Saved in:
Bibliographic Details
Main Authors: Zhu, Botao, Wang, Xianbin
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2405.20055
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913968767893504
author Zhu, Botao
Wang, Xianbin
author_facet Zhu, Botao
Wang, Xianbin
contents With the growing scale and intrinsic heterogeneity of Internet of Things (IoT) systems, distributed device collaboration becomes essential for effective task completion by dynamically utilizing limited communication and computing resources. However, the separated design and situation-agnostic operation of computing, communication and application layers create a fundamental challenge for rapid task-resource matching, which further deteriorate the overall task completion effectiveness. To overcome this challenge, we utilize hypergraph as a new tool to vertically unify computing, communication, and task aspects of IoT systems for an effective matching by accurately capturing the relationships between tasks and communication and computing resources. Specifically, a state-of-the-art task-resource matching hypergraph (TRM-hypergraph) model is proposed in this paper, which is used to effectively transform the process of allocating complex heterogeneous resources to convoluted tasks into a hypergraph matching problem. Taking into account computational complexity and storage, a game-theoretic hypergraph matching algorithm is proposed via considering the hypergraph matching problem as a non-cooperative multi-player clustering game. Numerical results demonstrate that the proposed TRM-hypergraph model achieves superior performance in matching of tasks and resources compared with comparison algorithms.
format Preprint
id arxiv_https___arxiv_org_abs_2405_20055
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Hypergraph-Aided Task-Resource Matching for Maximizing Value of Task Completion in Collaborative IoT Systems
Zhu, Botao
Wang, Xianbin
Systems and Control
With the growing scale and intrinsic heterogeneity of Internet of Things (IoT) systems, distributed device collaboration becomes essential for effective task completion by dynamically utilizing limited communication and computing resources. However, the separated design and situation-agnostic operation of computing, communication and application layers create a fundamental challenge for rapid task-resource matching, which further deteriorate the overall task completion effectiveness. To overcome this challenge, we utilize hypergraph as a new tool to vertically unify computing, communication, and task aspects of IoT systems for an effective matching by accurately capturing the relationships between tasks and communication and computing resources. Specifically, a state-of-the-art task-resource matching hypergraph (TRM-hypergraph) model is proposed in this paper, which is used to effectively transform the process of allocating complex heterogeneous resources to convoluted tasks into a hypergraph matching problem. Taking into account computational complexity and storage, a game-theoretic hypergraph matching algorithm is proposed via considering the hypergraph matching problem as a non-cooperative multi-player clustering game. Numerical results demonstrate that the proposed TRM-hypergraph model achieves superior performance in matching of tasks and resources compared with comparison algorithms.
title Hypergraph-Aided Task-Resource Matching for Maximizing Value of Task Completion in Collaborative IoT Systems
topic Systems and Control
url https://arxiv.org/abs/2405.20055