Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
Concept
Clock recovery
Summary
In serial communication of digital data, clock recovery is the process of extracting timing information from a serial data stream itself, allowing the timing of the data in the stream to be accurately determined without separate clock information. It is widely used in data communications; the similar concept used in analog systems like color television is known as carrier recovery.
Serial data is normally sent as a series of pulses with well-defined timing constraints. This presents a problem for the receiving side; if their own local clock is not precisely synchronized with the transmitter, they may sample the signal at the wrong time and thereby decode the signal incorrectly. This can be addressed with extremely accurate and stable clocks, like atomic clocks, but these are expensive and complex. More common low-cost clock systems, like quartz oscillators, are accurate enough for this task over short periods of time, but over a period of minutes or hours the drift in these systems will make timing too inaccurate for most tasks.
Clock recovery addresses this problem by embedding clock information into the data stream, allowing the transmitter's clock timing to be determined. This normally takes the form of short signals inserted into the data that can be easily seen and then used in a phase-locked loop or similar adjustable oscillator to produce a local clock signal that can be used to time the signal in the periods between the clock signals. The advantage of this approach is that a small drift in the transmitter's clock can be compensated as the receiver will always match it, within limits.
The term is most often used to describe digital data transmission, in which case the entire signal is suitable for clock recovery. For instance, in the case of early 300 bps modems, the timing of the signal was recovered from the transitions between the two frequencies used to represent binary 1 and 0. As some data might not have any transitions, a long string of zeros for instance, additional bits are added to the signal, the start and stop bits.
Official source
About this result
This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.