DSP | Theory | Math | Bilinear Transform

Calculators Theory Z-Domain Filters


  Quantization & Sampling Theorem Math Linear Systems Convolution Fourier Transform

  Z-Transform Bilinear Transform

← Back to Digital Signal Processing

Introduction

The bilinear transformation may be defined by;

\[\displaylines{ s = c\frac{1-z^{-1}}{1+z^{-1}} \\ z^{-1} = \frac{1-s/c}{1+s/c} }\]

where \(c\) is an arbitrary positive constant that we may set to map one analog frequency precisely to one digital frequency. In the case of a lowpass or highpass filter, \(c\) is typically used to set the cut-off frequency to be identical in the analog and digital cases.

Second Order

Given the analog prototype;

\[H_{a}(s) = \frac{1}{(s+a)(s+\bar{a})}\]

where \(a = e^{j \pi /4}\), so that;

\[H_{a}(s) = \frac{1}{(s+e^{j \pi /4})(s+e^{-j \pi /4})} = \frac{1}{s^{2} + \sqrt{2}s + 1}\]

To convert this to digital form, we apply the bilinear transform;

\[s = c\frac{1-z^{-1}}{1+z^{-1}}\]

where we set;

\[c = cot(\omega_{c}T/2) \overset{\Delta}{=} \frac{cos(\omega_{c}T/2)}{sin(\omega_{c}T/2)}\]

to obtain a digital cut-off frequency at \(\omega_{c}\) radians per second. For example, choosing \(\omega_{c} T = \pi/2\) (a cut off at one-fourth the sampling rate), we get;

\[c = \frac{cos(\pi/4)}{sin(\pi/4)} = 1\]

and the digital filter transfer function is;

\[\displaylines{ H_d(z) = H_a\left(\frac{1-z^{-1}}{1+z^{-1}}\right) = \frac{1}{\left(\frac{1-z^{-1}}{1+z^{-1}}\right)^2 + \sqrt{2}\left(\frac{1-z^{-1}}{1+z^{-1}}\right) + 1} \\ = \frac{(1+z^{-1})^2}{(1-2z^{-1}+z^{-2}) + (\sqrt{2} - \sqrt{2}z^{-2}) + (1+2z^{-1}+z^{-2})} \\ = \frac{(1+z^{-1})^2}{(2+\sqrt{2}) + (2-\sqrt{2})z^{-2}} \\ = \frac{1}{2+\sqrt{2}}\frac{(1+z^{-1})^2}{1 + \frac{2-\sqrt{2}}{2+\sqrt{2}}z^{-2}} }\]

Note that the numerator is \((1+z^{-1})^2\), as predicted earlier.

As a check, we can verify that the dc gain is 1;

\[H_d(1) = \frac{2^2}{2+\sqrt{2} + 2-\sqrt{2}} = 1\]

It is also immediately verified that \(H_d(-1) = 0\), i.e., that there is a (double) notch at half the sampling rate. In the analog prototype, the cut-off frequency is \(\omega_a=1\) rad/sec, where the amplitude response is \(G_a(j)=1/\sqrt{2}\) . Since we mapped the cut-off frequency precisely under the bilinear transform, we expect the digital filter to have precisely this gain. The digital frequency response at one-fourth the sampling rate is;

\[H_d(j) = \frac{(1-j)^2}{2+\sqrt{2} - (2-\sqrt{2})} = -\frac{j}{\sqrt{2}}\]

and \(20\log_{10}(\left\vert H_d(j)\right\vert)=-3\) dB as expected.

Note that the phase at cut-off is exactly -90 degrees in the digital filter. This can be verified against the pole-zero diagram in the \(z\) plane, which has two zeros at \(z = -1\), each contributing +45 degrees, and two poles at \(z=\pm j\sqrt{\frac{2-\sqrt{2}}{2+\sqrt{2}}}\), each contributing -90 degrees. Thus, the calculated phase-response at the cut-off frequency agrees with what we expect from the digital pole-zero diagram.

In the \(s\) plane, it is not as easy to use the pole-zero diagram to calculate the phase at \(\omega_a=1\), we quickly obtain;

\[H_a(j\cdot 1) = \frac{1}{j^2 + \sqrt{2}j + 1} = -\frac{j}{\sqrt{2}}\]

and exact agreement with \(H_d(e^{j\pi/2})\) is verified.

Example

With the values;

\[\displaylines{ H_{pass}dB = 3db \\ H_{stop}dB = 26db \\ f_{pass} = 12600Hz \\ f_{stop} = 2100Hz \\ f_{s} = 69300Hz }\]

Determine the analogue prototype using the appropriate prototype transformation and express the result in the form;

\[H(s) = \frac{b_{0}s^{2}}{s^{2} + b_{1}s + b_{2}}\]

So, the analog prototype of a second order lowpass Butterworth filter is;

\[H(s) = \frac{1}{s^{2} + \sqrt{2}s + 1}\]

For a highpass filter, this is further expanded as;

\[H(s) = \frac{1}{\left(\frac{\omega_{c}}{s}\right)^{2} + \sqrt{2}\left(\frac{\omega_{c}}{s}\right) + 1}\]

To design our highpass filter, we first obtain the digital frequency in radians per second;

\[\omega_{d} = 2\pi f = 2\pi(12600) = 25200 \pi \text{ rad/sec, and } T = 1/f_{s} = 1/69300 \text{ sec.}\]

Following the steps of the design procedure, we compute the prewarped analog frequency as;

\[\omega_{a} = \frac{2}{T} tan\left(\frac{\omega_{d}T}{2}\right) = 89072.81\]

After applying the prototype transformation, we have;

\[\displaylines{ A = \omega_{a}^{2} = 7933965481 \\ B = \sqrt{2}\omega_{a} = 12596.80 \\ H(s) = \frac{7933965481}{s^{2} + 12596.80s + 7933965481} }\]

← Back to Digital Signal Processing

Email

me@jah.red