DSP Blockset

Example: Using the IFFT Block

In the model below, the Sine Wave block again generates two frame-based sinusoids, one at 15 Hz and the other at 40 Hz. The sinusoids are summed point-by-point to generate the compound sinusoid

which is transformed to the frequency domain using an FFT block. The frequency-domain signal is then immediately transformed back to the time domain by the IFFT block, and the difference between the original time-domain signal and transformed time-domain signal is plotted on the scope.

To build the model, make the following parameter settings (leave unlisted parameters in their default settings):

• In the Sine Wave block, set:
• Amplitude = 1
• Frequency = [15 40]
• Phase offset = 0
• Sample time = 0.001
• Samples per frame = 128
• In the Matrix Sum block, set Sum along = Rows.
• In the FFT block, set Output in bit-reversed order =
• In the IFFT block, set:
• Input is in bit-reversed order =
• Input is conjugate symmetric =
• In the Sum block, set List of signs = |++.
• In the Gain block, set Gain = -1.
• In the Scope properties panel of the Vector Scope block, set Input domain = Time
• Set the Stop time in the Parameters dialog box to inf, and start the simulation by selecting Start from the Simulation menu.

The flat line on the scope suggests that there is no difference between the two signals, and that the IFFT block has perfectly reconstructed the original time-domain signal from the frequency-domain input.

More precisely, the two signals are identical to within round-off error, which can be seen by selecting Autoscale from the right-click menu on the scope. The enlarged trace shows that the differences between the two signals are on the order of 10^-15.

Example: Using the FFT Block

Statistics, Estimation, and Linear Algebra