Self-organization scheme for balanced routing in large-scale multi-hop networks

Abstract

We propose a self-organization scheme for cost-effective and load-balanced routing in multi-hop networks. To avoid overloading nodes that provide favourable routing conditions, we assign each node with a cost function that penalizes high loads. Thus, finding routes to sink nodes is formulated as an optimization problem in which the global objective function strikes a balance between route costs and node loads. We apply belief propagation (its min-sum version) to solve the network optimization problem and obtain a distributed algorithm whereby the nodes collectively discover globally optimal routes by performing low-complexity computations and exchanging messages with their neighbours. We prove that the proposed method converges to the global optimum after a finite number of local exchanges of messages. Finally, we demonstrate numerically our framework's efficacy in balancing the node loads and study the trade-off between load reduction and total cost minimization.

Publication DOI: https://doi.org/10.1088/1751-8121/abd34b
Divisions: College of Engineering & Physical Sciences > Mathematics
College of Engineering & Physical Sciences > Systems analytics research institute (SARI)
College of Engineering & Physical Sciences
Additional Information: Funding: This work was supported by the Independent Research Fund Denmark under Grant ID DFF–5054-00212, the Leverhulme Trust Grant No. RPG-2018-092, EPSRC programme grant number EP/R035342/1 TRANSNET (D.S.) and EPSRC grant number EP/N002350/1 (“Spatially Embedded Networks”)
Full Text Link:
Related URLs: https://iopscie ... 751-8121/abd34b (Publisher URL)
PURE Output Type: Article
Published Date: 2021-01-29
Published Online Date: 2020-12-14
Accepted Date: 2020-12-14
Authors: Badiu, Mihai Alin
Saad, David (ORCID Profile 0000-0001-9821-2623)
Coon, Justin P

Download

[img]

Version: Accepted Version

License: Creative Commons Attribution

| Preview

[img]

Version: Published Version

License: Creative Commons Attribution

| Preview

Export / Share Citation


Statistics

Additional statistics for this record