Chris Pearson – August 2022
What comes to mind when I mention the word “automation”? Some of you may immediately imagine a row of robotic arms along an assembly line for constructing a car. Others might think of a self-driving vehicle itself, or even a series of electronic transactions that take place in a financial institution when I insert my card in an ATM.
Today, automation is pervasive. We live in a world where more basic processes are relying on less human input than ever. With automation spanning across virtually every known industrial sector, we are seeing the impacts of machine decision-making and action in areas as diverse as entertainment, transportation, manufacturing, public safety, agriculture, natural resource exploration, and so many more.
Over the first few years of 5G network deployments, there was much focus on how 5G networks would work for consumers. However, new connectivity requirements for enterprise and industrial use cases can be very different from those of consumer use cases. Data rate and throughput may serve as the main benchmarks for consumer use cases like internet browsing and video streaming, but the requirements for enterprises and industrial sectors can go much further than these simple kinds of connections – and much further into deeper levels of business process automation. Luckily 5G was designed for these new deeper levels of precision.
In many cases, automated systems require extremely precise timing to ensure consistent operation. Think about the fine camera movement needed to capture a big budget action sequence replete with explosions and car chases, or automated locking mechanisms in the county jail, or precision drilling techniques used by modern rigs in the oil and gas industry. All the interlocking machinery and provisioning of data must happen very fast, must be synchronized, and must be delivered to a degree of extremely high reliability. In some cases, the accuracy requirement for such systems can approach one microsecond (1usec) or faster.
In a pre-5G network world, it has been challenging to be able to deliver data in a way that fulfills some of these new requirements. However, 5G has been designed to fulfill a vision to help digitally transform various enterprises and industries. In general, 5G has been standardized with the objective to deliver up to 10 gigabits per second of data at peak theoretical speeds, with as low as 1 millisecond (msec) of network latency, to up to one million devices per square kilometer – with five nines (99.999%) of uptime reliability. However, 5G-based precision timing systems may push the boundaries of these general specifications even further.
In our latest white paper, “Understanding 5G & Time Critical Services,” 5G Americas explores the future opportunity of precision timing in 5G networks and how they are beginning to tackle some of the most challenging basic requirements of time-critical services. Compared with satellite-based precision that forms the basis for so many timing systems, 5G networks offer a unique capability: providing indoor coverage at micro-second level precision and wireless timing delivery.
For instance, in commercial banking, asset exchanges and high frequency trading typically require 1 usec to 1 msec of timing accuracy, depending on regulatory standards and other factors established by governments. These are often extremely challenging use cases for certain precision timing solutions like Global Navigation Satellite Systems (GNSS) because they are typically indoor deployments. 5G networks could possibly be delivered more successfully indoors using small cell technologies.
Another example would be electrical grids applications with multiple generators that need to be phase-synchronized with each other. This requires very precise timing at each substation. While these outdoor systems typically use GNSS, 5G networks might be able to provide a solid backup solution – thus improving the resiliency of the electrical grid.
One huge area for precision timing involves manufacturing. Modern factories operate many types of machinery that are required to work in complete coordination with the rest of the factory system – and modern networks must take into consideration a host of legacy machinery and systems to allow for interoperation. Historically, precision-timed systems have historically relied on GNSS and others to provide the required timing accuracy, but different timing protocols and sources have been developed with varying degrees of precision. The table below provides a good overview of the typical levels of timing precision based on the timing system.
|Atomic Clock||1 part per1015||An extremely expensive method of generating very high precision time. The root time reference for all other systems is typically an atomic clock|
|NTP||<20ms||Network Time Protocol: Dominant protocol for distributing time over Internet. Not suitable for precision time delivery.|
|PTP||<1us||Precision Time Protocol: Requires end-to-end support at all switches. Used in controlled deployments (e.g., factory, radio access network)|
|GPS||<50 ns||Global Positioning System: Over the air, free and very accurate. Requires outdoor unobstructed antenna.|
|GLONASS||<200 ns||Same as GPS|
|5GS||<1us||5G system may provide timing based on several references; Universal Time Coordinated (UTC), GNSS or Local Time.|
In addition, varying levels of timing precision will also impact the number of user devices which can be successfully synchronized in the system. A user-specific clock synchronicity accuracy level will dictate not only the number of synchronized devices, but also the service area and the kind of scenario in which the precision-timing can be used. As you can see below, the accuracy of the clock will impact the ability of a system to fully synchronize all its dependent parts.
|User-specific clock synchronicity accuracy level||Number of devices in one communication group for clock synchronisation||5GS synchronicity budget requirement (note 1)||Service area||Scenario|
|1||up to 300 UEs||≤ 900 ns||≤ 100 m x 100 m||Motion control Control-to-control communication for industrial controller|
|2||up to 300 UEs||≤ 900 ns||≤ 1,000 m x 100 m||Control-to-control communication for industrial controller|
|3||up to 10 UEs||< 10 µs||≤ 2,500 m2||High data rate video streaming|
|3a||up to 100 UEs||< 1 µs||≤ 10 km2||AVPROD synchronisation and packet timing|
|4||up to 100 UEs||< 1 µs||< 20 km2||Smart Grid: synchronicity between PMUs|
|4a||up to 100 UEs||< 250 ns to 1 µs||< 20 km²||Smart Grid: IEC 61850-9-2 Sampled Values|
|4b||up to 100 UEs||<10-20 µs||< 20 km²||Smart Grid: IEC 61850-9-2 Sampled Values – Power system protection in digital substation|
|4c||54/km² (note 2) 78/km2 (note 3)||< 10 µs||several km²||Smart Grid: Intelligent Distributed Feeder Automation|
|4d||up to 100 UEs||<1 ms||< 20 km²||Smart Grid: IEC 61850-9-2 Sampled Values – Event reporting and Disturbance recording|
|5||up to 10 UEs||< 50 µs||400 km||Telesurgery and telediagnosis|
|NOTE 1: The clock synchronicity requirement refers to the clock synchronicity budget for the 5G system, as described in Clause 5.6.1.
NOTE 2: When the distributed terminals are deployed along overhead line, about 54 terminals will be distributed along overhead lines in one square kilometre. The resulting power load density is 20 MW/km2.
NOTE 3: When the distributed terminals are deployed in power distribution cabinets, there are about 78 terminals in one square kilometre. The resulting power load density is 20 MW/km2
With all these complexities, 5G networks are now being enhanced to be able to address many new verticals and industrial mission-critical use cases. Beginning in Third Generation Partnership Project (3GPP) Release- 16 and further enhanced in Release-17, technologies in Ultra-Reliable Low Latency Communication (URLLC) and Time Sensitive Network (TSN) standards have been created to specifically address these industrial vertical use case needs.
Momentum in finding industrial and automated solutions is pervasive as the Institute of Electrical and Electronics Engineers Time-Sensitive Networking (IEEE 802.1 TSN) standard provides a potential converged networking technology for factories to enable deterministic and low-latency communication in delay-sensitive industrial applications. Due to its ability to provide flexible wireless data access and full connectivity for a smart factory, IEEE 802.1 TSN is looking like an attractive addition to Precision Time Protocol (PTP) or its variants, such as IEEE 802.1AS, which requires wired ethernet connectivity.
Overall, we are just beginning to tap into the capabilities that 5G networks have to offer when it comes to precision timing and Timing as a Service (TaaS). Over the coming years, as business requirements are fleshed out and 5G-based timing systems integrated into processes, we could see many new doors opened with automation. We are only on the cusp of some great things. Let’s see what the future will bring.