Age-of-Task-Aware AAV-Based Mobile Edge Computing Techniques in Emergency Rescue Applications

In the case of extreme natural disasters like typhoons, earthquakes, and forest fires, the terrestrial communication infrastructure often suffers from severe damage, which seriously undermines the effectiveness of emergency response efforts, leading to critical challenges, such as the timely assessment of disasters, the quick emergency response strategy development, and the rapid implementation of reconnaissance and search-and-rescue operations. To address these challenge issues, autonomous aerial vehicles (AAVs)-based mobile edge computing (MEC) techniques had attracted research attention to effectively support emergency communication, disaster assessment, and rescue strategy decisions. In order to characterize the time-critical requirements of many emergency rescue applications, the concept of “Age of Task” (AoT) was introduced in this article as a metric for assessing the timeliness of task, and the minimization of the weighted AoT across all the terrestrial user equipments (UEs) was formulated. By leveraging the Lyapunov optimization analysis framework, the problem of minimizing the time-averaged weighted AoT was transformed into a series of real-time subproblems that involve task offloading scheduling decision, computational resource allocation, UE transmit power control, and AAV flight trajectory planning, all of which enable an AoT-aware AAV-based MEC network for emergency rescue applications. To highlight the effectiveness, four benchmark schemes were included for comparison to show the advantages of the AoT-aware adaptive AAV-based MEC algorithm (AAAUMA) in terms of the realized task freshness performance, lower energy consumption by the AAV, and smaller data buffer backlog sizes at all ground source nodes.

I. Introduction

Large-scale sudden-onset natural disasters always result in severe and unpredictable loss of properties and lives. During a natural disaster, it is crucial to maintain real-time and fast communications to quickly assess the situation, and to provide timely communication and computing services to locate disaster victims for enhancing the effectiveness of rescue operations. Unfortunately, in most cases, natural disasters like earthquakes, tsunamis, floods, fires, and hurricanes, tend to destroy most of ground communication equipments. Even those that are not damaged tend to stop working due to power outages. Taking the Wenchuan earthquake as an example, the communication infrastructure in Sichuan province was devastated by the earthquake. Data released by China Telecom showed that 28714 mobile and PHS base stations were destroyed, along with 28765 km of fiber optic cables, and 142078 telecommunication poles were snapped [1]. During Hurricane Harvey in the United States, the FCC announced that only one of the 19 cell towers in Arkansas County, Texas, was operational, and 85% of the cell towers in nearby counties were offline [2]. In sudden-onset natural disasters, rescuers always encounter challenging conditions, including damaged ground infrastructure, unstable satellite communication/navigation signals, and limited communication and computation capabilities. These factors significantly impede the effective assessment of the disaster, hinder the timely development of emergency rescue strategies and the rapid forward reconnaissance, as well as in-depth rescue search and efforts [3]. The challenging environment following a disaster imposes significant limitations on the use of conventional portable communication and sensing equipment. Therefore, it is essential to establish an emergency communication network that can respond promptly and adapt flexibly in the aftermath of a disaster [4], [5].