Enhancing graph-routing algorithm for industrial wireless sensor networks

Alharbi, Nouf Helal (2024) Enhancing graph-routing algorithm for industrial wireless sensor networks. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 2023AlharbiPhD.pdf] PDF
Download (6MB)


Industrial Wireless Sensor Networks (IWSNs) are gaining increasing traction, especially in domains such as the Industrial Internet of Things (IIoT), and the Fourth Industrial Revolution (Industry 4.0). Devised for industrial automation, they have stringent requirements regarding data packet delivery, energy consumption balance, and End-to-End Transmission (E2ET) time. Achieving effective communication is critical to the fulfilment of these requirements and is significantly facilitated by the implementation of graph-routing – the main routing method in the Wireless Highway Addressable Remote Transducer (WirelessHART), which is the global standard of IWSNs.

However, graph-routing in IWSN creates a hotspot challenge resulting from unbalanced energy consumption. This issue stems from the typical configuration of WirelessHART paths, which transfers data packets from sensor nodes through mesh topology to a central system called the Network Manager (NM), which is connected to a network gateway. Therefore, the overall aim of this research is to improve the performance of IWSNs by implementing a graph-routing algorithm with unequal clustering and optimisation techniques.

In the first part of this thesis, a basic graph-routing algorithm based on unequal clustering topologies is examined with the aim of helping to balance energy consumption, maximise data packet delivery, and reduce the number of hops in the network. To maintain network stability, the creation of static clusters is proposed using the WirelessHART Density-controlled Divide-and-Rule (WDDR) topology. Graph-routing can then be built between Cluster Heads (CHs), which are selected according to the maximum residual energy rate between the sensor nodes in each static cluster. Simulation results indicate that graph-routing with the WDDR topology and probabilistic unequal clustering outperforms mesh topology, even as the network density increased, despite isolated nodes found in the WDDR topology.

The second part of this thesis focuses on using the Covariance-Matrix Adaptation Evolution Strategy (CMA-ES) algorithm. This addresses the three IWSN requirements that form the focus of this research, by proposing three single-objective graph-routing paths: minimum distance (PODis), maximum residual energy (POEng), and minimum end-to-end transmission time (POE2E). The research also adapts the CMA-ES to balance multiple objectives, resulting in the Best Path of Graph-Routing with a CMA-ES (BPGR-ES). Simulation results show that the BPGR-ES effectively balances IWSN requirements, but single-objective paths of graph-routing does not achieve balanced energy consumption with mesh topology, resulting in a significant reduction in the efficiency of the network.

Therefore, the third part of this thesis focuses on an Improvement of the WDDR (IWDDR) topology to avoid isolated nodes in the static cluster approaches. The IWDDR topology is used to evaluate the performance of the single-objective graph-routing paths (PODis, POEng, and POE2E). The results show that in IWDDR topology, single-objective graph-routing paths result in more balanced energy consumption.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Additional Information: Supported by funding from Taibah University, Saudi Arabia.
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Colleges/Schools: College of Science and Engineering > School of Computing Science
Supervisor's Name: Mackenzie, Dr. Lewis and Pezaros, Professor Dimitrios
Date of Award: 2024
Depositing User: Theses Team
Unique ID: glathesis:2024-84135
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 28 Feb 2024 17:00
Last Modified: 28 Feb 2024 17:00
Thesis DOI: 10.5525/gla.thesis.84135
URI: https://theses.gla.ac.uk/id/eprint/84135
Related URLs:

Actions (login required)

View Item View Item


Downloads per month over past year