QTCAJOSA: Low-Complexity Joint Offloading and Subchannel Allocation for NTN-Enabled IoMT

TL;DR

This paper proposes a low-complexity algorithm for joint offloading and subchannel allocation in NTN-enabled IoMT, optimizing task delay across UAVs, HAPS, and LEO satellites.

Contribution

It introduces a greedy heuristic algorithm for resource allocation in non-terrestrial IoMT networks, reducing complexity while improving delay performance.

Findings

Including non-terrestrial nodes reduces task delay.

The proposed algorithm outperforms baseline methods.

Simulations confirm the effectiveness of the joint offloading approach.

Abstract

In this work, we consider the resource allocation problem for task offloading from Internet of Medical Things (IoMT) devices, to a non-terrestrial network. The architecture considers clusters of IoMT devices that offload their tasks to a dedicated unmanned aerial vehicle (UAV) serving as a multi-access edge computing (MEC) server, which can compute the task or further offload it to an available high-altitude platform station (HAPS) or to a low-earth orbit (LEO) satellite for remote computing. We formulate a problem that has as objective the minimization of the weighted sum delay of the tasks. Given the non-convex nature of the problem, and acknowledging that the complexity of the optimization algorithms impact their performance, we derive a low-complexity joint subchannel allocation and offloading decision algorithm with dynamic computing resource initialization, developed as a greedy heuristic based on convex optimization criteria. Simulations show the gain obtained by including the different non-terrestrial nodes against architectures without them.