<< Chapter < Page Chapter >> Page >

Although developed here for the linear-phase filter, [link] is a very general design approach for the FIR filter that allows arbitraryphase, as well as uneven frequency sampling and a weighting function in the error definition. For the arbitrary phase case, a complex F is obtained from sampling Equation 28 from FIR Digital Filters and the full h ( n ) is used. For the special case of the equally-spaced frequency samples and linear- phasefilter with unity weighting, the solution of [link] or [link] is the same as given by the frequency sampling design formulas.

One of the important uses of the unequally spaced frequency samples is to create a transition band between the pass and stopbands where there are nosamples. This “don't care" band does not contribute to the error measure q and allows better approximation to occur over the pass and stopbands.

Of the many ways to solve [link] or [link] , one of the easiest and most reliable is the use of Matlab, which has a special command to solve this least-mean-squared errorproblem. Equation [link] should not be solved directly. For large L , it is ill-conditioned and a direct solution will probably have large errors.Matlab uses special algorithms to minimize these numerical errors.

This approach was applied to the same problems that were solved by frequency sampling in the previous section. For N = L , the same results are obtained, thus verifying the theoretical prediction. As L becomes larger compared to N , more control is exerted over the behavior between the original sample points. As L becomes large compared to N , the solution approaches the same results as obtained where theerror is defined as a continuous function of frequency and the integral of the squared error is minimized.Although the solution of the normal equations is a powerful and flexible technique, it can be slow, have numerical problems, and require largeamounts of computer memory.

Examples of discrete least squared error filter design

Here we will give examples of several least squared error designs of FIR filters.

 This graph is labeled Length-15 FIR lowpass filter designed by least squared error. The x axis is labeled Normalized Frequency, and the y axis is labeled Amplitude Response, A. This figure consist of a box formed by the x and y axes and a line that extends perpendicularly from the y at y=14 and a line extending perpedicularly to the x axis at x=.5. In addition to this box there is a waveform that begins on the y axis just below the line perpendicular to the y axis. The waveform travels above the line and then back below and then back across the line further above the line. Then the line takes a very negative slope crossing the line perpendicular to the y axis and also the line perpendicular to the x axis. The line continues below the x axis crossing the axis just before x=.6. The line then undulates above and below the axis until the graph ends. There is an arrow pointing to the middle of the wave between existing inside the box. This arrow labels the area Achieved Amplitude and then below that is another arrow that labels the line perpendicular to the x axis Desired Amplitude.
Frequency Response of Length-15 FIR Filter Designed by Least Squared Error

As for the frequency sampling design, we see a good lowpass filter frequency response with the actual amplitude interpolating the desiredvalues at different points from the frequency sampling example in [link] even though the length and band edge are the same. Notice there is less over shoot but more ripple near f = 0 . The Gibbs phenomenon is the same as for the Fourier series.

If a transition band is introduces in the ideal amplitude response between f = 0 . 4 and f = 0 . 6 with a straight line, the overshoot is reduced significantly but with a slightly slower transition from the pass to stop band. This is illustrated in [link] .

Frequency Response of Length-15 FIR Filter with a Transition Band Designed by Least Squared Error

Continuous frequency definition of error

Because the energy of a signal is the integral of the sum of the squares of the Fourier transform magnitude and because specifications are usuallygiven in the frequency domain, a very reasonable error measure to minimize is the integral squared error given by

Get Jobilize Job Search Mobile App in your pocket Now!

Get it on Google Play Download on the App Store Now




Source:  OpenStax, Digital signal processing and digital filter design (draft). OpenStax CNX. Nov 17, 2012 Download for free at http://cnx.org/content/col10598/1.6
Google Play and the Google Play logo are trademarks of Google Inc.

Notification Switch

Would you like to follow the 'Digital signal processing and digital filter design (draft)' conversation and receive update notifications?

Ask