Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

 
 

Adaptive Clocking Overview for CTP Bundles

The goal of adaptive clocking is to prevent buffer anomalies by adjusting the clocks so that they are the same at each end of the network. If the clocks are not the same at each end of the network, the data rate entering and exiting buffers will not be the same, which causes a buffer underflow or overflow.

Adaptive clocking works by gathering information about packets arriving from the IP network and using that information to determine whether adjustments need to be made to the local clock to maintain frequency lock with the remote end. This process is called adaptive time domain processing (ATDP). ATDP provides rapid convergence to the correct clock, and does not vary due to changes in the average jitter buffer fill. As a result, a circuit continuously operates without a buffer recenter, even when clock references are not used.

There are two types of adaptive clocking:

  • Adaptive clocking with internal clock—Recovers the clock from the user equipment connected to the remote CTP device and uses it to generate both transmit and receive timing. All clocking is performed by the DDS, which is initially configured to be locked to the local system clock. When packets begin to flow between the CTP devices, the adaptive clock begins time domain analysis of the packets that arrive from the remote CTP device. Based on this analysis, adjustments are made to the DDS clock to approximate the frequency of the clock used to create network-bound packets on the remote CTP. In this way, the local CTP port can maintain long-term frequency lock with the remote CTP and pass this clock to the locally connected user equipment.

  • Adaptive clocking with external TX clock—Data received from the local user equipment that is bound to the IP network is clocked using the CTP external user clock (the transmit timing clock). Data received from the remote CTP device and bound for the interface is adaptively clocked with the recovered clock from the user equipment connected to the remote CTP device. This configuration allows for independent adaptive configuration in each direction. With this method, the user equipment can send packets into the network with their local clock, and the remote end CTP devices adaptively recover this clock. This clocking method is useful when the port speed is high or the cable length between the user equipment and CTP device is large.