Symbol-Synchronous Buses: Deterministic, Low-Latency Wireless Mesh Networking with LEDs

Industrial real-time control applications often require performance guarantees that contemporary wireless networks fail to offer. The cause of this shortcoming lies in the complexity and inefficiency of current wireless network architectures, which must coordinate multiple nodes across dense deployments, and hence today cannot match the latency levels offered by wired bus networks used for real-time control, such as CAN and Profibus.

This talk describes a novel networking paradigm that leverages the abundance of optical spectrum to address that gap. The paradigm, referred to as a symbol-synchronous bus, has nodes that concurrently transmit optical signals and thus delivers a wireless mesh network with a performance envelope resembling that of a wired bus in terms of deterministic latency and throughput. Our prototype platform, called Zero-Wire, confirms that symbol-synchronous buses unlock a novel end-to-end performance envelope for wireless mesh networks: latency under 1 ms across four hops, and jitter on the order of 10s of μs. These early results suggest a bright future for the under-explored area of optical wireless mesh networks in delivering ubiquitous connectivity through a low-complexity physical layer.

A presentation by Jonathan Oostvogels, Doctoral Researcher at imec-DistriNet, KU Leuven.

The view of the speaker

Question 1: What drives you?
Curiosity and the opportunity to innovate.

Question 2: Why should the delegate attend your session?
The presentation covers an emerging technology that unlocks new application areas for optical wireless communication and enables wireless networks in settings that currently still depend on their wired counterparts.

Question 3: What emerging technologies / trends do you see as having the greatest potential in the short and long run?
The abundance of bandwidth in the optical part of the spectrum stands in stark contrast with the scarcity of radio spectrum, which has guided decades of wireless system design. This difference in kind may lead to much more profound innovations in networking technology than appreciated today.

Question 4: What kind of impact do you expect them to have?
The exploration of network architectures that leverage the unique properties of optical links to fundamentally rethink network operation at all levels may revolutionise wireless networking in unexpected ways and quickly solve problems that appear fundamentally hard on traditional network stacks.

Question 5: What are the barriers that might stand in the way?
There is a tremendous amount of technological lock-in surrounding the way networked applications are built. Our appetite for new applications and deployment areas tends to change much faster than the ability to fundamentally rethink the way these applications should be implemented.

About Jonathan Oostvogels
Jonathan Oostvogels is a doctoral researcher at KU Leuven and a member of the imec-DistriNet research group. Jonathan works on the design of next-generation network architectures and techniques for low-latency and high-throughput wireless cyber-physical systems. His work has been featured in several top-tier publications, including both academic and popularising media.

About imec-DistriNet, KU Leuven
The imec-DistriNet research group is part of the Department of Computer Science at the KU Leuven. The general domain of expertise and innovation of DistriNet is the development of open, distributed object support platforms for advanced applications. The research is always application driven and is often conducted in close collaboration with industry.

Jonathan Oostvogels is speaker at the 2022 edition of LiFi Conference.

Leave a Reply

Your email address will not be published.