Digital Signal Processing/Discrete-Time Fourier Transform

From Wikibooks, the open-content textbooks collection

Jump to: navigation, search
Digital Signal Processing

The Discrete-Time Fourier Transform is a version of the fourier transform that is used to convert a discrete data set into a continuous-frequency representation. The DTFT is used mostly in theory, and less in practice, because computers are not usually capable of handling continuous-time frequency data. The DTFT is also useful because it provides a theoretical basis for the Z transform.

Contents

[edit] DTFT


[DTFT]

X(e^{j\omega}) = \sum_{n=-\infty}^\infty x[n]e^{-j\omega n}

The resulting function, X(ejω) is a continuous function that is interesting for analysis. It can be used in programs, such as Matlab, to design filters and obtain the corresponding time-domain filter values.

[edit] DTFT Convolution Theorem

Like the CTFT, the DTFT has a convolution theorem associated with it. However, since the DTFT results in discrete-frequency values, the convolution theorem needs to be modified as such:

DTFT Convolution Theorem
Multiplication in the continuous time domain becomes discrete convolution in the discrete frequency domain. Convolution in the continuous time domain becomes mulitplication in the discrete frequency domain.

[edit] Energy

It is sometimes helpful to calculate the amount of energy that exists in a certain set. These calculations are based off the assumption that the different values in a set are voltage values, however this doesn't necessarily need to be the case to employ these operations.

We can show that the energy of a given set can be given by the following equation:

\mathcal{E}_x = \sum_{n=-\infty}^\infty \left| x[n] \right| ^2

[edit] Energy in Frequency

Likewise, we can make a formula that represents the power in the continuous-frequency output of the DTFT:

\mathcal{E}_x = \frac{1}{2\pi} \int_{-\infty}^\infty \left|X(e^{j \omega})\right|^2 d\omega

[edit] Parseval's Theorem

Parseval's theorem states that the energy amounts found in the time domain must be equal to the energy amounts found in the frequency domain:

\sum_{n=-\infty}^\infty \left| x[n] \right| ^2 = \frac{1}{2\pi} \int_{-\infty}^\infty \left|X(e^{j \omega})\right|^2 d\omega

[edit] Power Density Spectrum

We can define the power density spectrum of the continuous-time frequency output of the DTFT as follows:

S_{xx}(e^{j\omega}) = \left|X(e^{j \omega})\right|^2

The area under the power density spectrum curve is the total energy of the signal.

Retrieved from "http://en.wikibooks.org/wiki/Digital_Signal_Processing/Discrete-Time_Fourier_Transform"