core Estimated learning time: 10 h

S5-3.2 IIR Filter Design & Realization

You can design digital Butterworth and Chebyshev filters using impulse invariance and the bilinear transform, apply spectral transformations, compare IIR with FIR, and realize filters in direct form I and II.

IIR design borrows the analog filter catalogue and maps it into discrete time, which is why Butterworth and Chebyshev appear in a digital course at all. The bilinear transform is preferred over impulse invariance because it cannot alias, and the price is frequency warping - prewarping is the correction, and knowing why the step exists is more useful than the formula. Direct form II uses half the delay elements of form I, and in finite precision that structural choice changes the filter's actual behaviour.

Work through these

  • Butterworth approximation

    The first of the two classic analog responses, borrowed into a digital course because the design route goes through it. Flat in the passband, and gentle in its transition.

  • Chebyshev approximation

    The second, trading ripple for a sharper transition at the same order. Choosing between the two is the first real design decision in this topic.

  • Impulse invariant transformation technique

    The first way of mapping an analog design into discrete time, and the one with a flaw. It can alias, which is exactly what the next item avoids.

  • Bilinear transform technique and prewarping

    The preferred route, because it cannot alias. The price is frequency warping, and prewarping is the correction; knowing why the step exists is worth more than the formula for it.

  • Digital Butterworth and Chebyshev filters

    The two classic responses now realised as digital filters, which is where the topic has been heading. Design one end to end at least once.

  • Spectral transformation techniques

    Turning one design into another, so a single lowpass prototype serves highpass and bandpass needs. It saves an enormous amount of repeated work.

  • Comparison between FIR and IIR filters

    The choice you actually make in practice, stated as a trade rather than a preference. Phase linearity against filter order is the axis, and the next topic is the other side of it.

  • Realization of IIR filters: direct form I and direct form II

    Two structures computing the same response with different amounts of memory. In finite precision that structural choice changes how the filter really behaves, which is the subject of a later topic.

  • Lab: design Butterworth and Chebyshev type 1 and 2 LPF, HPF, BPF and BSF in MATLAB

    The bench version of the design work, across all four filter shapes. Doing it in software first is what makes the hardware version in the next item tractable.

  • Lab: implement second order IIR filters on the DSK

    The same designs running on a real processor. Second order sections are the standard building block, and this is where you find out why cascading them is preferred to one big filter.

Sign in to keep your progress.

Free resources

Links last checked 30 Aug 2026.

Stuck here?

Ask a mentor. A real person answers, and they can see exactly which topic you're on. Usually within a couple of working days.

Checking your session…

Topics shown in module order.