Highlight your achievements on People@HES-SO More info
PEOPLE@HES-SO – Directory and Skills inventory
PEOPLE@HES-SO – Directory and Skills inventory

PEOPLE@HES-SO
Directory and Skills inventory

Help
language
  • fr
  • en
  • de
  • fr
  • en
  • de
  • SWITCH edu-ID
  • Administration
« Back
Karimian Mahboob

Karimian Mahboob

Chargé de Ra&D

Main skills

Wireless Communication

IoT Internet Of Things

Embedded Systems

  • Contact

  • Conferences

Main contract

Chargé de Ra&D

Desktop: C0.07

Haute Ecole d'Ingénierie et de Gestion du Canton de Vaud
Route de Cheseaux 1, 1400 Yverdon-les-Bains, CH
HEIG-VD
Institute
IICT - Institut des Technologies de l'Information et de la Communication
No data to display for this section

2025

Optimum clustering for multi-border router wireless mesh networks
Conference ArODES

Mahboob Karimian, Yann Charbon, Pierre Favrat

Proceedings of the 2025 International Conference on Artificial Intelligence, Computer, Data Sciences and Applications (ACDSA)

Link to the conference

Summary:

Low-power wireless networks constructed of thousands of nodes forming a mesh topology are becoming increasingly common. Various standards notably Wi-SUN tried to provide a solid frame for the networking part of these networks. However, as the network size grows, congested nodes become a bottleneck and make the network unreliable. Many routing algorithms, including RPL used by most mesh network standards, are based on the local decision of each node without having a global prevision of the ongoing traffic in other links. Additionally, the propagation of traffic information from each link creates an extra load in the network and increases management complexity. To tackle this problem, one solution is to optimize the network for high Quality of Service (QoS) by adding multiple border routers. It allows splitting a large network into manageable sub networks. This remains a cheaper and simpler solution because instead of creating many control packets, we only redirect traffic to a new sink which has high bandwidth and reliable connection towards the cloud/destination. Still, adding border routers needs an arrangement of the whole network so that only the minimum number of border routers for the level of QoS requested by the service user is used. This arrangement consists of finding the best location for the new border routers. This is done in two essential steps: i) Clustering the current big network into small sub-networks ii) Finding the nodes in the cluster that are optimal to be a new border router.

Wireless mesh networks routing optimization using machine learning and deep learning
Conference ArODES

Yann Charbon, Eric Tran, Mahboob Karimian, Pierre Favrat

Proceedings of the 2025 International Conference on Artificial Intelligence, Computer, Data Sciences and Applications (ACDSA)

Link to the conference

Summary:

Wireless mesh networks (WMNs) are increasingly deployed in large-scale infrastructures, where efficient and reliable routing is essential to maximise performance. Conventional protocols such as RPL rely on local decision-making and often produce globally suboptimal Directed Acyclic Graphs (DAGs), particularly in dense or complex topologies. This work investigates whether machine learning (ML) and deep learning (DL) techniques can be used to globally optimise DAG-based routing in WMNs. To support this exploration, we developed a high-performance simulator that evaluates the routing efficiency of arbitrary spanning trees using realistic link quality metrics derived from physical parameters such as RSSI. This simulator generates a large labelled dataset of optimal and suboptimal trees across randomly generated topologies, which is then used to train various ML and DL models. We formulate the optimisation task as a multi-label classification problem, where models predict the set of edges forming the optimal DAG. Despite extensive experimentation with classical classifiers, MLPs, and RNNs, our results show that standard models fail to outperform RPL, primarily due to the difficulty of learning under strict DAG constraints, such as acyclicity and single-parent-per-node rules, which render the optimisation problem non-differentiable. However, the simulator consistently identifies DAGs that outperform RPL by up to 16% in terms of routing efficiency. These findings underscore the limitations of unconstrained learning in combinatorial domains and point toward future research into constraint-aware, structure-driven learning frameworks for global WMN optimisation.

Achievements

Media and communication
Contact us
Follow the HES-SO
linkedin instagram facebook twitter youtube rss
univ-unita.eu www.eua.be swissuniversities.ch
Legal Notice
© 2021 - HES-SO.

HES-SO Rectorat