NIST Tackles Wireless Industrial Systems

The National Institute of Standards and Technology (NIST) is working to provide best practice guidelines to help manufacturers use wireless systems.

There is a strong case to be made for using wireless networking for connected industrial systems. In a connected factory, the equipment for each stage of the manufacturing process is automatically controlled in a coordinated way. To make this work cost-effectively and safely, you also need to collect information from your devices. So, machines are networked together for both sensing and control.

For decades, this has been done using a variety of wired protocols. But hard-wiring machine systems in a factory has drawbacks. The costs and difficulties of the initial installation and of subsequent changes are high.

But installing wireless systems in industrial environments can be challenging because of reliability and safety concerns.

The following is excerpted from a discussion with Rick Candell, Mohamed Hany, and Kang Lee, of the Industrial Wireless Systems (IWS) project team at the Communications Technology Laboratory (CTL) at the National Institute of Standards and Technology (NIST).

Figure 1. Wireless communication is needed to operate mobile robots in a factory. (Image: AsgherAbbas/Adobe Stock)

Tech Briefs: Why use wireless networks, given the various difficulties with them?

Candell: The primary reasons are cost and device density. The costs associated with cabling can be quite high. When you have cables, you have a lot of material, a lot of labor to install the cables and a lot of weight that has to be accommodated. Wireless obviates all that.

Not only does wireless bring a large savings to the installation up front, if an existing installation is being upgraded, running new cables through all that mess could be really daunting.

However, with wired connectivity, there’s guaranteed reliability, given the simplicity of the technology. Although the installation of wireless is simpler, reliability may have to suffer a little bit.

Hany: I would add that there are some industrial applications where wireless is the only reasonable solution. In harsh environments or difficult locations, wireless is a need, not just an option. Also, there are all the new mobile robots and autonomous guided vehicles (AGVs). The number of applications that really need wireless keeps increasing with time.

Tech Briefs: What about problems of reliability with wireless networks, especially in an industrial environment?

Candell: There are ways to design systems to be more resilient and robust to harsh environmental conditions. A lot of that comes down to RF band selection. If there is machine noise, for example, you wouldn’t want to choose a communications band below one GHz. You’re probably going to want something above three GHz.

In addition, you can architect your system to apply diversity techniques. For example, spatial diversity, where you use reflections in the environment to your advantage — or multiple antennas.

Figure 2. Proper location of network nodes is critical. (Image: NIST)

Frequency hopping is a diversity technique that allows you to vary frequencies in real-time so certain types of interference won’t impact your signal. It will also protect you from other networks operating on the same channel.

And then there’s time diversity, which is basically just transmitting the same information at more than one moment in time.

Hany: To add to the main question of how wireless can achieve reliability in industrial scenarios, the initial network installation is critical, for example, where to install the different nodes of the network. If all of these basic principles of wireless are taken care of, it can achieve very high reliability — it’s all about understanding how wireless works and making sure that the proper methods are followed when installing the network.

Lee: Technologies like 5G wireless or the coming 6G wireless are enabling standards for improved real-time communications and control processes. In the old days, communications protocols were often proprietary. So, the fact that we have existing wireless standards expands the possibilities for industrial automation and smart manufacturing operations.

Candell: That’s a really good point, Kang. And that leads to another important topic. We’ve done a lot of research on wireless time-sensitive networking (WTSN), which is being adopted through the IEEE and various industry consortia.

If you have a deadline, if you’re transmitting information and that information doesn’t arrive on time, there could be consequences for that. It would be considered a missed packet — that goes against your reliability target.

We’re working with the Avnu Alliance right now to help understand and better support industry in that area, in developing wireless time-sensitive networking (TSN). And the IEEE also has an effort ongoing — the IEEE 60802 standard, which is a TSN profile for industrial applications.

Tech Briefs: What exactly is time-sensitive networking?

Candell: In a nutshell, certain data streams are given different priorities — higher or lower — on the wireless network. Initially it applied to wired networks, and nowadays, it is being adapted for wireless. There are many parts of the specification, but the time-sensitive networking part is all about guaranteeing quality of service for specific applications by providing higher priority for time-critical information flows.

And it all has to be scheduled within the network, be it a wired network or a wireless network, all of that has to be coordinated. That’s part of the problem: coordination of these networks and allotting time for the time-sensitive applications to provide priority access to devices when they need it.

I also want to talk about standardization. Research is a big component of what we do but another very important component is working with industry on standardization. We provide data to industry in terms of measurements and techniques to focus on reliability, latency, and changing propagation effects. We take measurements of wireless propagation, and have produced very large data sets.

In addition, we offer workshops and guidance on testing techniques and methodologies. We work with industry through standards development organizations (SDOs) such as the IEEE. In fact, I chair the IEEE P3388 working group to produce standards on how to test and evaluate the performance of industrial wireless systems.

Tech Briefs: Could you tell me something about how and what one would test.

Candell: It comes down to understanding the requirements of your application so you can determine what the communications system needs to be able to support. Testing then becomes a process of the selection of what you’re going to test and how. You can test on-site in the environment where you want to deploy, or you can test in the lab using RF test equipment. We have the ability to recreate the propagation environment and the interference environment in our lab. So, part of the IEEE 3388 Standard is specifying how to do that.

Figure 3. A reverberation chamber can be used to recreate field conditions in the lab. (Image: NIST)

For propagation, we have the capability to digitally, or using reverberation chambers, recreate what we’ve observed in the field. For example, we’ve taken measurements at automotive factories at Ford and in an autoclave at a Boeing factory. We have many data sets where we’ve gone in and measured how wireless signals propagate through particular environments. So, we can recreate them in the lab in specific frequency bands of interest all the way up to 60 GHz.

We’re trying to encourage people to do more testing prior to deploying their networks. Once it’s deployed, if it’s not working the way you wanted it to, well, it’s kind of late at that point, so standardized testing is very important.

Hany: The whole idea of replicating scenarios in the lab is that in order to get reliable test results they have to be repeatable and controllable. You have to be able to isolate different factors of the testing scenario.

We can do on-site testing but you will have a lot of limitations about what you can test if you are doing it in an industrial environment. But once we do it in the lab, we can isolate different factors to see their impact on performance. We can also do stress testing in the lab, which we could not easily do in in a real setting.

Figure 4. The wireless environment in each factory is somewhat unique. For example, an airplane factory has different conditions than a car plant. (Image: Marius/Adobe Stock)

And, standardized tests make things fair. Every integrator could claim ‘this system works perfectly’ in a specific environment. But when we have a set of standardized tests that any system will go through, that will make things fair and transparent. If you really would like to install a wireless system, go through this test, and see the results. Then you can fairly compare different systems.

Lee: Something else to understand is, if I’m a manufacturer of wireless networks and I want to sell it to say, Ford as well as Boeing, I might think they can both use the network to achieve the same performance. But, in reality, it might have different performance characteristics, because Boeing has big airplanes in their factory, while Ford has cars and robots doing assembly. So, even though you use the same wireless network in both factories, it behaves somewhat differently. That’s why we are working with the IEEE 3388 Standard — it can help users to investigate how the wireless network will perform in their respective environmental settings.

Tech Briefs: So, is the Standard for testing?

Candell: Yes, performance evaluation — but it’s more than just testing. We present a systems engineering process to provide roles and responsibilities for the evaluation testing. It also includes a part where people have to understand what the requirements are. You can’t test if you don’t know what you’re testing. That’s essential because a lot of times people say: “I want to be able to measure temperature, pressure, flow, or whatever.” So, they just install it, or maybe a vendor says, ‘Use this, it’s a great product.” Then they get it to their site and discover they haven’t necessarily considered all of their requirements — it may or may not provide the functionality or the reliability that they really desired. Then they have to do something to mitigate what they could have identified up front.

In Conclusion

The IEEE 3388, Standard for the Performance Assessment of Industrial Wireless Systems was approved on Feb 12, 2025 by the IEEE Standards Board. This standard establishes a radio frequency (RF) reference environment model, test methodology, and evaluation process for the performance testing of industrial wireless networks. The IEEE 3388-2025 standard is protocol agnostic — it applies to all wireless protocols intended for use in industrial and mission-critical settings in which reliability concerns are essential to the functioning of the overlying operational systems.

A copy of the draft IEEE 3388-2025 standard can be acquired from the IEEE-SA Standards Store here .

This article was written by Ed Brown, editor of Sensing Technology. For more information go here .