The delivery of synchronization to next-generation base stations will rely on PTP grandmaster clocks deployed in the network. Sync packets flow from the grandmaster clock to the slave clocks in these base stations. Mobile base stations—including 3G, LTE, and LTE-Advanced—all require frequency synchronization of 16 ppb to assure high quality of service and avoid dropped calls. In addition, TDD and LTE-Advanced networks also require tight phase synchronization of 1.1 µs.
The loss of physical layer sync has generated a requirement for new base station designs incorporating PTP slave clocks that will meet the 16 ppb requirement using packet technology. Such PTP slaves in the base stations rely on access to a carrier-class PTP grandmaster clock deployed in the mobile access network.
PTP Slave Locking Considerations
The slave clock establishes communication with the grandmaster by requesting a reservation for a synchronization flow, specifying parameters such as message rate and reservation duration. Initially, the slave free-runs at an indeterminate frequency while the acquisition and qualification processes takes place. The process of aligning the frequency and time to the master takes place in the tracking stage. By the time the locked stage is reached, the slave is well within the application limits and able to maintain the target level of performance over the long term.
Network deployment and reference network test models have been established to assist carrier engineering staff develop specific deployment rules for IEEE 1588-2008 synchronization solutions in their networks. To provide failover protection, there are two main approaches to consider: network-based redundancy and built-in grandmaster hardware redundancy.
- Network-Based Redundancy
In network-based redundancy, two grandmasters are employed. These may be co-located, but they are most often at different geographic locations to provide maximum protection against failure, as shown in the following illustration.
Network-based redundancy protects against network failure in the vicinity of the grandmaster as well as failure of the grandmaster itself, because the second grandmaster is normally connected by a separate access link into a different part of the network. However, there are some significant issues created by switching between two different grandmasters.
- Built-in Grandmaster Hardware Redundancy
A more robust method of providing grandmaster redundancy is to implement a carrier class IEEE 1588-2008 grandmaster clock with built-in hardware redundancy. A fully redundant grandmaster clock employs an active and standby clock, synchronized to redundant primary reference sources (for example, GPS or T1/E1), as shown in the following illustration.
The major advantage of this configuration is that because the active and standby clocks share a common reference and common network location, the PTP slave devices see no synchronization offset during a failover switching scenario. The slave clocks will all remain locked to the redundantly protected grandmaster, and will not be forced to acquire, qualify, and track to a new grandmaster clock with an unknown offset, or connected through a different network path.
In summary, some reliability issues affect both redundancy methods. The failure of a primary reference source (PRS) is an important issue in the design of a grandmaster, particularly if accurate time is required as well as frequency. For frequency applications, backup may be provided using a T1 or E1 frequency reference. In the event of a PRS failure, a T1/E1 input reference that is traceable to a Stratum 1 clock source will keep the PTP time stamps from drifting and is an acceptable redundancy engineering practice. The failure of the slave or its access link will only affect the network element(s) dependent on that slave. The impact on the overall network is low. It is not common practice to provide redundant links or clocks to the base station in cellular networks due to the cost of such protection.
Leave a Reply
You must be logged in to post a comment.