[Scilab-users] Frequency response

[Rafael Guerra]

Just to refresh some definitions, for non-DSP specialists like me.

For a Linear Time Invariant system with impulse response h(t) and transfer function H(f), input/output signals x(t) / y(t) and Fourier transforms X(f) / Y(f)  we have:

    y(t) = x(t) * h(t)  (convolution)  ??  Y(f) = X(f).H(f)  (product)

When the input is a delta function *(t)  the output is the impulse response h(t) itself.

As indicated by Tim W., to obtain the system's transfer function H(f)  ("frequency response"), padding the (time) impulse response with enough zeros will produce enough spectral resolution in the FFT, and this is the most straightforward way.

Otherwise you can use for instance Scilab function's  poly, syslin, and bode;
which sounds like using a sledgehammer to crack a nut.

Regards,
Rafael

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20180916/4c8e672b/attachment.htm>