Jitter Requirements
A
jitter requirement is a
performance
requirement that specifies a maximum amount of the
performance
quality subfactor
jitter.
The typical objectives of a jitter requirement are to:
- Ensure (for all practical purposes) that:
- An individual event happens when it should.
- Periodic events happen when they should.
- Jitter does not grow large enough to cause failures.
Jitter requirements are typically specified in terms of the
following measurements:
- The maximum acceptable time interval.
- The maximum acceptable standard deviation in the time
distribution.
- The maximum acceptable variability in time.
The following are typical examples of jitter
requirements:
- “The application shall display temperature sensor
readings every 5 seconds plus or minus 1 second.”
- “The application shall read the pressure sensor
every 100 miliseconds with a standard deviation of 30
miliseconds.”
The following guidelines have been found to be useful when
producing jitter requirements:
- The scope of a jitter requirement is typically:
but may also include:
- Jitter is an error in time when one or more events occur
due to varying time delays. Jitter can be important because
the existance of jitter means that events that should happen
at an exact time do not and that events that should be
periodic do not occur on exactly regular intervals. If jitter
becomes too large, then failures can occur.
- Jitter requirements are typically only required for
real-time systems.
- Jitter requirements are different from and not to be
confused with the
architectural mechanisms that may be used to
implement them: