We investigate the suitability of various communication technologies and protocols for wireless sensor connectivity in industrial environments. Emphasis is given to the robustness of communication against fading and interference, energy efficiency, and the use of relay communication. The performance of ZigBee, ultra-wideband (UWB), and LoRa has been experimentally tested in production plants, airplanes, and space launch vehicles in collaboration with industrial partners.
Modern factories utilize large numbers of sensors to monitor and control processes, machines, and goods. These sensors measure physical or chemical parameters and typically report them to a control center, which analyzes the measurements and sends commands to actuators. Such industrial automation requires a reliable communication system, which is traditionally implemented through wired connections. However, there is a desire to replace these cables with wireless connections, which will generate massive benefits in terms of quick and cost-effective factory reconfiguration. Such flexibility is especially important for small lot sizes and short product cycles. Similar to the application in factories, sensor networks find their way into vehicles, trains, and airplanes, where wireless connectivity saves weight and reduces the complexity of installation and maintenance. Future airplanes, for example, will be equipped with thousands of embedded sensors and actuators for structural health monitoring, passenger and aircrew assistance, and aircraft control. These examples illustrate that sensor networks play a vital role in today’s industry, and that appropriate radio technologies and communication protocols are to be developed in order to meet the demands of industrial applications. These requirements include system security, reliability, energy efficiency, lifetime, coexistence, and real-time capabilities, to name just a few. In addition, the wireless connectivity has to be robust against signal fading and interference while meeting requirements in terms of data rate and delay. A key question in this context is: Which wireless technology is best suited for such industrial demands? To contribute to this important domain, my team and I have conducted a number of funded research projects with a strong experimental focus and in collaboration with industry partners.
Our first project designed and investigated novel communication techniques based on cooperative relays. The basic concept is as follows: A relay device supports the communication from a sender to a destination by overhearing packets and rebroadcasting the ones the destination cannot properly decode, thus providing a second chance to receive it. We studied the performance of relay communications for industrial applications based on real-world measurements. We employed low-cost, off-the-shelf IEEE 802.15.4 devices in a harsh and cluttered factory environment. Results show that relaying protocols outperform conventional retransmissions in terms of delivery ratio and number of retransmissions. “Reactive relay selection provides the best overall delivery ratio,” we state for the setting in our 2014 article published in the IEEE Transactions on Industrial Informatics and conclude that cooperative relaying can be beneficial in large industrial sensor networks but requires a careful choice of the associated protocols.
The goal of a collaboration with Airbus was to evaluate off-the-shelf wireless technologies for different applications in the aeronautic sector with emphasis on the connectivity of sensors in passenger airplane cabins. We specified industry-relevant applications, usage scenarios, and quality of service requirements and considered regulatory aspects. The first question was the choice of technology: IEEE 802.11 (WiFi) offers high data rates but is power hungry; IEEE 802.15.1 (Bluetooth) and IEEE 802.15.4 (ZigBee, WirelessHART, ISA100.11a) are cheaper and more energy efficient at lower data rates; but all three typically operate in unlicensed bands and thus may interfere with devices carried by passengers. This led us to ultra-wideband (UWB) communications standardized in IEEE 802.15.4-2011 as an alternative technology. UWB devices transmit short pulses, thus spreading the signals to be transmitted over a large bandwidth. UWB has many unique features that are relevant for industrial environments: It comes with precise self-localization and permits high dates rates up to 27 Mbit/s over short distances with low power consumption. State-of-the-art UWB transceivers have a small form factor, and their noise-like signal is difficult to detect and robust against narrowband jamming. In fact, UWB transceivers are mainly used for localization tasks today, but our objective was to asses their suitability for real data communications. As part of our activities, we presented a proof-of-concept for an UWB sensor network deployed in a passenger cabin of a commercial Airbus. We investigated the packet loss rate and studied combinations of spatial and temporal diversity to improve performance. These measurements were the first ones reported for state-of-the-art UWB devices in such a setting and shed light on the potential of UWB to support emerging aeronautic applications. Our work presented at the ACM International Conference on Modelling, Analysis and Simulation of Wireless and Mobile Systems (MSWiM) in 2017 received the best paper award. Related experiments with UWB devices were also conducted in two industrial scenarios, namely a large-size aircraft assembly hangar and a medium-size production hall, and inside an Ariane5 launch vehicle, assessing received signal power fluctuations and connectivity.
Our recent work in this area is done in two applied projects with regional partners. In UWB4I, we build on UWB to develop and test a self-powered industrial sensor network for condition monitoring. A number of applications with different requirements for communications and measurement processing are being investigated. We bring three design dimensions together: energy management, networking, and data analysis. The interactions between them are the core of our approach and drive the solutions for the particular applications. Another, very large project is “Dependable, secure and time-aware sensor networks (DESSNET)”. Several use cases for measurement collections have been defined with Treibacher Industrie, focusing on predictive maintenance. The main challenge is to establish a wireless network in an environment full of steel constructions and chemicals. For this purpose, we currently test a Long Range Wide Area Network (LoRaWAN), which combines robustness against small-scale fading with ultra low energy consumption at the expense of limited data rates.
- Towards industrial ultra-wideband networks: Experiments for machine vibration monitoring
IEEE Access, March 2020
- DISTY: Dynamic stochastic time synchronization for wireless sensor networks
IEEE Transactions on Industrial Informatics, June 2017
- An experimental study of selective cooperative relaying in industrial wireless sensor networks
IEEE Transactions on Industrial Informatics, August 2014
- Long-term LoRa experiments in a chemical plant
- Cooperative relaying in LoRa sensor networks
- Allocation of repetition redundancy in LoRa
Ultra-wideband UWB (IEEE 802.15.4-2011)
- HiPR: High-precision UWB ranging for sensor networks
- Study of a self-powered UWB sensor network for industrial applications
- UWB connectivity inside a space launch vehicle
- Experimental study of UWB connectivity in industrial environments
- Experimental study of packet loss in a UWB sensor network for aircraft (best paper award)
- Experiments with UWB aircraft sensor networks
Zigbee / IEEE 802.15.4
- Experimental evaluation of PCO synchronization in 802.15.4 networks
- Neighbor cardinality estimation with low-power transceivers: …
- Measurement-based analysis of adaptive relay selection in industrial wireless sensor networks
- Measurement-based analysis of cooperative relaying in an industrial wireless sensor network
Photo made by Daniel Neuhold at ZAL Zentrum für Angewandte Luftfahrtforschung with kind permission.