Understanding simulation settings

Sample rate, Simulation time & Symbol rate

The primary simulation settings for a project are defined within the Project settings/Simulation settings tab. The Number of iterations (when performing parameter sweeping or repeating test runs) can be defined here, along with the sampling requirements for our simulation.

SystemLab|Design is designed primarily for block sequence simulation. All signals are stored as time-based, sampled signals to allow for the ability to access instantaneous values of a signal based on its time point in the simulation (also called discrete-time signals). To achieve this, a Sample rate and Simulation time must be defined for the entire simulation.

The Sample rate defines how many samples we wish to capture per second within the simulation window and is generally set to the Nyquist rate (twice the value of the highest frequency component we want to capture in representing an analog signal). When multiplied by the Simulation time we obtain the Total samples that will be required for our simulation. This number should be closely monitored as it is used by all the functional blocks to manage the input and output sizes of arrays, which in turn affects the time to perform calculations and visualize data!

Note

When running simulations which contain a very large number of data samples (1e6 or higher), it is recommended to disable (uncheck) the Save port data check box in the Tool bar. This will improve the simulation performance and reduce memory loading linked to populating arrays for all data ports in the design.

Another parameter that can be set in the project simulation settings is the Symbol rate. This parameter is useful when analyzing digital communications systems. These types of systems are generally defined by the transport of symbol blocks (representing a binary data collection) and operate at defined symbols/sec rate(s) (also called baud rate).

The Symbol rate is defined separately from the Sample rate but as a general rule, when running communication models, it’s recommended to maintain a Sample rate that is twice or greater than that of the Symbol rate.

../_images/Simulation_Settings_1.png

Fig 1: Main settings group for running a simulation. Key elements include the Number of iterations, the Sample rate, the Simulation time and the Symbol rate

Confirming simulation settings with a signal data analyzer

The following example Electrical signal data analyzer views (for a sampled electrical pulse stream) can be used to verify our simulation settings.

In the time-domain graph (Fig 2), the blue dots represent the sampled data points as a function of time. The separation between these samples is equivalent to the Sample period calculation (1E-11 sec) from the Simulation settings tab.

../_images/Simulation_Settings_2.png

Fig 2: Example view of Time data tab (Electrical signal data viewer)

The inverse of the Sample period is the Sample rate (1E+11 Hz) and is equivalent to the delta between the maximum and minimum x-axis points (or simulation bandwidth) displayed on the frequency-domain graph (Fig 3). The spacing between the frequency points is linked to the inverse of the time window of the simulation. For a fixed sample rate, the longer the simulation time, the tighter the spacing (resolution) between the frequency points.

../_images/Simulation_Settings_3.png

Fig 3: Example view of Frequency data tab (Electrical signal data viewer)

Note

Information on the simulator settings can also be found in the upper left corner of the Time data tab (Total samples, Sample period, Time window) and Frequency data tab (Total samples, Sample rate) of a signal data analyzer.

Feedback settings

SystemLab|Design also has a feature for performing the dynamic analysis of systems. When Enable feedback is selected, SystemLab|Design will divide the Simulation time into Feedback segments. Each segment represents a shorter-time simulation that is run just like a normal simulation. These subsets are then concatenated together to complete a full picture of system performance over the defined time window of the simulation.

This flexibility allows for output signals from downstream functional blocks to be used as inputs to upstream functional blocks (feedback loops). The resolution of the feedback system can be changed by increasing the number of segments. For further details on how to run simulations with feedback, see Running simulations with feedback.

../_images/Simulation_Settings_4.png

Fig 4: Feedback settings group (located under Simulation settings tab of Project settings)