07 Nov 2018


Xavier Bush

Jesper Lindström
Prof. James Gross


It is common knowledge that Industry 4.0, or Industrial IoT, is eagerly seeking solutions for tomorrow’s manufacturing shop floor, in which automation equipment will have to become much more flexible, adaptable and modular. There is even a huge interest in having production setups prepared for product individualisation that would adapt to customer needs. Naturally, key components towards this higher flexibility are wireless networks, which in addition to flexibility can also facilitate mobile use cases, as well as fast operational scalability. However, wireless technologies struggle to be included in time-critical factory automation processes. Why is this so?

To answer this question, we draw a comparison to a time-period a couple of decades ago in which the automation industry was seeking ways to innovate manufacturing towards higher efficiencies. Obviously, the industry figured out that one of the factors that dramatically contributes towards these goals is the way machinery communicates. Thus, not surprisingly, automation systems adopted early on industrial fieldbus solutions allowing for cable reduction and simpler architectures. It is safe to say today that the introduction of wired fieldbus systems has been a significant success story in the automation industry, in which they are still dominating the market. It is clear that this success story was possible thanks to a perfect business-technology match.

The dominance of fieldbus systems, and later Ethernet-based fieldbus systems, in this sector contrasts with the small impact that wireless communications have had in industrial automation so far. While, wireless technologies have been thriving enormously over the last two decades in consumer scenarios (mainly by the advancement of cellular and Wi-Fi networking technologies) or even in some industrial applications (e.g. WirelessHART in process automation), their role is non-existent in the factory automation domain when it comes to time-critical applications within single digit milliseconds.

Two aspects come here into play so far that have been enhancing each other. On the one hand, over the past decades pressure to move towards wireless technologies due to cost savings have been low, if at all. Wired fieldbus solutions were good enough regarding performance, while the industry generally has a slow-moving and conservative approach to adopting new technologies. On the other hand, from a technological point of view, 25 years ago wireless technologies were still in its infancy, and hardly ready for the mass-deployment with respect to real-time constrained application scenarios. Simply speaking, the performance requirements of industrial automation were too high.

However, the already mentioned interest of the Industry 4.0 in the new shop-floor opens up new pressure towards the adoption of wireless scenarios. Which brings us to the question of technological readiness. Due to the proliferation of consumer-grade wireless systems, there is a firm viewpoint in factory automation that wireless, in general, is not reliable enough for real-time applications in the field. But wireless technical development, i.e. research and engineering, has constantly been evolving over the last 25 years bringing new solutions. Is the point of view of the automation industry about wireless technologies therefore still valid?

We will dig into the details of this question by highlighting the challenges faced when trying to optimise wireless technologies for high reliability and ultra-low latency. In the past, wireless technologies often have failed to deliver on these characteristics due to different random effects of the wireless medium not adequately accounted for. However, cutting-edge research over the last decade has devised concepts to overcome these random effects, which have been implemented in our wireless technology EchoRing.

We address these specific challenges along three interrelated blog posts. Blog Post I “Wireless Propagation Effects and Their Impact on Reliable Transmissions” deals with the random effects occurring purely from the so-called wireless propagation environment itself, i.e. scattering and reflections of radio waves, blocking through larger objects, etc. The next two blog posts address the complications arising when multiple nodes – or generally speaking multiple electromagnetic transceivers – transmit on the same radio resources. This leads to interference, which can have a devastating random impact on the reliability of wireless systems. This topic is covered in Blog Post II “Avoiding Collisions: Adding Determinism in My Network” and in Blog Post III “Interference: Other Wireless Networks Affect Mine”.

After following through the first three blog entries, the reader will understand that the challenges presented by the wireless channel can be tackled (when addresses appropriately) and that our technology EchoRing presents a cable-like performance suitable for time-critical industrial applications. Therefore, just as happened decades ago with fieldbus systems (going back to the parallelism), there is finally a business-technology match regarding wireless technologies.

Once we are done with the first blog series, we will continue this blog by focusing on the following three areas. Area I will cover additional technical requirements (e.g. co-existence with other wireless technologies) and its solutions (e.g. anticipating dynamic frequency hopping mechanism) for a scalable product. Area II will address our experience with real-live deployment of our EchoRing products and learnings in different industrial settings. Area 3 will discuss our experience of interfacing a high reliability and low-latency wireless network with Ethernet-based fieldbus systems transparently.