The objective of synchronization in telecommunications networks is to enable operators to transport digital information intact between network elements and across network and national boundaries. In order to accomplish this goal, all network elements that comprise the global switching and transport network infrastructure must be synchronized (or in some cases phase or time-aligned). Ideally, they'd be synchronized to the same reference, but having a single reference applicable across the globe is clearly impractical. As a "compromise" a certain level of performance has been determined and any frequency source that meets these accuracy and stability requirements is considered a Primary Reference Source (PRS). The output frequency of a PRS is within 1x10-11 of perfection. Consequently, the frequency offset between any two PRS devices is less than 2x10-11.
Distribution of accurate and stable synchronization in telecommunication networks is performed at two levels:
- Inter-office: Distribution of synchronization between offices. If an office does not have a local PRS, it can accept signals derived from traffic carrying transmission systems connected to a second office as a timing reference. Ideally, this second office has a PRS or, sub-optimally, has derived its timing reference from a PRS-traceable source linked to a third office.
- Intra-office: Distribution of synchronization within an office. Equipment that achieves this may be called a BITS (Building Integrated Timing Supply) clock, a TSG (Timing Signal Generator) or an SSU (Synchronization Supply Unit).
Network elements that perform the switching and transport functions in the network have internal frequency sources or oscillators that are not inherently as accurate as the frequency provided by a Primary Reference Source. In order to synchronize the flow of traffic between all these elements, the internal frequency sources or oscillators in the network elements must be locked (traceable) to the accuracy level of the distributed frequency provided by a PRS. The network elements are designed to phase or frequency-lock the internal frequency sources or oscillators to that of the external signal applied which, if the sync network is properly designed, will be traceable to a PRS. A synchronization failure occurs when the network element loses phase or frequency-lock to the externally applied signal or when traceability back to a PRS is interrupted so the network element is performing at the accuracy level of its internal frequency source or oscillator, an event referred to as "holdover". Such a failure is not immediately catastrophic since the holdover performance of the oscillator may keep the quality of sync reasonably good for a period of time to allow for craft intervention. Message switches typically have reasonable holdover mechanisms. However, terminal equipment like D4 Channel Banks and some SS7 link terminating equipment do not.
|
