When properly designed, timing networks are resilient and robust. When any activity is planned, step one is what I call the “information collection stage.” It is important to understand the current state of the timing system prior to undertaking any activities that may negatively impact supported services. Steps associated with the information collection process include retrieving office records that indicate fuse panel and BDFB power assignments. Determining how the Building Integrated Timing Supply (BITS) clock is referenced and retrieving its current alarm status can be extremely useful. Cabling connected to a BITS clock should have cable tags indicating the location of the far end equipment to which it’s providing timing.

The key point is this: knowledge is power. The more you know about the timing system that you are about to work on, the better. Having established a well thought out plan may be the one thing that keeps you from making a tough situation worse. An understanding of the network’s current configuration and alarm status will allow you to develop the appropriate procedure. All of the aforementioned information can be obtained without impacting the system’s operation.  

SyncCare offers peace of mind for network alarms and more.

Takeaways

• DO “Keep your hands in your pockets.” 
• DON’T randomly unplug and swap out cards.That may cause the system to fail. When I was first trained in the operation and maintenance of a BITS clock, keeping my hands in my pocket was the advice I received from a very wise Engineer. He explained that the system is designed for redundancy. Even though the system may be reporting alarms, there may be NO impact to service due to its resilient design.
   
• DO “Trust But Verify”
• DON’T EVER connect an input reference to a BITS clock without first confirming its viability.When operating as designed, ALL BITS clocks should be traceable to a Stratum 1 reference. GPS Receivers and Cesium Atomic Clocks meet this criteria. In some cases, traffic bearing DS1 signals or OC-N Derived DS1 signals from SONET nodes may also be configured to perform at a Stratum 1 level. The quote “trust but verify” comes to mind. Just because you have a DS1 signal properly framed and error free, does NOT mean it meets the stringent requirements to reference a BITS clock.

• DO have a plan – always!
• DON’T go it alone.If you are on-site and notice alarms in a BITS clock, it is good practice to notify the NOC and make them aware of the current alarms and conditions. It is helpful as well to understand the scope of a timing trouble. The issue may simply be a failed plug-in that has switched to protection and requires replacement causing no impact to service. It may also be a situation where multiple or all network elements in the office are slipping, i.e., experiencing degraded input references. The latter is obviously a more critical situation, a second set of eyes may help you to accurately assess the potential impact to service.

Summary Takeaways

To ensure the synchronization network performs optimally, the best thing we can do is respond to alarms in a timely manner. When a BITS clock loses its input references and goes into holdover, most oscillators are of a very high quality and can continue providing accurate timing to network elements. The problems arise when this condition is left unattended for an extended period of time. Make it a habit to walk by the BITS clock when you’re in an office. Report any anomalies to the NOC and develop a plan to safely restore the system to an alarm free state.  

The Rising Role of Precision Time Protocol (PTP)

A key component to resilient timing for 5G is the security of the GPS input. GPS vulnerability has long been a topic of discussion. While it has been widely acknowledged to be a security concern by the power utility, transportation, and communication industries, we have only recently begun taking action to address this issue. The object of this paper is to discuss the evolution of the timing network starting with the distribution of DS1 and Composite Clock signals for digital transport equipment and ending with the deployment of PTP packet timing in support of 5G wireless service.

Evolution of Telecommunications Timing 

Since its inception, the need for increased bandwidth and capacity in our telecommunications system has been a constant driver of technological advances. In the early 1940’s, AT&T developed the L-carrier system. As our population grew, so did the demand for increased capacity and the improved performance of our communications network. L-carrier systems were frequency division multiplexers. That is, a common transport medium that could be used to support multiple users operating simultaneously at different frequencies. It was an efficient way to increase the capacity of existing infrastructure. In the 1960s, Bell Laboratories developed T-carrier. Rather than frequency-based multiplexing, T-carrier is a form of time division multiplexing (TDM.) As with L-carrier, T-carrier also allows multiple users to share a common transport. Thus, increasing the capacity and efficiency of the transport network. TDM, however, assigns a timeslot rather than a frequency to each channel. To maintain the proper sequencing of the channels, end-end synchronization is required. More on this when we discuss their similarities in 5G’s use of FDD and TDD.  

The 5G Paradigm

Today, the same demands for increased bandwidth, capacity, and performance that caused the development of earlier multiplexing schemes are now driving innovation and technological advances in 5G wireless service. And with it resilient timing for 5G. Two spectrum usage techniques used in 5G are Frequency Division Duplex (FDD) and Time Division Duplex (TDD). There is an interesting analogy between the comparison of FDD to TDD and L-carrier to T-carrier. Like L-carrier, FDD transmits uplink and downlink information at different frequencies. Whereas with TDD, uplink and downlink transmissions are assigned different timeslots in the same spectrum frequencies, not unlike T-carrier. TDD frames include different time periods and timeslots that allows for a more efficient use of the spectrum. TDD requires uplink and downlink transmission signals to be precisely synchronized. Failure to do so will result in poor RF performance causing corrupted data and dropped calls. The evolution from frequency to time comes at a price as it requires greater than a tenfold improvement in network timing accuracy. This can be achieved by deploying multiple and diverse technologies such as GPS, SyncE, and IEEE 1588.  

Resilience in Timing 

Now that we have established the key role that highly accurate and precise timing plays in 5G networks, we can discuss ways to increase its resiliency. As previously mentioned, GPS vulnerability is a major concern for our nation’s power utility, transportation, and communication industries. Executive Order 13905 titled Strengthening National Resilience Through Responsible Use of Positioning, Navigation, and Timing Services states that our national and economic security is hugely dependent on our critical infrastructure. GNSS is widely used globally as a timing source. Concerns for regional or global GNSS outages have caused enterprises to pursue ways to protect their networks from reliability issues and potential security risks. Because of its low cost and simplicity in deployment, GNSS has been the primary solution for accurate time distribution. The downside to this solution is the operational expense incurred when GNSS failures occur. Dispatching personnel to locations to troubleshoot and repair equipment is costly from both a service and an economic perspective. GNSS is ubiquitous and therefore it represents a single point of failure that mandates a more secure approach to the distribution of accurate timing. Standards have been set describing how accurate time can be transported over a packet network. These standards define profiles for the operation of precision time protocol (PTP) and Synchronous Ethernet (SyncE). IEEE 1588-2008 is the standard for timekeeping. 1588 Packet Timing Protocol is a series of standards developed specifically for the transport of time over packet networks. The Primary Reference Time Clock (PRTC) is the new generation of time distribution system that is widely deployed by wireless service providers. Using PRTC systems, packet time distribution is a way to provide timing to a mobile access point as a back-up to locally provided GNSS time delivery.  

Resilient Timing for 5G – Technological Solutions  

PRTC systems deployed in the core network use PTP to provide backup timing to local GNSS time distribution. However, since we have not eliminated the dependency of GNSS in the core, our reliance on GNSS remains. The PRTC standard G.8272 includes the requirements for time and phase over a packet network. It describes a clock that delivers <100ns phase and time performance traceable to UTC. With the ever increasing need to improve the performance of emerging mobile technologies and increase protection against GNSS outages, a new standard G.8272.1 called enhanced primary reference clock ePRTC was created. This standard describes a clock that delivers <30ns phase and time performance traceable to UTC. An ePRTC consists of a GPS, an Atomic Clock, and an ePRTC system. Its objective is to produce an independent timescale that is autonomous. It provides time, phase, and frequency calibrated by GNSS. The timescale is then maintained based on the stability of the Atomic Clock. A PRTC receives time directly from GNSS. An ePRTC generates its own timescale locally. The powerful attribute of the ePRTC is that by generating its own independent time it is NOT subject to attacks on GNSS such as jamming or spoofing. Per the standard, once in holdover due to the loss of GNSS, the ePRTC output can increase from 30 ns to 100 ns over a 14-day period. Therefore, subtending PRTC systems can meet their accuracy budget of 100ns when referenced to the ePRTC.  

Outlook  

Although GPS has served us well and is likely to remain an integral component of the timing network, addressing its susceptibility to both intentional and unintentional service interruptions is essential. GPS dependent PRTC systems provide timing over packet networks that can be used to back up GPS. With the advent of ePRTC, GPS is used to calibrate the system, but it then creates a timescale based on the stability of the Atomic Clock. Downstream PRTC systems can take advantage of the improved accuracy created by the ePRTC along with the benefit of a 14-day holdover in the event of a GNSS outage. The next step in establishing a highly accurate and secure time source is the deployment of virtual PRTC (vPRTC) which is a node with a high-performance boundary clock that meets the <100 ns requirements of PRTC without requiring a local GPS antenna system at the site.  A “virtual PRTC” network offers resilient, extremely stable, sub-100 ns time to any point on a DWDM network. The High-Performance Boundary Clocks create a vPRTC node that saves the operators time, expense, and the complexity associated with deploying GPS at all sites. That, along with the added protection and security that ePRTC and vPRTC provides, makes it a critical design consideration that will resolve the present and future concerns surrounding GPS vulnerabilities.  

Conclusion   

The more things change, the more they stay the same. While the evolution of the timing network has gone through many different phases, the drivers for the innovations remain the same. Our customers require and expect continual improvements in the services we provide. Governments and industries will continually demand more reliable, resilient, and secure networks. Fortunately, ePRTC and vPRTC technology enable service providers to meet these requirements in a secure and cost-effective manner, providing robust and resilient timing to critical infrastructures.