Discrete-time Fourier Transform (DTFT)

A slecture by ECE student Xian Zhang

Partly based on the ECE438 Fall 2014 lecture material of Prof. Mireille Boutin.



Outline

  • Definition
  • Periodicity property
  • Example of computation of DTFT of a complex exponential
  • Conclusion
  • References


Definition

The discrete-time Fourier transform (DTFT) of a discrete set of real or complex numbers x[n] with n=all integers, is a Fourier series, which produces a periodic function of a frequency as the variable, $ \omega $. The Fourier series is:

$ \begin{align} \\ \mathcal{X}_1(\omega) & = \sum_{n=-\infty}^{\infty}x[n]e^{-j\omega n} \\ \end{align} $

Inverse DTFT is :

$ \begin{align} \\ \quad x[n] & =\frac{1}{2\pi}\int_{-\pi}^{\pi}\mathcal{X}(\omega)e^{j\omega n}d\omega \\ \end{align} $



Periodicity property

$ \mathcal{X}(\omega) $ is periodic with period $ 2\pi $. Because,

$ \begin{align} \\ \mathcal{X}(\omega+2\pi) & = \sum_{n=-\infty}^{\infty}x[n]e^{-j(\omega+2\pi)n}\\ & =\sum_{n=-\infty}^{\infty}x[n]e^{-j\omega n}e^{-j2\pi n}\\ & =\sum_{n=-\infty}^{\infty}x[n]e^{-j\omega n}\\ & =\mathcal{X}(\omega)\\ \end{align} $



Example of computation of DTFT of a complex exponential

Given: $ \begin{align} \\ x[n] =e^{j\omega_o n}\\ \end{align} $

Proof: $ \begin{align} \\ \mathcal{X}(\omega) & = 2\pi rep_{2\pi}(\delta(\omega - \omega_o))\\ \end{align} $


Firstly, let's try:

$ \begin{align} \\ \mathcal{X}(\omega) & = \sum_{n=-\infty}^{\infty}e^{j\omega_o n}e^{-j\omega n}\\ & = \sum_{n=-\infty}^{\infty}e^{j(\omega-\omega_o) n}\\ & = \infty, \omega = \omega_o; 0, else \\ \end{align} $

How to compute something that diverges:


$ \begin{align} \\ e^{j\omega_o n} & =\frac{1}{2\pi}\int_{0}^{2\pi}\mathcal{X}(\omega)e^{j\omega n}d\omega \\ \end{align} $

Assume $ \omega $ is between $ 0 $ to $ 2\pi $.

If $ \begin{align} \\ \mathcal{X}(\omega) & =2\pi \delta(\omega - \omega_o).\\ \end{align} $ The result works for $ \omega $ between $ 0 $ to $ 2\pi $.

But this is not the final answer. Because Fourier transform must be periodic but the answer we got above is not periodic.

Since $ \mathcal{X}(\omega) $ must be periodic with period $ 2\pi $.

The final answer is:

$ \begin{align} \\ \mathcal{X}(\omega) & = 2\pi rep_{2\pi}(\delta(\omega - \omega_o))\\ \end{align} $



Conclusion

Understand the basic calculation of Discrete-time Fourier Transform (DTFT) and inverse Fourier transform. Notice an important property of DTFT: the periodicity property. Master the basic DTFT computation of complex exponential.



Reference

[1]. Mireille Boutin, "ECE438 Digital Signal Processing with Applications," Purdue University August 26, 2009




Back to ECE438 slectures, Fall 2014

Alumni Liaison

Ph.D. on Applied Mathematics in Aug 2007. Involved on applications of image super-resolution to electron microscopy

Francisco Blanco-Silva