5G Timing Synchronization with Disaggregated Cell Site Gateways

by Will Chang

As the saying goes, “time is the most valuable resource”. It is one of the most important aspects of the transport network that is not often given enough credit for. Good synchronization will allow everything to run smoothly with no complaints, however poor synchronization can result in missed call handovers, capacity reductions and network down time. With the advent of 5G, the requirements for timing synchronization has become even tougher as we move from an asynchronous IP network to an end to end converged IP network. So, let's delve into what are the types of network synchronizations, the differences between 4G and 5G timing requirements, and how we can incorporate flexibility into our network to meet the evolving timing requirements of 5G.

What is Time Synchronization?

Synchronization within a network is where two clocks, or oscillators, are aligned either by their repeating interval (frequency), their phase angles (phase) and/or their time origin (time). Frequency, typically measured in Hertz (Hz) is the repeating interval or the speed of the clocks. Imagine two clocks both pulsating at 1 second intervals. When their repeating intervals are both the same (1 second), they are considered frequency synchronized.

Looking at the two wave forms from our example above, you’ll see why that only synchronizing the frequency is not enough. There is another type of synchronization called phase synchronization. Phase synchronization is when the phase angles are also synchronized together. Typically, when phase is synchronized, so will the repeating intervals would be in sync also.

The third type of synchronization is of course, time synchronization, which is when the two clocks are aligned by using the same time source. The time source or time origin is the master time keeper for that part of the network and is typically sourced through GPS/GNSS.

Why is Time Synchronization Important?

Time synchronization affects the quality of services telecoms can provide to their customers. Poor synchronization within the network would not only be very costly network providers but detrimental to those who are relying on the network to perform high precision and time sensitive functions.

Here are some just some examples of some every day activities that require good time synchronization. Just imagine what would happen to the quality of these services if it’s out of sync.

• Emails – For the accurate time stamping of when emails were sent and received.
• Video Streaming (webinars, movies on demand, conference calls) – Enabling a smooth video and audio quality without lag or jitter.
• Financial Markets – Allowing the stock market to know who executes the trade first.

The Problem with Timing Synchronization

Although a network is synchronized, it can still become out of sync due to a number of factors, called clock drift. Clock drift can be caused by the quality of internal oscillators, the distance between the clocks, and environmental factors such as temperature.

There is always a certain amount of tolerance, or time budget, allowed within the network to compensate for clock drift. However, as new time critical services such as telehealth, autonomous driving, and AR/VR applications are developed, that tolerance becomes more and more narrow. Therefore, the timing requirements of 5G has become drastically more stringent compared to LTE.

Therefore, when planning out the 5G network topology, special consideration is needed for time critical services to travel a path within the network where the timing requirements are met. This means, older parts of the network will need to be upgraded to do so.

4G vs 5G Timing Requirements for the Transport Network

The challenge for meeting 5G timing requirements is that it’s not so clear cut. Of course, in general the time budget throughout the network is reduced significantly for 5G. Here’s a comparison of some of the timing requirements for 4G LTE and 5G transport networks.

Aside from more stringent timing requirements, 5G brings about another challenge by introducing a more complex fronthaul and midhaul into the RAN. By allowing the split of the BBU into separate control and data units, CU and DU respectively, 5G inherently enables a lot of different RAN topologies that could complicate the equipment needed for timing synchronization.

Another complication would be the fact that there will be many brownfield scenarios where a network has a mix of 3G, 4G and 5G (especially 4G, since it’s projected that 4G and 5G will coexist for quite a while). So, we cannot simply upgrade all the equipment to meet 5G’s timing, it has to be able to coexist with older synchronization techniques as well, or at least bring them up to speed.

UfiSpace’s DCSG Meets the Evolving Timing Requirements of 5G

The UfiSpace S9500-22XST is a disaggregated cell site gateway that enables flexible timing synchronization at cell sites with the capability of aggregating 2G/3G/4G and 5G data for the backhaul. It is compatible with both legacy and next-generation BBUs and supports the timing profiles needed to enable each network clock to meet the timing requirements for an evolving transport network involving old and new synchronization types. It can even be set as a grand master clock for establishing a timing source within the transport network.

Timing Profiles:

ITU-T Synchronous Ethernet (SyncE): For Frequency Synchronization through Ethernet
IEEE 1588v2: For Frequency, Phase and Time Synchronization through Ethernet

• G.8265.1: Enables the deployment of PTP-based frequency synchronization
• G.8275.1: For full PTP aware network. Requires the network to provide full timing support with all PTP-aware devices deployed and is suitable when building greenfield networks.
• G.8275.2: For partially PTP aware network. Allows partial timing support to be operated on an existing network with legacy PTP-unaware devices and is suitable for brownfield networks.
• T-TC: Telecom Transparent Clock
• T-BC/OC: Telecom Boundary Clock/Ordinary Clock
• T-GM: Telecom Grand Master Clock

Timing interfaces:

• GNSS: For antennae to set Frequency/Phase/Time Synchronization
• 1PPS: For Phase Synchronization via 1 Pulse Per Second (1PPS)
• 10MHz: For Frequency Synchronization via 10MHz frequency
• TOD: For Phase Synchronization via ASCII string set combining 1PPS and Time of Day signals
• BITS: For Frequency Synchronization distribution
• Ethernet: For Frequency Synchronization using SyncE and PTP