Cell-based clock recovery device

Details Cell-based clock recovery device Cell-based clock recovery device

1997-02-19

1998-12-01

Vu, Huy D.

Multiplex communications

Data flow congestion prevention or control

Flow control of data transmission through a network

370395, 370412, 370503, H04J 314

Patent

active

058448918

DESCRIPTION:

BRIEF SUMMARY
FIELD OF THE INVENTION

This invention relates to digital transmission systems, and more particularly to a cell-based clock recovery (CBCR) device for providing a clock recovery function in the receiving entity of a system to implement adaptation of constant bit-rate (CBR) services over an asynchronous transfer mode (ATM) or ATM-like network.


DESCRIPTION OF PRIOR ART

Asynchronous Transfer Mode (ATM)) is a high-speed digital communications protocol for which the basic functional unit is a fixed-length 424-bit (53-byte) cell. Five bytes of each cell are allocated for routing and control, and the remaining 48 bytes are used for data transport. On entry to an ATM-based network, information is loaded into cells in accordance with standardized formatting protocols called adaptation protocols. Once the cell has being filled, it is transported through the network as soon as possible. Buffers are used throughout ATM-based networks to deal with congestion, i.e., cases where more than one cell is ready for transport over a given communications link.
As suggested by the name, constant bit rate (CBR) services transmit data bits at a nominally constant rate. More specifically, transmitting entities for CBR services use a reference clock to time the delivery of data bits. Receiving entities must access the same clock, either directly or indirectly, to retrieve the data.
Reference clocks used in CBR services must operate within standardized specifications for mean operating frequency, maximum jitter and maximum wander. Jitter and wander are constraints on high-frequency variability and low-frequency variability about the mean operating frequency, respectively. Jitter and wander are generally measured in terms of the offset of clock pulses from where they would be if the clock were consistently operating at precisely its mean operating frequency. Jitter is short-term deviation in the pulse center-points from where they would be if the clock was fixed at its mean frequency. Wander is the equivalent long term variation. Jitter is measured in the order of Hertz, whereas wander is measured in the order of hours or days.
The following specifications, drawn from the ANSI specification for T1 service, provide an example of requirements for CBR clocks. Here a UI is a unit interval or clock period, which for T1 is 648 ns: UI peak-to-peak, and jitter in the frequency band between 8 Khz and 40 Khz must be less than 0.07 UI peak-to-peak. must be less than 28 UI peak-to-peak over any 24-hour period.
Clock recovery is a process by which entities within a communications network gain access to a reference clock when needed. There are two basic approaches to clock recovery. The first approach entails applying relatively simple techniques to a clock signal that is separately transmitted to the receiving entity. The other approach is to extract the clock from an analysis of the periodicity of the received data signal. While the second approach is generally more complex to implement and more prone to error, it obviates the need for transmission of a separate clock signal. Cell-based clock recovery is an example of the second approach.
When CBR services are transported over an ATM network, a recommended basis for controlling the output clock in the receiver is to monitor the buffer Services: Functionality and Specification, Draft T1S1/92-570, Nov. 13, 1992!. If the receiver's output clock is slower than the sender's input clock, then the buffer fill-level will tend to increase with time. Conversely, if the receiver's output clock is faster than the sender's input clock, then the buffer fill-level will tend to decrease with time. The output clock can thus be adjusted based on trends in the buffer fill-level. In this arrangement, it is not necessary for both the sender and the receiver to have access to a common network clock. The approach can thus be used in more situations than the alternative recommended clock recovery method, i.e., the synchronous residual time stamp (SRTS) method. Such a method is described, for example, in PCT applicati

REFERENCES:
patent: 5027351 (1991-06-01), De Prycker et al.
patent: 5486784 (1996-01-01), Eriksson
Singh et al., "Jitter and Clock recovery for periodic traffic in broadband packet networks," IEEE Transactions on Communications, pp. 2189-2196, May 1994.
"Rate based end-to-end clock synchronization," IBM Technical Disclosure Bulletin, vol. 37, No. 04B, Apr. 1994.

Affiliated with

Also associated with

Rating

Cell-based clock recovery device does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Cell-based clock recovery device, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Cell-based clock recovery device will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2401110