[Paper Review] Energy Aware Trajectory Optimization for Aerial Base Stations

Jingcong Sun, Christos Masouros|arXiv (Cornell University)|Jan 23, 2019

UAV Applications and Optimization25 references4 citations

TL;DR

This paper proposes an iterative algorithm combining successive convex optimization and block coordinate descent to jointly optimize UAV trajectory and user scheduling for energy-constrained aerial base stations, achieving up to 50% higher coverage probability and 20% lower energy consumption than benchmarks, with robust techniques compensating for imperfect user location information.

ABSTRACT

By fully exploiting the mobility of unmanned aerial vehicles (UAVs), UAV-based aerial base stations (BSs) can move closer to ground users to achieve better communication conditions. In this paper, we consider a scenario where an aerial BS is dispatched for covering a maximum number of ground users before exhausting its on-board energy resources. The resulting trajectory optimization problem is a mixed integer non-linear problem (MINLP) which is non-convex and is challenging to solve. As such, we propose an iterative algorithm which decomposes the problem into two sub-problems by applying both successive convex optimization and block coordinate descent techniques to solve it. To be specific, the trajectory of the aerial BS and the user scheduling and association are alternately optimized within each iteration. In addition, to achieve better coverage performance and speed up convergence, the problem of designing the initial trajectory of the UAV is also considered. Finally, to address the unavailability of accurate user location information (ULI) in practice, two different robust techniques are proposed to compensate the performance loss in the existence of inaccurate ULI. Simulation results show both energy and coverage performance gains for the proposed schemes compared to the benchmark techniques, with an up to 50% increase in coverage probability and an up to 20% reduction in energy.

Motivation & Objective

• Maximize the number of ground users covered by a single aerial base station before on-board energy exhaustion.
• Address the challenge of joint trajectory and user scheduling optimization under non-convex, mixed-integer, non-linear constraints.
• Improve convergence speed and coverage performance through a carefully designed initial trajectory.
• Compensate for performance degradation caused by inaccurate user location information (ULI) in practical deployments.

Proposed method

• Formulate the problem as a mixed-integer non-linear program (MINLP) with energy and flight constraints.
• Apply successive convex optimization to approximate the non-convex trajectory subproblem into a convex one iteratively.
• Use block coordinate descent to alternate between optimizing UAV trajectory and user scheduling/association in each iteration.
• Design an initial trajectory (CIT and DIT) that ensures fair user access and accelerates convergence.
• Introduce two robust techniques—worst-case (WC) and maximum excess data method (MEDM)—to mitigate performance loss from imperfect ULI.
• Model propulsion power as a function of velocity and acceleration to accurately capture energy consumption during flight.

Experimental results

Research questions

• RQ1 How can UAV trajectory and user scheduling be jointly optimized to maximize covered users under strict energy and flight constraints?
• RQ2 What is the impact of initial trajectory design on convergence speed and coverage performance in energy-constrained UAV deployment?
• RQ3 How does imperfect user location information degrade system performance, and what mechanisms can effectively compensate for this loss?
• RQ4 What trade-offs exist between energy efficiency, coverage gain, and convergence speed in iterative trajectory optimization algorithms?
• RQ5 To what extent can robust design techniques restore performance when ULI is inaccurate?

Key findings

• The proposed IA-DIT algorithm achieves up to 50% higher coverage probability compared to benchmark schemes under the same energy budget.
• Energy consumption is reduced by up to 20% with the proposed scheme compared to conventional approaches.
• The designed initial trajectory (DIT) reduces convergence time, with IA-DIT converging in at most 9 iterations compared to 15 for IA-CIT.
• With imperfect ULI, coverage probability drops by approximately 25% without robust techniques, but MEDM and WC methods nearly fully compensate for this loss.
• When on-board energy is increased, the WC and MEDM techniques achieve higher coverage gains, with MEDM outperforming WC by about 7% across all energy levels.
• IA-DIT consumes more energy than IA-CIT in short time periods due to faster, more agile maneuvers, but IA-CIT’s energy use grows more rapidly with time, exceeding IA-DIT at T = 80s.