Software Defined Networking for the Industrial Internet of Things

Student thesis: Doctoral ThesisDoctor of Philosophy (PhD)


While the increasing ubiquity of embedded devices has given rise to the `smart' moniker applied to everyday objects, the underpinning wireless communication protocols collectively incorporate them into the Internet of Things (IoT). These protocols facilitate communication to, from, and between objects, and have spawned ever more sophisticated applications in both the home and in industry. While much of the publicity around this phenomenon has focused on domestic uses, a quiet revolution has been taking place within industrial and commercial sectors: as the initial promises of IoT are finally met by technological capability. It is here, in the Industrial Internet of Things (IIoT), where there is the opportunity to streamline operations, such as managing warehouse stock for Just-in-Time (JIT) manufacturing; support safety-critical systems through the complex monitoring of sensor and actuator networks; and offer new business models, where IoT infrastructure can be offered as a service (IoTaaS).

Traditionally, low-power wireless mesh networks have been at the forefront of the IoT conversation. However, the increasing complexity of these networks has laid bare the shortcomings inherent in current control architectures and protocols, where lossy channels and multi-hop topologies present significant challenges. To address these issues, there has been considerable interest in applying the concepts of Software Defined Networking (SDN), which over the past decade has liberated data centre and campus network management from reliance on vertical infrastructure practices. However, the centralised SDN approach faces considerable challenges in the constrained environments present in low-power wireless networks.

This thesis explores not only how SDN concepts can be used to provide dynamic and flexible control in IIoT, but crucially how to address and manage the SDN overhead. It presents analytical, simulated, and experimental results, as well as the design and implementation of two novel SDN architectures for low-power wireless networks: μSDN and Atomic-SDN. These results demonstrate both that the challenges of applying SDN within constrained IoT networks can be overcome, and that SDN can be used to address the complex and diverse traffic requirements of IIoT applications across low-power wireless networks. The synchronous flooding approach of Atomic-SDN, in particular, provides an effective means of achieving the one-to-many traffic pattern required in distributed control systems, and makes it a highly promising solution for deploying SDN within low-power wireless IIoT networks.
Date of Award24 Mar 2020
Original languageEnglish
Awarding Institution
  • The University of Bristol
SupervisorReza Nejabati (Supervisor) & George Oikonomou (Supervisor)


  • SDN
  • Mesh Networks
  • IoT
  • IEEE 802.15.4
  • Low-Power Wireless
  • RPL
  • 6TiSCH
  • Concurrent Transmissions
  • Synchronous Flooding
  • TSCH
  • Multi-Hop
  • Programmable Networks
  • Adaptive Networking
  • IIoT
  • Smart Metering
  • Smart City

Cite this