Analyzing Floating-Point Filters (Digital Filter Design Toolkit)

After you enter the target specifications for a digital filter, you can analyze the characteristics of the resulting filter in the Configure Classical Filter Design dialog box of the Classical Filter Design Express VI by evaluating the pole-zero plot, the magnitude response, and the filter order.

Magnitude Response

The frequency response of a digital filter is defined by H(e^j2πf), and the magnitude response is defined by |H(e^j2πf)|. For discrete-time systems, H(e^j2πf) is periodic with a period of f_s. For real-valued digital filters, the magnitude response is symmetric with respect to 0, ±f_s, ±2f_s, …. Therefore, you can calculate the magnitude response for only [0, f_s/2], which contains the frequencies between 0 and the Nyquist frequency. The magnitude response graph in the Configure Classical Filter Design dialog box includes a green vertical line to indicate the location of f_s/2.

Filter Order Specification

The Classical Filter Design Express VI automatically computes the minimal filter order required to fulfill the given filter specification and displays the order in the Filter order indicator. With the same specification, you can use different algorithms to create digital filters with different filter orders. You can estimate the computational complexity and cost based on the filter order. If you have strict requirements for the system, the filter order can help you determine if the filter is acceptable.