ground users over a finite flight period while jointly optimizing its trajectory and user scheduling under a total energy constraint.The fixed-wing model includes minimum-speed and maximum-acceleration constraints, with constant altitude, quasi-static users, and LoS channels.
  • Trajectory behavior: With Emax = 13000 J, the UAV follows a smooth trajectory close to both ground users with relatively large turning radii.The trajectory reflects limited propulsion energy and avoids sharp directional changes.
  • Trajectory behavior: With Emax = 23000 J, the trajectory approaches the design that ignores propulsion energy consumption, using sharper direction changes to reduce path loss.These sharper changes improve proximity to the users but require excessive propulsion energy.
  • Speed behavior: Limited propulsion energy keeps the UAV speed near 30 m/s, whereas sufficient energy supports an initial 50 m/s approach followed by 5 m/s flight around a ground user.The latter profile exploits controllable mobility to maximize throughput.
  • Throughput-energy tradeoff: Common throughput rises rapidly with available propulsion energy before approaching a constant strictly below the upper-bound throughput.At sufficiently large energy, such as Emax = 23000 J, throughput is limited by minimum UAV speed and maximum acceleration constraints rather than propulsion energy.

C. Further Discussion and Future Work

The discussion extends UAV communication tradeoffs to IoT energy consumption, energy harvesting, altitude selection, and multi-UAV coordination. These extensions expose additional design couplings, including balancing UAV propulsion costs against device energy savings and communication performance.

  • IoT energy tradeoffs: The throughput-energy tradeoff extends to a tradeoff between IoT devices’ communication energy consumption and UAV propulsion energy consumption.The extension is especially relevant to UAV-enabled data collection in IoT networks.
  • IoT energy tradeoffs: A mobile UAV collector can move close to IoT devices to reduce their transmission energy or completion time, but the saved device energy costs UAV movement.Reduced circuit-operation energy can also benefit energy-limited IoT devices.
  • Energy harvesting: For solar-powered UAVs, higher altitude increases free-space path loss but can harvest more solar energy to support faster flight and smaller turning radii.This creates a non-trivial altitude optimization tradeoff that could potentially alleviate the throughput-energy tradeoff.
  • Multi-UAV coordination: Multiple-UAV missions require energy cooperation in addition to inter-UAV signal-transmission coordination.Possible schemes include sequential energy replenishment, wireless power transfer, and collaborative trajectory design.
  • Multi-UAV coordination: Judicious trajectory design can balance propulsion energy consumption across different UAVs to support multi-UAV mission tasks.This adds energy balancing to coordination objectives for multi-UAV systems.

V. CONCLUSIONS

The conclusion presents throughput, delay, and energy as jointly manageable tradeoffs in UAV-enabled communication and trajectory design. It emphasizes joint trajectory and communication-resource optimization while noting the paper’s focus on UAVs as aerial base stations.

  • Contributions: The article revisits fundamental throughput, delay, and energy tradeoffs in emerging UAV-enabled wireless communication.These tradeoffs form the paper’s central scope.
  • Contributions: Controllable UAV mobility and trajectory design can be jointly optimized with communication resource allocation to balance users’ throughput and delay requirements and UAV energy consumption.The conclusion frames these objectives as jointly designable rather than independent.
  • Scope: The proposed tradeoffs also apply to UAV users, although the paper focuses on UAVs serving as aerial base stations.The conclusion explicitly identifies this as an extension beyond the paper’s main setting.
  • Future work: Other design considerations remain open, including improving throughput or reducing delay through additional trajectory and communication design choices.The conclusion points to further practical research rather than claiming the three tradeoffs exhaust the design space.
  • Future work: The article aims to provide insights for practical UAV-enabled wireless-network communication and trajectory design and motivate further research.The stated outlook includes continued work in this research field.
Loading 1805.07038v1…