Synchronization (sync) and timing are essential to telecommunications networks to ensure optimal performance and prevent packet loss, dropped frames and degradation of quality of experience that will affect end-user services. Most critical telecom applications require precise time and frequency and synchronization to operate properly such as VoIP, video streaming, TDM (time division multiplex) services, voice switching, mobile services, and any LBS (location based services, with E911 being the most important). Holdover is the time period required to keep the network sync-stabilized when the source of sync is disrupted or temporarily unavailable.
In traditional digital telecommunications networks (TDM), sync was maintained by employing two types of synchronization elements, Primary Reference Clocks and distribution Clocks, over a physical circuit. The PRC or PRS (using either Cesium or GPS) provides the reference signal for the synchronization of other clocks within a network, or section of a network. Distribution clocks (called BITS, SSU or SASE depending on configuration) select one of the external synchronization links coming into a station as the active synchronization reference. These two types of clocks attenuate jitter and wander, maintain operation in holdover mode, and provide synchronization outputs to all Central Office network elements.
As networks transition from TDM to packet-based next generation networks, choosing a sync technology becomes more challenging, because packet-based networks do not deliver synchronization naturally as the TDM network elements did. So synchronization (and QoS, quality of service) must be engineered into the packet backhaul. Precise synchronization is especially critical in mobile networks for the successful call signal handoff and proper transmission between base stations, as well as for the transport of real-time services. If individual base stations drift outside the specified frequencies, mobile handoff performance decays, calls interfere, and calls cannot be made, resulting in high dropped-call rates, impaired data services, and, ultimately, lost customers. In the event that timing or synchronization reference is temporarily lost, a network's ability to maintain time or "holdover" becomes critical to ensure optimal network performance.
Why is Holdover Important?
Holdover is critical to ensuring service continuity as well as keeping the operator's OPEX to a minimum (e.g. elimination of emergency truck rolls to the BTS site). In certain geographical areas where GPS signals are received only intermittently, holdover is crucial for the operation of base stations. Holdover technologies are also necessary to maintain sync during GPS outages caused by external events, such as in 2007 in San Diego when the US Air Force's wide-scale denial of GPS caused a major GPS outage. Criminals can also jam GPS signals locally, given the commercial availability of GPS jammers, and environmental factors such as sun spots also contribute to GPS disruptions. Out-of-sync base stations dramatically increase the operator's OPEX. Further, if telecommunications regulations such as E911 requirements are violated, the operator may incur additional expenses of reporting outages and carrying out audits.
Holdover Requirements and Technologies
Holdover is achieved by equipping base stations transceivers with oscillators that temporarily "hold over" sync signals. Holdover capability can range from several hours to several days depending on the quality of the BTS's oscillator. Holdover requirements are not standard; they vary depending on the type, complexity, and operator' requirements. LTE TDD networks have stringent timing requirement, + 1.5 microseconds. (See Network Frequency and Time Requirements table).
Location based services (LBS) and E911 impose exacting sync requirements in order to accurately locate the handset by triangulating from base stations; this imposes extremely precise sync requirements (up to 0.2 microseconds).
The most commonly used holdover source is an Oven Controlled Crystal Oscillator (OCXO), which can ensure 8 microseconds of holdover from 8 to 24 hours, depending on its performance level (different grades of OCXO deliver varying holdover performance). The OCXO performance can be enhanced (with software and processes) to provide either longer holdover or more precise holdover. Rubidium oscillators deliver a much higher level of performance, -i.e. holdover of up to 7 days-- or more precise holdover (they can deliver holdover of 2.5 microseconds for more than 24 hrs, or 8-microsecond holdover for up to 7 days. (See Sample Holdover Module Performance graph).
Holdover Technology of the Future
A better solution that will further extend the holdover range of OCXO and use the newer packet network transport systems is currently in the works. Sync modules extending the holdover of OCXOs in base stations by combining Sync-E, IEEE-1588, and GPS are currently being designed and tested for deployment. A simple combination of 1588, GPS and syncE is not enough. Sophisticated optimization and monitoring algorithms are used to combine GPS (primary) and other sync signals to produce the 'best of all worlds' sync output. PTP and GPS can deliver the time (phase sync) and frequency sync while syncE and the OCXO provide frequency output (extension of 'SETS' concepts). In this solution, multiple technologies aid one another to achieve optimum holdover.
Down the road, a better approach is a sync module that combines PTP packets, other sync inputs, and GPS with a rubidium oscillator using an algorithm that features intelligent selection of inputs, optimal use of clock input accuracy, and produces the best possible clock signal output. In this scenario, sync technologies are used to reinforce each other, and requirements are met with multi-standard clocks. Multi-standard clocks are desirable because some sync technologies may not work in certain deployment instances, and in this situation, multi-standard clocks ease operational pressures, and give a simple, universally deployable product.
While some base stations are moving to integrate all the functions onto their single board implementations, this approach has several advantages. Different performance profiles (for sync) can be implemented on the same system, using / activating different options on a timing module. These timing modules can be characterized separately from the main base station design, and therefore enable a more streamlined production schedule. Symmetricom's trademark BesTime algorithm combined with its best-in class rubidium oscillators (as well as its expertise in selecting and using optimally design OCXOs) may provide a promising solution for the next gen sync module.
Base stations are being optimized for performance and price in that order. It is essential that a mission critical function, such as sync, be properly implemented, since otherwise the performance of the base station is negatively impacted, and may cause increase the operator's operational expenses and decrease the quality of their network performance and service delivery, which in turn can result in customer churn.
Selecting a well thought out sync strategy is the first step. It should be followed up with selecting a robust sync solution that, when integrated into a base station, extends it operational life, and results in a low touch, well performing base station and wireless network.
About the Author:
Rajendra N. Datta (Rajen) is the Director for Symmetricom's Embedded Solutions. Mr. Datta worked in various roles and capacities at AT&T Bell Labs (later Lucent, Bell Labs). Later, he founded and led eBizAutomation Inc., designer of infrastructure systems for e-commerce, as CEO until it was acquired by SpinCircuit (later acquired by Cadence). Mr. Datta holds a Master of Computer Science from the University of Oregon and a Masters in Management of Technology from the Wharton School of Business and Moore School of Eng, University of Pennsylvania.