Main Content

Derive Ranges at the Subsystem Level

You can derive range information for individual atomic subsystems and atomic charts. When you derive ranges at the model level, the software takes into account all information in the scope of the model. When you derive ranges at the subsystem level only, the software treats the subsystem as a standalone unit and the derived ranges are based on only the local design range information specified in the subsystem or chart. Therefore, when you derive ranges at the subsystem level, the analysis results might differ from the results of the analysis at the model level.

For example, consider a subsystem that has an input with a design minimum of -10 and a design maximum of 10 that is connected to an input signal with a constant value of 1. When you derive ranges at the model level, the range analysis software uses the constant value 1 as the input. When you derive ranges at the subsystem level, the range analysis software does not take the constant value into account and instead uses [-10..10] as the range.

When to Derive Ranges at the Subsystem Level

Derive ranges at the subsystem level to facilitate:

  • System validation

    It is a best practice to analyze individual subsystems in your model one at a time. This practice makes it easier to understand the atomic behavior of the subsystem. It also makes debugging easier by isolating the source of any issues.

  • Calibration

    The results from the analysis at subsystem level are based only on the settings specified within the subsystem. The proposed data types cover the full intended design range of the subsystem. Based on these results, you can determine whether you can reuse the subsystem in other parts of your model.

Derive Ranges at the Subsystem Level

The complete procedure for deriving ranges is described in How to Derive Ranges.

To derive ranges at the subsystem level, the key points to remember are:

  • The subsystem or subchart must be atomic.

    An atomic subsystem executes as a unit relative to the parent model. Atomic subsystem block execution does not interleave with parent block execution. You can extract atomic subsystems for use as standalone models.

  • In the Fixed-Point Tool, under System Under Design (SUD), select the subsystem of interest.

  • Under Range Collection Mode, select Derived ranges as the method of range collection.

Related Topics