24 matrix formulation of convolution from above we

This preview shows page 5 - 8 out of 12 pages.

2.4 Matrix formulation of convolution From above, we have w [ m ]=( h v )[ m ] = N 1 k = 0 h [ m k ] v [ k ] . We want to write this operation as vector w = vector h vector v = L vector v . To make it easier to see, consider an input sequence of length n = 4 and an impulse response of length m = 3: vector v = v 0 v 1 v 2 v 3 vector h = h 0 h 1 h 2 .
Image of page 5

Subscribe to view the full document.

EE 505 B, Autumn, 2011 Linear Systems 6 We know that from the previous example that the output is of length 6, so w 0 w 1 w 2 w 3 w 4 w 5 = bracketleftBigg 6 × 4 matrix bracketrightBigg v 0 v 1 v 2 v 3 . The answer is w 0 w 1 w 2 w 3 w 4 w 5 = h 0 0 0 0 h 1 h 0 0 0 h 2 h 1 h 0 0 0 h 2 h 1 h 0 0 0 h 2 h 1 0 0 0 h 2 v 0 v 1 v 2 v 3 . To interpret this result, note that each column is a shifted impulse response, and for matrix multiplication, each column is scaled by the input vector vector v elements ( v k ). Compare this description with the description of convolution for discrete systems above. (They are the same.) The matrix above is a Toeplitz matrix, and has the general form vector w = h 0 0 · · · 0 0 h 1 h 0 · · · 0 0 h 2 h 1 · · · . . . . . . . . . h 2 · · · h 0 h M 2 . . . . . . h 1 h 0 h M 1 h M 2 · · · . . . h 1 0 h M 1 · · · h M 3 . . . 0 0 · · · h M 2 h M 3 . . . . . . . . . h M 1 h M 2 0 0 0 0 h M 1 bracehtipupleft bracehtipdownrightbracehtipdownleft bracehtipupright L is a M × N matrix v 0 v 1 v 2 . . . v N 1 . A note on Toeplitz matrices (you’ll see the relevance in a moment). For finite n , we don’t know of a closed from expression for the eigenvalues and eigenvectors of the Toeplitz matrix. However, for doubly infinite Toeplitz matrices, the eigenvectors are complex exponentials.
Image of page 6
EE 505 B, Autumn, 2011 Linear Systems 7 Doubly infinite matrices are difficult to write, so we will concentrate on the cyclic convo- lution case. Actually, we care a lot about the cyclic case because it is cyclic (or circular) convolution which is related to the discrete Fourier transform (DFT), and the DFT is efficiently implemented by the fast Fourier transform (FFT). 2.5 Cyclic/circular convolution For N = 4, w 0 w 1 w 2 w 3 = h 0 h 3 h 2 h 1 h 1 h 0 h 3 h 2 h 2 h 1 h 0 h 3 h 3 h 2 h 1 h 0 v 0 v 1 v 2 v 3 . Cyclic convolution is used a lot in digital signal processing (e.g., EE 518). Regular (non-cyclic) convolution can be implemented using circular convolution (and therefore using the FFT) by zero padding the input and impulse response, if the impulse response and input are finite in length (e.g., for finite impulse response (FIR) systems, not infinite impulse response (IIR) systems).
Image of page 7

Subscribe to view the full document.

Image of page 8
You've reached the end of this preview.

{[ snackBarMessage ]}

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern