Joint Optimization of Trajectory Control, Task Offloading, and Resource Allocation in Air-Ground Integrated Networks

In an air-ground integrated network (AGIN), low-altitude unmanned aerial vehicles (UAVs) and a high-altitude platform (HAP) operate synergistically to support computationally expensive and delay-critical applications of mobile ground devices (GDs). UAVs obtain tasks from GDs, execute the tasks, and...

Full description

Saved in:
Bibliographic Details
Published in:IEEE internet of things journal 2024-07, Vol.11 (13), p.24273-24288
Main Authors: Morshed Alam, Muhammad, Moh, Sangman
Format: Article
Language:English
Subjects:
Air-ground integrated networks (AGINs)
Autonomous aerial vehicles
Computation offloading
Delay
Delays
Energy consumption
High altitude
Internet of Things
joint optimization
Low altitude
Mixed integer
multiagent deep reinforcement learning (MA-DRL)
Multiagent systems
Resource allocation
Resource management
Swarming
Task analysis
task offloading
Trajectory
Trajectory control
Trajectory optimization
Unmanned aerial vehicles
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In an air-ground integrated network (AGIN), low-altitude unmanned aerial vehicles (UAVs) and a high-altitude platform (HAP) operate synergistically to support computationally expensive and delay-critical applications of mobile ground devices (GDs). UAVs obtain tasks from GDs, execute the tasks, and offload some of the tasks to the HAP. In AGINs, the trajectory control of a UAV swarm should provide optimal coverage to randomly distributed mobile GDs. The limited resources of UAVs, such as energy, computation, caching, and bandwidth, result in further challenges. Therefore, a joint optimization problem is formulated in this study to minimize the task execution delay and energy consumption of UAVs by optimizing the UAV's trajectory, GD association, task-offloading ratio, and resource allocation. The limited resources, maximum task execution delay, task queue size, and mobility of UAVs are regarded as key constraints. Solving the problem is intricate owing to the complex mixed-integer nonlinear constraints coupled with a large continuous and discrete decision space. To track the dynamics in AGINs and efficiently solve the problem above, we utilize a swarming behavior-integrated multiagent gated recurrent unit-based actor and multihead attention-based critic network (SMA-GAC) framework. Results of simulative evaluation show that the proposed SMA-GAC outperforms baseline methods.
ISSN:2327-4662
2327-4662
DOI:10.1109/JIOT.2024.3390168