Digital Filter Cascade Implementation Example

Digital Filter implementation in direct form has significant sensitivity to coefficient quantization. The larger the order of the filter, the more sensitive the frequency response and impulse response of the filter are to coefficient quantization. It is known that breaking the transfer function into the 1st order and 2nd order stages and implementing them in cascade form (interconnected stages in series) is considerably more robust filter structure in terms of preserving impulse response and frequency response of the filter resulting in considerably lower sensitivity to coefficient quantization. This example shows that for a desired transfer function of a digital filter how we can find the needed stages coefficients and implement the cascade canonical realization.