Testing Time-Sensitive Networking Over 5G: High Availability and Reliability

The new 5G system has been designed from the ground up, with many different verticals in mind, not just the enhanced mobile broadband (eMBB) use case. One of the most important areas where 5G brings innovations and enables massive change is Industrial IoT or mission critical applications in industrial environments.

5G introduces key features related to the following three main application scenarios:

1. Increased data rates

2. Lowered latency and increased reliability

3. Increased number of devices serviced

The 5G evolution spans over multiple 3GPP releases, with each release adding or expanding features to address one or more of the application scenarios listed above:

1. Enhanced mobile broadband (eMBB)

2. Ultra-reliable and low-latency communication (URLLC)

3. Massive machine type communications (mMTC)
   

Release 15, which was completed back in 2019 was mainly focused on the eMBB use case. Release 16 on the other hand, completed in mid-2020, was mainly focused on the enhancements to the URLLC (eURLLC) and integration of the 5G system with the TSN system. Release 17, currently in development, will further improve the eURLLC and the mMTC mainly.

Time-sensitive networking (TSN) is a set of IEEE standards that provide deterministic, highly reliable and low-latency communication over wired Ethernet. It is the technology of the future for convergent industrial communications and Industry 4.0. Interworking of the 5G system and the TSN system is a key objective in making the 5G system a better fit for future Industrial Internet of Things (IIoT) applications.

More on how 5G and the TSN systems are integrated and what role they play in the Industry 4.0 transformation can be found in our previous blog, Testing Time-Sensitive Networking Over 5G: Time Synchronization.

High availability in 5G systems increases with the new 3GPP release 16

For mission critical applications in industrial environments, high availability and reliability is a must. In traditional communication systems, to achieve high availability and reliability retransmission-based techniques were used. However, retransmissions introduce significant delays and jitter to the system that the mission critical applications would not be able to tolerate. New techniques are needed to address the challenges of providing highly reliable communication within tight latency and jitter budgets.

In release 16, the 5G system introduces new techniques and improves existing ones to offer high availability. These methods can be classified in two different categories:

1. Packet duplication-based methods. One such method is Packet Data Convergence Protocol (PDCP)

2. Signal combining based methods. One such method is coordinated multi-point transmission using multi-TRP (CoMP)

There is a big increase in availability from 3GPP release 15 to release 16: from five nines (99.999%) to six nines (99.9999%). Although on the first glance it might not look impressive, when converting it into average expected downtime the progress is obvious:

Integration of TSN and 5G systems enables end-to-end seamless redundancy

The TSN toolbox has a perfect tool to address high availability in environments sensitive to latency and jitter. This is the IEEE 802.1CB-2017 standard, or Frame Replication and Elimination for Reliability (FRER). Using FRER, mission critical traffic is replicated and sent over two or multiple disjunct network paths and duplicates are being eliminated at the destination.
To achieve end to end seamless redundancy, disjoint paths between the FRER end points are required over both the 5G and the TSN systems, as shown below.

3GPP defines an interface between the TSN CNC and the 5G control plane to exchange information for the 5G system to properly select UPFs and 5G base stations (gNBs) so that the user plane paths are disjointed.

For increased redundancy it is possible to use devices that are equipped with multiple UEs. The 5G system does not need to implement FRER functionality itself, since the FRER end points are outside of the 5G system. It just needs to provide the mechanisms for multiple disjoint PDU sessions and an interface to request and configure the different paths.

Ensure end-to-end high-availability targets are met

When testing the end-to-end redundancy of the system, there are special considerations to take in two main areas:

1. The interfaces between the TSN CUC and CNC on one hand side and the TSN application functions on the 5G control plane, on the other hand. Since 802.1CB is not implemented inside the 5G network itself, but only carried over disjunct PDU sessions as ethernet payload, the correct configuration of the disjunct paths plays a key role.

2. The application on the end system 802.1CB uses a sequence number to keep track of the duplicated frames. Nonetheless, the protocol itself does not guarantee in-order delivery of the frames. It is up to the application to do any reordering when necessary. This is especially important in a wireless system, where the latency over different paths can change due to for example the movement of a robot or a sensor, like it can be seen in Fig 3. Or different path could be blocked by moving object on the factory floor.

High availability plays a key role for mission-critical Industry 4.0 applications. Even the shortest downtime on the factory floor can cause significant losses. The new advances in 5G networks coupled with the new enhancements to the Time-Sensitive Networks protocol set offer innovative solutions for this problem.

Ready to get started with TSN over 5G? Learn more about testing Time-Sensitive Networking and 5G testing and assurance.