PaperlayerView exact revision on arXiv

Source-linked AI summary

Cyclical Multiple Access in UAV-Aided Communications: A Throughput-Delay Tradeoff

Jiangbin Lyu, Yong Zeng, Rui Zhang

arXiv:1608.03180v2cs.IT

TL;DR

Distributed ground terminals served by a cyclically moving UAV experience periodic channel-strength variations that conventional static operation does not exploit. The paper proposes cyclical multiple access with optimized position-based time allocation to maximize minimum throughput. Simulations show substantial throughput gains over a static UAV base station, with increased access delay and suitability for delay-tolerant applications.

  • Problem

    The paper studies how to exploit cyclically varying UAV–GT channels when a UAV serves distributed ground terminals as a mobile base station.

  • Method

    Cyclical multiple access schedules GTs in cyclical time division according to UAV position and optimizes allocations to maximize minimum throughput.

  • Results

    87% throughput gain over the static UAV base station is achieved with optimized normalized trajectory length D̄*=1.10, at the cost of increased access delay.

  • Takeaways & Limitations

    The mobile-UAV CMA design is most suitable for delay-tolerant applications such as periodic sensing and large data transfer.

Abstract

from arXiv · show

This letter studies a wireless system consisting of distributed ground terminals (GTs) communicating with an unmanned aerial vehicle (UAV) that serves as a mobile base station (BS). The UAV flies cyclically above the GTs at a fixed altitude, which results in a cyclical pattern of the strength of the UAV-GT channels. To exploit such periodic channel variations, we propose a new cyclical multiple access (CMA) scheme to schedule the communications between the UAV and GTs in a cyclical time-division manner based on the flying UAV's position. The time allocations to different GTs are optimized to maximize their minimum throughput. It is revealed that there is a fundamental tradeoff between throughput and access delay in the proposed CMA. Simulation results show significant throughput gains over the case of a static UAV BS in delay-tolerant applications.

I. INTRODUCTION

The paper studies a mobile UAV base station serving distributed ground terminals through periodically varying channels. It proposes cyclical multiple access to exploit stronger channels near each terminal, improving max-min throughput while increasing access delay.

  • The UAV operates as a mobile base station providing wireless connectivity to distributed ground terminals.
  • Cyclical UAV motion creates periodic variations in the strengths of the UAV–GT channels.
  • Cyclical multiple access schedules terminals in position-based cyclical time division to exploit stronger channels when the UAV flies closer to them.
  • The proposed algorithm allocates transmission time to maximize the minimum terminal throughput.
  • The mobile-UAV CMA design significantly improves max-min throughput over a static UAV base station, at the expense of increased access delay.

II. SYSTEM MODEL

The system models equally spaced ground terminals along a line and a UAV flying cyclically above them at fixed altitude. Downlink links use a line-of-sight free-space path-loss model with known periodic channel variation.

  • K ground terminals are equally spaced along a straight line of length Δ, with positions defined by x_k = −Δ/2 + (k−1)Δ/(K−1).
  • The UAV follows a cyclic horizontal trajectory with period T and fixed altitude H.
  • The model considers downlink communication over a given frequency band, with analogous application to uplink communications.
  • UAV–GT channels are modeled as line-of-sight links, with Doppler effects from UAV mobility assumed perfectly compensated.
  • Each GT’s channel varies periodically with the UAV position, and the UAV knows this variation over period T.
  • Throughput is measured in bps/Hz, also called spectrum efficiency, under constant UAV transmission power.

III. CYCLICAL MULTIPLE ACCESS

Cyclical multiple access exploits periodic channel variations by assigning ground terminals cyclical time-division access to the UAV.

  • CMA schedules ground terminals to communicate with the UAV in a cyclical time-division manner.
  • The scheme is designed to exploit the periodic channel variations experienced by different ground terminals.
  • CMA provides the basis for the subsequent position-dependent transmission scheduling design.

A. Cyclical TDMA

Cyclical TDMA assigns contiguous UAV trajectory segments to terminals according to their position-dependent rates. The resulting throughputs are computed from rate integrals over the allocated segments.

  • Each GT’s rate is symmetric and unimodal in UAV position, reaching its maximum when the UAV is directly above that GT.
  • Cyclical TDMA divides the one-way UAV trajectory into K contiguous segments and assigns each segment to one GT.
  • Segments near each GT’s location are allocated to that GT to exploit its highest achievable rate.
  • Fig. 2 compares the UAV–GT rate distribution with the corresponding transmission-time allocation for 10 GTs.
  • The average throughput of each GT is obtained by integrating its position-dependent rate over its allocated trajectory segment.

B. Max-Min Throughput

The section formulates max-min throughput optimization for cyclical TDMA and solves it by iteratively equalizing neighboring GT throughputs. The resulting mobile-UAV scheme improves throughput over a static UAV, but longer trajectories increase access delay.

  • Problem formulation: The non-convex problem maximizes the minimum GT throughput by optimizing trajectory delimiting variables b1, b2, · · ·, bK−1 for a fixed trajectory length D.Fairness is enforced through constraints θk ≥ τ and ordered segment boundaries.
  • Structural properties: At the max-min solution, all GTs have equal throughput, and each delimiting variable affects only its two neighboring GT throughputs.Increasing a boundary raises one neighboring throughput while lowering the other, enabling local equalization.
  • Algorithm: Algorithm 1 repeatedly selects the largest neighboring throughput gap and updates its boundary by solving the equal-throughput equation until every gap is below ǫ.The algorithm initializes uniformly spaced boundaries and uses ǫ = 10−5 as the stopping threshold.
  • Optimality: Algorithm 1 converges because the largest throughput gap is nonnegative and decreases each iteration, then achieves the max-min throughput by contradiction.At convergence, all GT throughputs are equal; the proof rules out any higher feasible minimum throughput.
  • Numerical comparison: 33.7% higher max-min throughput is achieved by the mobile UAV with D = ∆/2 than by the static UAV, whose example throughput is 0.3488 bps/Hz.The mobile case achieves 0.4663 bps/Hz, while increasing D also increases the period and access delay for fixed UAV speed.

C. Access Delay

Cyclical TDMA creates access delay because each GT communicates in two separated windows during each UAV flight period. Delay varies by GT location and increases with trajectory length, motivating an RMS-delay constraint.

  • Access-delay definition: Each GT communicates during two non-consecutive windows corresponding to its allocated UAV-position segment, creating two mute intervals per flight period.The mute intervals occur on the left and right sides of the UAV trajectory.
  • Access-delay definition: Access delay is the longest contiguous mute time experienced by a GT during one UAV flying period.
  • Access-delay patterns: Middle GTs generally have smaller delays, while edge GTs have larger delays because their left- and right-side mute intervals are respectively more balanced and unbalanced.These location-dependent patterns should be considered when designing upper-layer protocols or applications with different delay requirements.
  • RMS access delay: The overall delay metric is RMS access delay, which accounts for both the average delay and its variation across GTs.
  • RMS access delay: With fixed UAV speed, RMS access delay generally increases with trajectory length because the round-trip period is T = 2D/V.The system imposes the constraint φrms ≤ Φ.
  • Access-delay definition: Access delay differs from conventional communication delay and can be arbitrarily small for a static UAV BS by using arbitrarily short TDMA minislots.

IV. NUMERICAL RESULTS

The numerical study optimizes trajectory length under RMS-delay tolerances and compares optimal versus equal time allocation. Results show that mobility can substantially improve max-min throughput, especially for larger GT ranges, but with increased access delay.

  • IV. NUMERICAL RESULTS: The study compares max-min throughput under different RMS access-delay tolerances using normalized trajectory length and fixed simulation parameters.For each tolerance, trajectory length is searched subject to the RMS-delay constraint.
  • IV. NUMERICAL RESULTS: Optimal trajectory length is selected by one-dimensional search because RMS access delay generally increases monotonically with normalized trajectory length.Equal time allocation, δk = 1/K, is used as a comparison scheme.
  • IV. NUMERICAL RESULTS: For ∆ = 1000m, optimal and equal time allocation achieve normalized trajectory lengths D̄* = 1.10 and 1.11, respectively.
  • IV. NUMERICAL RESULTS: The corresponding RMS delays are φrms = 52.37s and 52.85s for optimal and equal allocation, respectively.
  • IV. NUMERICAL RESULTS: When Φ > 60s, equal allocation with D̄* = 1.11 achieves near-optimal performance, whereas for Φ < 30s, optimal allocation significantly outperforms equal allocation.
  • IV. NUMERICAL RESULTS: Optimized mobile-UAV operation yields throughput gains of 33.7% at D̄ = 0.5 and 87% at D̄* = 1.10 over the static-UAV case.The gain reaches up to 236% when the GT location range is ∆ = 2000m.

V. CONCLUSIONS

The letter proposes cyclical multiple access for UAV-aided communications and characterizes max-min throughput through position-based transmission-time allocation. Simulations show significant gains over a static UAV BS in delay-tolerant scenarios.

  • V. CONCLUSIONS: The proposed CMA scheme allocates GT transmission time according to UAV position to characterize and optimize max-min throughput.
  • V. CONCLUSIONS: Simulation results show significant throughput gains over a static UAV BS in delay-tolerant scenarios.
  • V. CONCLUSIONS: Future extensions include arbitrary GT locations in 2D or 3D, variable UAV speed or altitude, and multiple-UAV deployment or cooperation.
Loading 1608.03180v2…