[Paper Review] Hierarchical Aerial Computing for Internet of Things via Cooperation of HAPs and UAVs

Ziye Jia, Qihui Wu+3 more|arXiv (Cornell University)|Feb 12, 2022

UAV Applications and Optimization4 citations

TL;DR

This paper proposes a hierarchical aerial computing framework using high-altitude platforms (HAPs) and unmanned aerial vehicles (UAVs) to maximize computation offloading for IoT devices in remote or disaster-affected areas. It employs a matching game theory-based algorithm for IoT-to-UAV offloading and a heuristic algorithm for UAV-to-HAP offloading, achieving near-optimal performance with reduced complexity, outperforming standalone UAV or HAP solutions in total computed data and user support.

ABSTRACT

With the explosive increment of computation requirements, the multi-access edge computing (MEC) paradigm appears as an effective mechanism. Besides, as for the Internet of Things (IoT) in disasters or remote areas requiring MEC services, unmanned aerial vehicles (UAVs) and high altitude platforms (HAPs) are available to provide aerial computing services for these IoT devices. In this paper, we develop the hierarchical aerial computing framework composed of HAPs and UAVs, to provide MEC services for various IoT applications. In particular, the problem is formulated to maximize the total IoT data computed by the aerial MEC platforms, restricted by the delay requirement of IoT and multiple resource constraints of UAVs and HAPs, which is an integer programming problem and intractable to solve. Due to the prohibitive complexity of exhaustive search, we handle the problem by presenting the matching game theory based algorithm to deal with the offloading decisions from IoT devices to UAVs, as well as a heuristic algorithm for the offloading decisions between UAVs and HAPs. The external effect affected by interplay of different IoT devices in the matching is tackled by the externality elimination mechanism. Besides, an adjustment algorithm is also proposed to make the best of aerial resources. The complexity of proposed algorithms is analyzed and extensive simulation results verify the efficiency of the proposed algorithms, and the system performances are also analyzed by the numerical results.

Motivation & Objective

• Address the challenge of providing low-latency, high-capacity MEC services for IoT devices in remote or disaster-affected areas lacking terrestrial network coverage.
• Formulate an integer programming problem to maximize total successfully computed IoT data under joint constraints of delay, computation, energy, and transmission resources.
• Overcome the prohibitive complexity of exhaustive search in large-scale aerial computing networks through computationally tractable algorithms.
• Enable efficient cooperation between HAPs and UAVs by modeling offloading decisions across two hierarchical layers: IoT-to-UAV and UAV-to-HAP.
• Improve resource utilization through an adjustment algorithm that optimizes aerial platform deployment and task distribution.

Proposed method

• Develop a hierarchical aerial computing framework where UAVs serve as intermediate relays and HAPs act as high-capacity edge servers for heavy computation tasks.
• Apply a matching game theory-based algorithm to model and solve IoT device offloading decisions to UAVs, incorporating preference lists that encode delay, energy, and computation constraints.
• Introduce an externality elimination mechanism to handle interdependencies between IoT devices' offloading decisions, ensuring stable and efficient matching outcomes.
• Design a heuristic algorithm for UAV-to-HAP offloading decisions, prioritizing energy efficiency and timely computation under resource limitations.
• Propose an adjustment algorithm to dynamically reconfigure UAV and HAP roles and positions to maximize resource utilization and system throughput.
• Analyze the computational complexity of all proposed algorithms and validate performance via extensive simulations against exhaustive search and baseline schemes.

Experimental results

Research questions

• RQ1 How can HAPs and UAVs be jointly coordinated to maximize the total amount of IoT data successfully computed in aerial MEC networks?
• RQ2 What is the optimal offloading strategy for IoT devices between UAVs and HAPs under strict delay and resource constraints?
• RQ3 How can interdependencies (externalities) among IoT devices’ offloading decisions be mitigated to ensure stable and efficient matching outcomes?
• RQ4 What impact do variations in HAP and UAV computation capabilities have on system performance and energy consumption?
• RQ5 Can a low-complexity algorithm achieve near-optimal performance compared to exhaustive search in large-scale aerial computing deployments?

Key findings

• The proposed hierarchical HAP-UAV computing framework outperforms both standalone UAV-only and HAP-only MEC modes in terms of total computed data and number of served IoT users.
• The matching game-based algorithm for IoT-to-UAV offloading achieves near-optimal performance with significantly reduced computational complexity compared to exhaustive search.
• HAP computation capability has a stronger impact on system performance than UAV computation capability, as HAPs serve as central processing hubs for multiple UAVs and IoT devices.
• Increasing HAP or UAV computation capacity leads to higher total energy consumption, with HAP capability having a more pronounced effect due to its broader coverage and higher load.
• The adjustment algorithm effectively enhances resource utilization, enabling better adaptation to dynamic network conditions and improved system throughput.
• Simulation results confirm that the proposed algorithms achieve performance close to the optimal solution while maintaining low computational overhead, validating their practicality for real-world deployment.