Circular Convolution - Discrete Sequences

Discrete Sequences

Similarly, for discrete sequences and period N, we can write the circular convolution of functions h and x as:

\begin{align} (x_N * h) \ &\stackrel{\mathrm{def}}{=} \ \sum_{m=-\infty}^\infty h \cdot x_N \\ &= \sum_{m=-\infty}^\infty \left( h \cdot \sum_{k=-\infty}^\infty x \right). \end{align}

This corresponds to matrix multiplication, and the kernel of the integral transform is a circulant matrix.

Read more about this topic:  Circular Convolution

Famous quotes containing the word discrete:

    “One can describe a landscape in many different words and sentences, but one would not normally cut up a picture of a landscape and rearrange it in different patterns in order to describe it in different ways. Because a photograph is not composed of discrete units strung out in a linear row of meaningful pieces, we do not understand it by looking at one element after another in a set sequence. The photograph is understood in one act of seeing; it is perceived in a gestalt.”
    —Joshua Meyrowitz, U.S. educator, media critic. “The Blurring of Public and Private Behaviors,” No Sense of Place: The Impact of Electronic Media on Social Behavior, Oxford University Press (1985)