Timing and synchronization options to support 5G

Network operators are deploying new 5G networks to address mobile broadband growth and enable new business opportunities. In doing so, they are upgrading the equipment at cell sites and finalizing their deployment plans, which involves ensuring that synchronization requirements for both frequency and time/phase are met.

Synchronization in many ways is the lifeblood of the radio access network (RAN), which cannot operate without it. Clock specifications are becoming more stringent for 5G, particularly in fronthaul-connected sites. As a result, there’s a growing need to have a backup synchronization source to one from a Global Navigation Satellite System (GNSS).

Natural phenomenon such as solar flares and weather-related equipment damage can knock out service. Similarly, intentional jamming and spoofing can interfere with the relatively weak GNSS signals picked up by receivers. Network operators cannot fully control these vulnerabilities, so a reliable source of backup to GNSS is required.

Without a backup source, the gNBs would need to maintain proper holdover, which requires expensive oscillators to maintain synchronization until the cell can lock onto the restored GNSS signal. However, relying on built-in oscillators can be problematic because eventually the gNB synchronization would degrade and wander from the primary reference time clock, potentially causing it to interfere with other neighboring cells.

Using the packet transport network as a backup synchronization source

Fortunately, packet transport networks that provide connectivity to cell sites can also distribute highly accurate synchronization data. With 5G gNBs requiring both frequency and time/phase synchronization, the transport nodes can provide synchronization using Synchronous Ethernet (SyncE) and the IEEE 1588v2 Precision Time Protocol (PTP).  The PTP can work with various timing profiles (see Figure 1).

Figure 1. Timing profiles

Use cases for full and partial timing support

The choice of Full Timing Support (FTS) or either type of partial timing support depends on a network operator’s needs.

FTS is suited:

• For applications requiring highly accurate synchronization distribution
• To avoid having GNSS receivers at each site
• As the primary synchronization source for line-of-sight limited cell sights.

Partial Timing Support (PTS) or Assisted Partial Timing Support (APTS) is suited for:

• Small, indoor, last-mile deployments where the network has limited or no timing support
• Bridging between areas of FTS as a cost-effective means to extend the footprint by leveraging third-party networks that lack PTP awareness
• GNSS failure protection serving as a backup to local GNSS receivers.

FTS profile

The FTS profile requires that each transport node participating in the synchronization chain support both SyncE and PTP. This means that each node (router or switch) performs as a Telecom Boundary Clock (T-BC) that receives timing from the Telecom Grand Master (T-GM), which it uses to update its local clock. Each T-BC node, in turn, acts as a PTP master for downstream nodes to deliver accurate time/phase synchronization to the end application.

This profile also allows for the use of Telecom Transparent Clocks (T-TCs), which update the PTP messages with their transit times to remove uncertainty and allow accurate time recovery.

We recommend the FTS profile because it provides the most accurate timing with predictable performance. FTS is also mandatory for fronthaul, stipulated as a requirement in the IEEE 802.1CM TSN for fronthaul standard.

PTS profile

For mobile operators who want to leverage third-party leased networks that do not have PTP-aware nodes installed or transport operators who do not want the expense of replacing all their equipment with PTP-aware equipment, the ITU-T has defined a telecom PTP profile with partial timing support that can operate over existing networks.

In the PTS profile, the use of T-BCs or T-TCs is permitted but not required. Using this profile, T-BCs can be placed at strategic intermediate locations to reduce noise as the timing signal passes through the network. The profile operates over existing switches and routers using unicast IP. This makes upgrades easier than needing to rip and replace all nodes that lack PTP awareness.

However, partial on-path timing support is usually found in older networks and only targets the “coarse” accuracy of ±1.5 µs. The accuracy of a network using PTS is impacted by the number of hops, the per-interface traffic (distribution, load, burstiness) and network reconfigurations (e.g., due to path reconfigurations resulting from failover protection or from SDN reconfiguring paths to optimize resources).

Due to the limitations in synchronization accuracy of this profile, it is typically limited to small, last-mile deployments where the network has limited or no timing support and where accurate network planning and testing is needed to ensure that synchronization performance requirements are met.

APTS profile

Improving on the PTS profile, networks using the APTS profile can provide a backup source for timing from the core T-GM to the edge T-GM. Using APTS eliminates the need to have T-BCs along the synchronization path in an Ethernet network.

At the local site, which could be the cell site, the transport node can receive a GNSS signal to provide the timing to the RAN equipment. While the GNSS signal is locked, there is a calibration of the PTP flow coming through the partial timing network; this calibration calculates the timing error (TE) offset as shown in Figure 2.

The TE offset is used if the local GNSS source is lost, to help improve the holdover time and ultimately minimize the impact of network reconfiguration changes.

Figure 2. Backhaul timing redundancy via APTS network

Nokia IP Anyhaul offers flexible timing options

In the Nokia IP Anyhaul solution, the Nokia 7250 IXR IP routers support the distribution of highly accurate frequency, time and phase synchronization distribution over the transport network. The routers provide a highly accurate source of timing through integrated GNSS receivers (including dual band receivers) and GNSS SFPs.

The routers also provide the flexibility of supporting various timing profiles, including FTS, PTS and APTS, for delivery of timing through packet networks depending on the application requirements. This flexibility includes the option of using multiple profiles through a profile interworking function (IWF).

The IWF is useful when the radio equipment is configured to work with the FTS profile but an operator subsequently chooses to obtain synchronization from a network using APTS. In such cases, the routers can perform an IWF (as shown in the bottom of Figure 2) where they support the APTS profile by default as the primary profile and support FTS as an alternate profile on different ports. In this way, the routers provide synchronization to those radios configured for FTS without needing to reconfigure the radio equipment.

Figure 3 shows some of the possible configurations using timing protocols to back up GNSS. We recommend a combination of GNSS and FTS because this option results in the highest synchronization accuracy. This option helps to make the network future ready because having high-quality boundary clocks in each node can support all applications, including highly stringent fronthaul services.

Figure 3. Nokia IP Anyhaul routers using various timing profiles as backup to GNSS

Recognizing that operators will also be faced with the need to bridge networks through non-PTP aware networks, the PTS and APTS profiles can be deployed with careful planning to ensure performance requirements are met. Operators need to consider several factors, such as the number of hops, different traffic profiles, queuing characteristics and reroute events—because any of these can impact packet delay variation and delay asymmetry, and therefore impact performance.

Increasing accuracy with dual band receivers

The Nokia 7250 IXR portfolio includes IP routers that use dual band GNSS receivers capable of operating using both the L1 and L5 bands (as shown at the top of Figure 3). An advantage of using the L5 band is that it has higher power and operates at a lower frequency than the L1 band. This results in better reception of the signal by receivers compared to using only the L1 band.

In addition, the use of a dual band receiver helps to improve synchronization accuracy by reducing the time error. The largest source of time error in GNSS receivers is caused by signal delay though the ionosphere, which is dependent on space weather and variations in sun activity.

The impact of ionosphere behavior is different for each frequency, and multi-band GNSS receivers can use these differences in delay between the bands to estimate and compensate for the absolute delay to improve accuracy. In this way, multi-band GNSS receivers enable the routers to attain PRTC-B (Primary Reference Time Clock – Class B) performance accuracy to within 40 ns of time error relative to UTC; this is an improvement over single band GNSS receivers, which meet PRTC-A performance accuracy to within 100 ns.

Learn more

Read the synchronization white paper to learn more about how your network can benefit from these capabilities and how the Nokia IP Anyhaul solution delivers high-quality, network-based synchronization.