5G for Industrial Automation
What is 5G and Why Should an Industrial Automation Engineer Care?
Introduction
With the major manufactures releasing new 5G compatible devices and network operators rolling out 5G networks, you could be forgiven for thinking that this latest generation of cellular network is nothing more than a means to stream more video content and download games faster.
While some network operators are using 5G as an alternative to fixed lines for home broadband services in built up areas the main promise of this latest generation of cellular network is as an enabler of machine communications. The widespread deployment of 5G technologies promises ubiquitous connectivity which will enable many IoT, IIoT, Smart Cities and Industry 4.0 applications.
History of Cellular Networks
Cellular mobile networks were first launched in 1979 by Nippon Telegraph and Telephone (NTT) in Japan. This first-generation network only supported voice calls.
Approximately each decade after this a new generation of cellular network technology has been rolled out with each generation incorporating more features.
The fifth generation of networks (5G) is currently being deployed by networks operators (and 6G networks are under development, but that is a subject for another post).
Figure 1 Source: https://www.ericsson.com/en/5g/what-is-5g
What is 5G?
5G is the 5th generation of mobile network and is a significant break from the previous 4G networks as it includes the so called ‘New Radio’ (NR) technology.
3GPP
The 5G standards are developed by the 3rd Generation Partnership Project (3GPP) which comprises seven international telecommunications standards developing organisations (ARIB, ATIS, CCSA, ETSI, TSDSI, TTA, TTC).
5G technology standards have been planned to be developed as three releases (15, 16 and 17). Current networks are currently deploying the release 15.
Many of the features required for industrial applications are planned to be included in releases 16 and 17 only.
Release Status
What is the Difference between 4 and 5G?
Speed (Bandwidth)
5G is 10 to 50 times faster than 4G.
Latency
Latency is reduced from 200 mS to around 1 mS.
Cost
The expectation is that costs for IoT devices will be lower under 5G while handset costs will be unlikely to change.
Power Consumption
Some estimates of devices power consumption reductions of 90% are expected with 5G networks.
Connectivity
The devices supported per km2 increases from 10k-100k to 1 million with 5G.
Location
The 5G standards will support positioning services down to centimetre accuracy for indoor private networks and 0.1m for combined public 5G and Global Navigation Satellite Systems (GNSS).
Reliability
The 5G standard implements multiple features to support high reliability.
Range (& Penetration)
The range and penetration of 5G signals is significantly less than for 4G networks (operators are compensating for this by building many more base stations)
So, What Makes 5G so Different?
Frequency
5G networks operate at a much higher frequency than 4G. The exact frequency bands used vary depending on your region and carrier with 4G typically below 6 GHz and 5G operating at 30 GHz and above (up to 300 GHz).
Operating a radio at a higher frequency has several advantages,
- Higher bandwidth (i.e., more data can be transferred). Refer to the following description of why this is the case.
- The signals are more directional.
- Antennas are smaller (allowing for more devices to be served from a single base station).
All engineering decisions include trade-offs and the disadvantages of operating at a higher frequency are,
- 5G signals have less range and penetration which requires more base stations to be constructed.
- Base stations include more equipment leading to higher base station power (said to be over 3 times greater)
Network slicing
5G networks run multiple virtual networks on the same physical hardware (the same principle that is used in cloud computing where multiple virtual machines run on the same physical host).
Sperate network ‘slices’ can be defined with characteristics to suite the use case. This might include a highly reliable but low bandwidth IoT solution which has different network requirements to a low latency, high bandwidth network to support autonomous vehicles.
Massive MIMO
Multiple Input Multiple Output (MIMO) technology makes use of large numbers of antennas and advanced algorithms to expand the capacity of the network.
Some of the features that MIMO implements are,
- Spatial diversity. Sending the same signal along multiple paths to create redundancy and prevent interruption.
- Beamforming. Directing targeted signals towards recipients to increasing data rates for all users.
- Multi-user. Directing signals to multiple users simultaneously.
Network Management
5G networks include advanced algorithms that tailor network speeds to devices dependent on the traffic type.
Support for Devices
The 5G standards include (or will include in release 16 and 17) a number of features to support IoT devices which include,
- Sleep mode to manage battery operated devices (5G Efficiency)
- Ultra Reliable Low Latency Communications (URLLC)
- Non-public networks
- Edge computing
Industry Applications
From the discussion above you can see that while 5G networks are being rolled out and hold great promise to support industrial automation applications, most of the features required by industry are contained in the 3PGG release 16 and 17 specifications that are unlikely to be widely deployed for the next five to six years.
Thales, have suggested the following schedule in relation to the viability of 5G use cases,
- Fixed wireless access (from 2018–2019 onwards)
- Enhanced mobile broadband with 4G fall-back (from 2019–2020–2021)
- Massive M2M / IoT (from 2021–2022)
- Ultra low-latency IoT critical communications (from 2024–2025)
3GPP: 5G for Industry 4.0
5G supports Industry 4.0 with,
- Integration with the IEEE 802 LAN standards
- Support for time sensitive communications
- Support for non-public networks
Proposed Use Cases
Network equipment suppliers (Nokia, Ericsson, Siemens, etc.) and network operators (Telstra, Optus, Vodaphone, etc.) have nominated specific industries which would benefit from 5G deployment to support Digitisation, Digitalisation and Digital Transformation.
- Airports
- Manufacturing
- Mining
- Ports
- Utilities
- Digital Twins
- Logistics
- Public Safety
Private Networks
Private 5G networks are support and are expected to be widely using in Industry 4.0 applications. For organisations with facilities requiring private networks and a field force requiring use of a public network, private and slices of a public network can be combined to provide an integrated solution.
Reality Check
While it is clear that the enhanced capabilities of 5G networks will support new use cases that previously have not been viable in relation to Smart Cities, Logistics, Autonomous Vehicles etc. the use case in relation to manufacturing is not so clear.
Is 5G Meeting a need in Manufacturing
Below are the three key benefits of 5G technologies to manufacturing that are being promoted:
Lower cost of wired vs wireless communications:
In a typical manufacturing plant this assertion is unlikely to be valid.
In the case of attaching a 5G base station to an industrial automation control network or a fieldbus network is hardly likely to be cost effective compared to the fibre optic backbone network which exists in most facilities.
In the case of 5G instrument and actuators, battery life will be a serious limitation and the cost of battery replacement will be significant.
Communications will be pervasive:
Current manufacturing facilities are not suffering from lack of communications options and bandwidth. Most modern industrial facilities will have a fibre optic backbone network typically operating in the Gigabit range.
Ability to migrate control functions to the cloud:
It is proposed to move control functions currently executing in programable logic controllers (PLCs) located at the industrial facility to the cloud.
From a resilience perspective this would not be recommended as it is typical to keep the control function close to the actuator. In addition, this would not be permissible for any safety related function.
Ability to collect more data, analyse with ML and AI to better control and maintain:
The implication here is that industrial facilities are suffering from the inability to collect data when in fact the converse is true. Industrial facilities through their SCADA and DCS systems are currently collecting 10s or 100s of thousands of data points per second, most of which is currently not analysed.
This problem has only been made worse with the widespread deployment of ‘smart’ instrumentation since the early 90s where each instrument, apart from the measurement value can also provide 10s to 100s of points of diagnostic and configuration data.
Currently industrial facilities are drowning in data and really need better data models and analysis tools to help them get value from the existing data rather than plan to gather even more.
Cybersecurity
Shifting from wired to wireless communications is at the very least going to make users more susceptible to cybersecurity threats even if this is only a denial-of-service (DoS) attack.
Resiliance
Living in a land of bushfire, flood, and cyclones the resilience of any proposed solution is always a consideration.
Even without 5G networks, many businesses have become highly reliant on cloud-based solutions. Have you considered how your business would cope without access to could based services for a day, week, or month? How about loss of access to GPS?
Resilience is an issue to consider and it needs to align with your business model and risk appetite.
Data Models
As mentioned previously, industrial automation is currently lacking in data models and useful high-level abstractions. The ISA and OASIS have attempted to develop these in the past with limited success. Integrating and analysing large volumes of process data will rely on the creation of useful data models.
Competing IoT Networks
There are a number of low powered wide area network technologies that are currently in various stages of deployment including LoRaWAN, SigFox and NB-IoT.
NB-IoT has been incorporated into the 5G standard.
While 5G will be widely deployed, it is unclear if it will be the dominant IoT and Industry 4.0 network.
Review
We have now reviewed what 5G is, what makes it different from 4G, some potential applications and challenges in industry.
To return to the original question, why should an automation engineer care about 5G? I suggest that the answer is in two parts:
5G Technology May Support Applications that are Appropriate to Your Business
Pervasive Communications:
5G networks will provide pervasive, high bandwidth communications which will be able to support applications which were not previously viable (Asset tracking, Augmented Reality, Wearable devices etc.)
Access to Pervasive Computing:
The high bandwidth of 5G networks will allow low power devices to access compute power (via the cloud) which was not previously available.
There is a lot of hype around 5G — it might be pushed onto you by others
Given that there is a lot of hype around 5G, it might just catch the attention of the CEO, CTO, CDO etc. and you may well find yourself being pressured to adopt a 5G solution even if it is not a good match for your business.
Much of the current promotional material implies that 5G will be a good fit for everyone and produce incredible operational benefits.
What to do?
Determine the applicability of 5G to your industry
Examine the services that 5G will offer, talk to you service and equipment providers and determine if there are valid use cases for you.
Do not forget to consider that 5G may well support activity sets that are not currently present in your organisation but could be beneficial.
Incorporate 5G (or not) into you Automation Stratergy
Update your automation strategy to include 5G if it is beneficial and if not, clearly document why it has been excluded.
Your automation strategy should also identify you position in relation to new technology adoption. Will you be an innovator and early adopter, a fast follower or wait for the technology to become mainstream?
Note: While it is expected that 5G will be fully deployed (including release 16 & 17) by the middle of the decade, this does not mean that it will have become widely adopted in industrial automation.
References
https://www.thalesgroup.com/en/markets/digital-identity-and-security/mobile/inspired/5G
https://www.qualcomm.com/news/onq/2019/06/20/how-5g-massive-mimo-transforms-your-mobile-experiences
https://www.ericsson.com/en/blog/2020/12/5g-positioning--what-you-need-to-know
https://www.allaboutcircuits.com/industry-articles/how-will-5g-networks-improve-location-awareness/
O’Connell, E., Moore, D., Newe, T., 2020. Challenges Associated with Implementing 5G in Manufacturing. Telecom 1, 48–67.. doi:10.3390/telecom1010005
Gu, F., Niu, J., Jiang, L., Liu, X., Atiquzzaman, M., 2020. Survey of the low power wide area network technologies. Journal of Network and Computer Applications 149, 102459.. doi:10.1016/j.jnca.2019.102459
Read Further
Thanks for reading, hope you enjoyed this article.
To explore further,
- Subscribe to email notifications
- Click on the ‘follow’ button at the top of the article
- For all things Industry 4.0, check out my Industrial Digital Transformation & Industry 4.0 publication
- Feel free to join my network on LinkedIn (remember to mention that you have read the article)
To support medium authors, consider a subscription.
Remember to click on the subscribe and follow button,
