The Internet of Things (IoT) (a digital interconnection of everyday objects by means of the internet) has been promising to revolutionize the world for many years. Today, IoT is present in simple devices. For instance, we can all track our sleep, the temperature of the house, or the lights of the living room. However, our lives are still more or less the same. The promised revolution is taking longer than expected…

 

Ideally, society would need objects that can reliably communicate with low latency and that do not consume their batteries too rapidly. Without these characteristics, it is difficult that IoT would be used in the truly “important” things. Rather, IoT would become the “Internet of superfluous things” instead of the “Internet of Important Things”.

 

Currently, the approach that is realizing this ambition more successfully is the Industrial IoT. Because of the mission-critical nature of their requirements, industrial-grade IoT use cases such as asset monitoring in offshore drilling platforms or complex management of industrial processes in refineries, do really require IoT technologies that guarantee deterministic performance, low communication latency (being late may be as bad as a dropped packet) and low power consumption.

 

In this blog entry, I take the Industrial IoT as a starting point, and I propose the following question: “Can we still squeeze the Industrial IoT to get more out of it?” Based on the Ph.D. thesis I defended almost two years ago, I would say the answer is yes. Two of the main contributions were focused on usage cases of Industrial IoT in unprecedented ideas.

 

• Think of a music festival, where thousands of people wear smart bracelets (similar to the ones you use to measure how many calories you burn running). Could we use industrial but cheap IoT technologies to enable new and better applications in this context? For example, imagine that you could pay for drinks through the bracelet, synchronize the LEDs on the bracelets in the audience to show patterns and figures during a song, or that the bracelets could interact with each other for social purposes. Industrial IoT networks are typically designed for static industrial environments (with a few nodes that remain stationary). Could we scale industrial IoT networks, up to several thousand nodes (wristbands) in environments with a high density of devices? In my thesis, I investigated the scalability of 6TiSCH, one of the most effective and widespread standardized solutions in the field of Industrial IoT, and identified the main problems that arise when networks of this type grow beyond a few dozen. To do this, I proposed two approaches. First, is the inclusion of multi-channel capability in some network nodes. This consists of simultaneously using multiple wireless channels to avoid resource bottlenecks. Second, is the use of a new technology called DeBraS, specially designed for large and dense networks, to increase the scalability of the network. To better understand what is achieved with DeBraS, imagine yourself trying to cross two types of streets. A tram circulates along with the first one, moving along its tracks following certain timetables. The second street is… a highway! One where many cars pass constantly. As you can imagine, it would be much easier for you to find a suitable moment to cross the street of the tram without being run over. On the highway, you would have to look very well both ways several times to start running towards the other side. The idea of ​​DebraS is precisely to force you to take a good look at both sides of the highway and choose the right moment when no cars pass to cross it (that is, choose the frequency and the exact moment to send a package). DebraS makes network nodes “listen” to their neighbors when they send data packets to the network, in order to avoid collisions. This is something that the standard 6TiSCH technology did not need because tram traffic is much less and they pass at known times. The results showed that DeBraS increases the forwarding rate per node, significantly increasing the scalability of the network. The price to pay is higher energy consumption. However, for cases like the music festival, where smart wristband batteries are expected to last several days, it should not be a problem. In return, the bracelets can benefit from the advantages of the Industrial IoT. That is deterministic operation, low communication delay, and relatively low power consumption. If you are a little more interested in the subject, in my work I developed a 6TiSCH Simulator with a free software license to quickly narrow down and prototype new technologies based on 6TiSCH. See [1,2,3].

 

Photo by Melissa Askew in Unsplash

• Imagine now that you are cycling with a group of friends in the Pyrenees, emulating a queen stage of the Tour de France. Could we make it possible for cyclists to see the performance and stats of each of their friends, such as their speed, power, or location in real-time? WiFi technology is not energy efficient enough and mobile 3G/4G network is not always available in rural areas. Again, could we leverage industrial IoT technologies like 6TiSCH to create a dynamic wireless mesh network between cyclists and still ensure deterministic performance, low communication latency, and low power consumption? The CONAMO project, where I played an important role in the design and development of the network system, focused precisely on this problem. In this project, I proposed to modify some traditional mechanisms of 6TiSCH to optimize the operation of the network in such a dynamic environment. This consisted of making each node in the network choose, at a given moment, which node is the best to send a packet to a certain destination and select the correct time interval and frequency to send it. I also increased the standard communication range of 6TiSCH by lowering its frequency band from 2.4 GHz to 868 MHz. These tasks were some of the most challenging of my Ph.D., involving complicated software and hardware development, and field-testing with cyclists. Real (including the Royal Belgian Cycling Society). Fortunately, this effort paid off. A final demo with over 20 cyclocross riders was held in Heverlee (Leuven)! A video about the final CONAMO demo can be found here: https://www.vrt.be/vrtnws/nl/2018/11/07/nieuwe-datatechnologie-volgt-prestaties-wielrenners-in-real-time/ (Dutch). This work on 6TiSCH is worldwide unique in terms of network size and node mobility, and it is really in the frontier of what Industrial IoT technologies are able to offer outside of their industry niche. See [4].

 

 

Finally, I would like to point out that it would not have been possible to do this work alone. As the saying goes, all research is built “on the shoulders of giants”. In my case, the giants of this doctorate were, on the one hand, my colleagues, who were fundamental in my thesis by filling these research ideas with guidelines, suggestions, and sometimes slightly controversial nuances. On the other hand, the great open source community has generously shared many interesting projects, tools, and datasets, making the research journey more friendly and humane.

 

 

[1] Esteban Municio, and Steven Latré. "Decentralized broadcast-based scheduling for dense multi-hop TSCH networks" In Proceedings of the MobiCom Workshop on Mobility in the Evolving Internet Architecture (MobiArch), pp. 19-24. ACM, 2016. [2] Esteban Municio, Kathleen Spaey, and Steven Latré. "A distributed density optimized scheduling function for IEEE 802.15.4e TSCH networks" In Transactions on Emerging Telecommunications Technologies 29, no. 7 (2018): e3420. [3] Esteban Municio, Glenn Daneels, Mališa Vučinić, Steven Latré, Jeroen Famaey, Yasuyuki Tanaka, Keoma Brun, Kazushi Muraoka, Xavier Vilajosana, and Thomas Watteyne. "Simulating 6TiSCH networks." In Transactions on Emerging Telecommunications Technologies 30, no. 3 (2019): e3494. [4] Esteban Municio, Glenn Daneels, Mathias De Brouwer, Femke Ongenae, Filip De Turck, Bart Braem, Jeroen Famaey, and Steven Latré. "Continuous Athlete Monitoring in Challenging Cycling Environments using IoT Technologies" In IEEE Internet of Things Journal 6.6 (2019): 10875-10887. [5] Esteban Municio, Johann Marquez-Barja, Steven Latré, and Stefano Vissicchio. "Whisper: Programmable and Flexible Control on Industrial IoT Networks" In Sensors, no. 11 (2018): 4048. [6] Esteban Municio, Steven Latré, and Johann Marquez-Barja. "Extending Network Programmability to the Things Overlay using Distributed Industrial IoT Protocols" In IEEE Transactions on Industrial Informatics (2020)

 

-

 

This article was sent by Esteban Municio. Esteban is a Telecommunication Engineer and Researcher specialized in computer networks currently working in the AI-driven System group of the i2CAT Foundation. He received a B.Sc+M.Sc. degree of Telecommunication Engineering from the Madrid Polytechnic University (UPM) in 2013, a M.Sc degree in Networks and Computer Systems from the King Juan Carlos University (URJC) in 2014, and a PhD degree in Computer Science in imec - University of Antwerp in 2020. After finishing his PhD, he continued in imec - University of Antwerp as Postdoctoral Researcher for two years, working in the context of ultra-reliable IoT networks and end-to-end network programmability. He has been involved in some European research projects such as FP7 TUCAN3G, Celtic+ FlexNet, H2020 ProTego, H2020 InterConnect, H2020 Vital-5G and H2020 DAEMON. His main research interests are: traffic engineering, SDN, programmable wireless networks, TSN, network orchestration and ultra-reliable Industrial IoT. He is also interested in heterogeneous backhaul networks, smart cities deployments, community networks and connectivity provision in rural environments. You can find more about him in his web page: https://emunicio.github.io/