DOCSLIB.ORG

10. the Discrete-Time Fourier Transform (DTFT)

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
10. the Discrete-Time Fourier Transform (DTFT) 10. The Discrete-Time Fourier Transform (DTFT) 10.1. Definition of the discrete-time Fourier transform The Fourier representation of signals plays an important role in both continuous and discrete signal processing. In this section we consider discrete signals and develop a Fourier transform for these signals called the discrete-time Fourier transform, abbreviated DTFT. The discrete-time Fourier transform of a discrete sequence x( m) is defined as follows: ∞ ~ ~ X( ejω )= ∑ x() m e −jm ω (10.1) m=−∞ where ω~ is called the normalized frequency. ~ The notation X (e jω ) is justified by the observation that the frequency dependency is in exponential form. In order for the DTFT of a sequence to exist, the summation in (10.1) must converge. It will hold if x() m is absolutely summable, that is ∞ ∑ x() m < ∞ . (10.2) m=−∞ Note that the DTFT of a discrete-time sequence is a function of a continuous variable ω~ . 222 Since ∞ ∞ ~ ~ ~ X (e j()ω+ 2 π )= ∑ x() m e −jm() ω+ 2 π = ∑ x() m e−jm ω e − jm 2 π = m=−∞ m=−∞ ∞ ~ ~ = ∑ x() me−jm ω = X ( e jω ) m=−∞ then the DTFT is periodic in ω~ with a period of 2π . ~ Since X(e jω ) is periodic in ω~ with the period equal to 2π , we can express it by an exponential Fourier series in variable ω~ . Therefore ∞ ∞ jω~ ~ jmω~ ~ −jm ω~ X (e )= ∑cm e = ∑ c−m e (10.3) m=−∞ m=−∞ where π 1 ~ ~ c~ = X (ejω ) e jmω dω~ . (10.4) −m 2π ∫ −π Comparison of equations (10.1) and (10.3) shows that the discrete signal x( m) jω~ ~ corresponding to the spectrum X (e) ixs( m) = c−m . Therefore, from equation (10.4) the inverse discrete-time Fourier transform is π 1 ~ ~ x() m= X (ejω) e jm ω dω~ . (10.5) 2π ∫ −π To shed more light on the subject we will derive the DTFT using an alternative approach. Let us consider the Fourier transform of a continuous-time signal x() t ∞ X()jω = ∫ )x te (−j ωt d t . −∞ We approximate the integral as follows: 223 ∞ −j ωmTs X ()jω ≅ ∑ x ( mTs )e Ts (10.6) m=−∞ and replace the product ωTs by the normalized frequency ~ ω ω = ωTs = . f s If we ignore the scale factor Ts and replace x( mTs ) by x( m), the resulting ~ summation, denoted by X (e jω ), is the DTFT ∞ ~ ~ X (e jω )= ∑ x() m e−j ωm . (10.7) m=−∞ Example 10.1 Let us consider the signal x( m) = am u( m) a <1 . The DTFT of this signal is ∞ ∞ ~ ~ ~ m X (e jω )= ∑ame−jm ω = ∑(ae−j ω ) . m=0 m=0 Since the expression on the right hand side is the geometric series, we obtain jω~ 1 X (e )= ~ 1− a e−j ω provided a <1. Example 10.2 Let x( m) be the unit sample δ(m) . The DTFT of this signal is ∞ ~ ~ X (e jω )=∑ δ()m e−j ωm = 1. m=−∞ 224 10.2. Some properties of the DTFT In this section we formulate some properties of the discrete time Fourier transform. Periodicity This property has already been considered and it can be written as follows ~ ~ X (e j(ω+ 2 π) )= X (e jω ). (10.8) Linearity The DTFT is a linear operator, i.e. the discrete-time Fourier transform of a signal x( m) = a1 x 1( m)+ a2 x 2 ( m) is jω~ jω~ jω~ X(e)= a1 X 1(e )+ a2 X 2 (e ) jω~ where X k (e ) is the DTFT of xk ( m) ( k=1 , 2). Shifting Let us consider a shifted signal ˆx( m) = x( m− m0 ). The DTFT of this signal is (see (10.7)) ∞ ∞ ~ ~ ~ ~ ˆ jω −jm ω −j()m − m0 ω −jm0 ω X (e )=∑ x() m − m0 e =∑ x () m − m0 e e . m=−∞ m=−∞ Let k= m − m0 , then ∞ ~ ~ ~ ~ ~ Xˆ (ejω )= e−jm0 ω ∑ x() k e−jk ω = e−jm0 ω X ( e jω ). k =−∞ Thus, we conclude that shifting in time results in the multiplication of the DTFT ~ by a complex exponential e−jm0 ω . 225 Example 10.3 Let us consider the shifted unit sample x( m) = δ( m− m0 ). Using the shifting property and knowing that the DTFT of δ(m) is 1, we obtain ~ ~ X (e.jω )= e−jm0 ω Frequency shifting Let us consider a signal x( m) multiplied by e jmω0 ˆx( m) = x( m)e jmω0 . The DTFT of this signal is ∞ ∞ ~ ~ ~ ~ Xˆ (e jω )= ∑ x() m ejmω0 e−jm ω = ∑ x() m e −jm() ω−ω0 = X ()e j ()ω−ω0 . m=−∞ m=−∞ Thus, multiplying a sequence by a complex exponential e jmω0 results in shifting in frequency of the DTFT. Convolution theorem The convolution of signals x( m) and y( m) is given by ∞ x() m∗ y () m = ∑ x ()( k y m− k ). k =−∞ The DTFT of the convolution is ∞ ∞ ∞ ∞ ⎛ ⎞ ~ ⎛ ~ ⎞ ⎜ ⎟ −jm ω ⎜ −jm ω ⎟ ∑∑⎜ x() k y ( m− k )⎟ e = ∑∑⎜ x() k y ( m− k )e ⎟ = mk=−∞ ⎝ =−∞ ⎠ k =−∞ ⎝ m=−∞ ⎠ ∞ ∞ ∞ ∞ ⎛ ~ ~ ⎞ ⎛ ~ ~ ⎞ = ⎜ x() k y ( m− k )e−j(m − k ) ω e −jk ω ⎟ = ⎜ x() k e−jk ω y () p e−jp ω ⎟ = ∑∑⎜ ⎟ ∑∑⎜ ⎟ k =−∞ ⎝ m=−∞ ⎠ k =−∞ ⎝ p=−∞ ⎠ ~ ~ = XY(ejω) ( e j ω ) 226 where p= m− k . Thus, the DTFT of a convolution of signals x( m) and y( m) is the product of the DTFTs of x( m) and y( m) . Parseval’s theorem Let us consider a discrete signal x( m). Parseval’s theorem states that ∞ π 2 1 ~ 2 x() m = X (ejω ) dω~ . (10.9) ∑ 2π ∫ m=−∞ −π In Section 12 it will be shown that this equation gives signal energy in the time domain and in the frequency domain. 10.3. Comparing of the DTFT to the DFT Recall that the DFT of a sequence {xm = x( m)} where m= 0 , 1 ," , N −1 is 2π N −1 N −1 − j mn −mn N Xn = ∑ xm w= ∑ xme n= 0 , 1 ,"" , N −1. (10.10) m=0 m=0 On the other hand, the DTFT of the same sequence is N −1 jω~ −jm ω~ X (e )= ∑ xme . (10.11) m=0 Comparing (10.10) to (10.11) we find jω~ XXn = (e ) 2πn n= 0 , 1 ,"" , N −1. (10.12) ω=~ N Equation (10.12) states that the coefficients of the DFT are samples of the 2π continuous spectrum given by the DTFT at ω~ = n . N Note that the DFT coefficients correspond to N samples of the X( z) taken at N equally spaced points around the unit circle 227 2π j n z = e N n= 0 , 1 ," , N − 1. 10.4. Generalized DTFT Some discrete-time signals do not have a DTFT but they have a generalized DTFT as explained below. ~ Let the DTFT of a signal x( m) be X (e jω )= δ( ω~). To find this signal, we use the inverse DTFT: π 1 ~ 1 x() m = δ() ω~ejmω d ω~ = . 2π ∫ 2π −π 1 This result states that the constant signal x() m = has the DTFT equal to δ() ω~ . 2π Hence, the constant signal x() m =1 has the DTFT equal to 2πδ(ω~), or ~ x() m =1 ↔ X (ejω )= 2 πδ( ω~) . (10.13) Note that the signal x() m =1 does not have the DTFT in the ordinary sense because the series ∞ ~ ∑e−jm ω m=−∞ is not convergent. Therefore we say that 2πδ(ω~) is a generalized DTFT of the signal x() m =1. Now we consider a discrete signal x( m) having the DTFT jω~ ~ X (e )= δ( ω + ω0 ). Then π 1 ~ 1 x() m = δ() ω~ + ωejmω d ω~ = e−jm ω0 2π ∫ 0 2π −π or ~ −jm ω0 jω ~ x() m = e ↔ X (e)= 2 πδ( ω + ω0 ) (10.14) holds. 228 Likewise, we find ~ jmω0 jω ~ x() m = e ↔ X (e)= 2 πδ( ω − ω0 ). (10.15) Since 1 j(ω0m +α) 1 −j( ω0m +α) cos()ω0 m + α = e + e = 2 2 1 1 = ejα e jω0m + e−j α e−j ω0m 2 2 then using (10.14) and (10.15) we determine the DTFT of the signal x() m =cos ( ω0m + α ) as follows: jω~ 1 jα ~ 1 −j α ~ X (e )=e 2 πδ() ω− ω0 +e 2 πδ() ω + ω0 = 2 2 jα ~ −j α ~ =π(e δω−ω (0 ) + e δω+ω (0 )) or jω~ jα ~ −j α ~ x() m =cos ( ω0m +α↔ ) X (e) =π( e δω−ω+ (0 ) e δω+ω (0 )). (10.16) A similar approach leads to the DTFT of the signal x( m) = sin(ω0m + α) 1 1 x() m =sin ( ωm + α ) = e j(ω0m +α) − e−j( ω0m +α) = 0 2j 2j 1 1 = ejα e jω0m − e−j α e−j ω0m . 2j 2j To determine the DTFT of x( m), we apply (10.14) and (10.15) jω~ 1 jα ~ 1 −j α ~ X (e )=e 2 πδ() ω− ω0 − e 2πδ() ω+ ω0 = 2j 2j −j α ~ jα ~ =πj( e δω+ω (0 ) − e δω−ω (0 )) or jω~ −j α ~ jα ~ x() m =sin( ω0m +α↔) X (e) =π j( e δω+ω− (0 ) e δω−ω (0 )). (10.17) 229 In the special case when α = 0 we have: jω~ ~ ~ x() m =cos ω0m ↔ X (e)=πδω−ω( (0 ) +δω+ω (0 )) (10.18) jω~ ~ ~ x() m=sin ω0 m ↔ X (e)= j πδω+ω( (0 ) −δω−ω (0 )). (10.19) 10.5. Frequency response of LTI discrete systems Let an LTI discrete system be represented by its unit sample response h( m) (see Fig.10.1).
Load more

Recommended publications
  • Convolution! (CDT-14) Luciano Da Fontoura Costa
    Convolution! (CDT-14) Luciano da Fontoura Costa To cite this version: Luciano da Fontoura Costa. Convolution! (CDT-14). 2019. hal-02334910 HAL Id: hal-02334910 https://hal.archives-ouvertes.fr/hal-02334910 Preprint submitted on 27 Oct 2019 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Convolution! (CDT-14) Luciano da Fontoura Costa [email protected] S~aoCarlos Institute of Physics { DFCM/USP October 22, 2019 Abstract The convolution between two functions, yielding a third function, is a particularly important concept in several areas including physics, engineering, statistics, and mathematics, to name but a few. Yet, it is not often so easy to be conceptually understood, as a consequence of its seemingly intricate definition. In this text, we develop a conceptual framework aimed at hopefully providing a more complete and integrated conceptual understanding of this important operation. In particular, we adopt an alternative graphical interpretation in the time domain, allowing the shift implied in the convolution to proceed over free variable instead of the additive inverse of this variable. In addition, we discuss two possible conceptual interpretations of the convolution of two functions as: (i) the `blending' of these functions, and (ii) as a quantification of `matching' between those functions.
    [Show full text]
  • The Convolution Theorem
    Module 2 : Signals in Frequency Domain Lecture 18 : The Convolution Theorem Objectives In this lecture you will learn the following We shall prove the most important theorem regarding the Fourier Transform- the Convolution Theorem We are going to learn about filters. Proof of 'the Convolution theorem for the Fourier Transform'. The Dual version of the Convolution Theorem Parseval's theorem The Convolution Theorem We shall in this lecture prove the most important theorem regarding the Fourier Transform- the Convolution Theorem. It is this theorem that links the Fourier Transform to LSI systems, and opens up a wide range of applications for the Fourier Transform. We shall inspire its importance with an application example. Modulation Modulation refers to the process of embedding an information-bearing signal into a second carrier signal. Extracting the information- bearing signal is called demodulation. Modulation allows us to transmit information signals efficiently. It also makes possible the simultaneous transmission of more than one signal with overlapping spectra over the same channel. That is why we can have so many channels being broadcast on radio at the same time which would have been impossible without modulation There are several ways in which modulation is done. One technique is amplitude modulation or AM in which the information signal is used to modulate the amplitude of the carrier signal. Another important technique is frequency modulation or FM, in which the information signal is used to vary the frequency of the carrier signal. Let us consider a very simple example of AM. Consider the signal x(t) which has the spectrum X(f) as shown : Why such a spectrum ? Because it's the simplest possible multi-valued function.
    [Show full text]
  • The Discrete-Time Fourier Transform and Convolution Theorems: a Brief Tutorial
    The Discrete-Time Fourier Transform and Convolution Theorems: A Brief Tutorial Yi-Wen Liu 25 Feb 2013 (Revised 21 Sep 2015) 1 Definitions and interpretation 1.1 Units Throughout this semester, we will use the integer-valued variable n as the time variable for discrete-time signal processing; that is, n = −∞, ..., −1, 0, 1, ..., ∞. In this convention, the unit of n is dimensionless, but be aware that in reality the sampling period is 1/fs, where fs denotes the sampling rate. In some text books, 1/fs is denoted as T , and its unit is second. In this course we do not do this, thereby favoring conciseness of notations. 1.2 Definition of DTFT The discrete-time Fourier transform (DTFT) of a discrete-time signal x[n] is a function of frequency ω defined as follows: ∞ ∆ X(ω) = x[n]e−jωn. (1) n=−∞ X Conceptually, the DTFT allows us to check how much of a tonal component at fre- quency ω is in x[n]. The DTFT of a signal is often also called a spectrum. Note that X(ω) is complex-valued. So, the absolute value |X(ω)| represents the tonal component’s magnitude, and the angle ∠X(ω) tells us the phase of the tonal component. Also note that Eq. (1) defines the DTFT as the inner product between the signal and the complex exponential tone e−jωn. Because complex exponentials {e−jωn,ω ∈ [−π, π]} form a complete family of orthogonal bases for (practically) all signals of interest, what Eq. (1) essentially does is to project x[n] onto the space spanned by all complex exponen- tials.
    [Show full text]
  • Fourier Transforms & the Convolution Theorem
    Convolution, Correlation, & Fourier Transforms James R. Graham 11/25/2009 Introduction • A large class of signal processing techniques fall under the category of Fourier transform methods – These methods fall into two broad categories • Efficient method for accomplishing common data manipulations • Problems related to the Fourier transform or the power spectrum Time & Frequency Domains • A physical process can be described in two ways – In the time domain, by h as a function of time t, that is h(t), -∞ < t < ∞ – In the frequency domain, by H that gives its amplitude and phase as a function of frequency f, that is H(f), with -∞ < f < ∞ • In general h and H are complex numbers • It is useful to think of h(t) and H(f) as two different representations of the same function – One goes back and forth between these two representations by Fourier transforms Fourier Transforms ∞ H( f )= ∫ h(t)e−2πift dt −∞ ∞ h(t)= ∫ H ( f )e2πift df −∞ • If t is measured in seconds, then f is in cycles per second or Hz • Other units – E.g, if h=h(x) and x is in meters, then H is a function of spatial frequency measured in cycles per meter Fourier Transforms • The Fourier transform is a linear operator – The transform of the sum of two functions is the sum of the transforms h12 = h1 + h2 ∞ H ( f ) h e−2πift dt 12 = ∫ 12 −∞ ∞ ∞ ∞ h h e−2πift dt h e−2πift dt h e−2πift dt = ∫ ( 1 + 2 ) = ∫ 1 + ∫ 2 −∞ −∞ −∞ = H1 + H 2 Fourier Transforms • h(t) may have some special properties – Real, imaginary – Even: h(t) = h(-t) – Odd: h(t) = -h(-t) • In the frequency domain these
    [Show full text]
  • Fourier Analysis
    Chapter 1 Fourier analysis In this chapter we review some basic results from signal analysis and processing. We shall not go into detail and assume the reader has some basic background in signal analysis and processing. As basis for signal analysis, we use the Fourier transform. We start with the continuous Fourier transformation. But in applications on the computer we deal with a discrete Fourier transformation, which introduces the special effect known as aliasing. We use the Fourier transformation for processes such as convolution, correlation and filtering. Some special attention is given to deconvolution, the inverse process of convolution, since it is needed in later chapters of these lecture notes. 1.1 Continuous Fourier Transform. The Fourier transformation is a special case of an integral transformation: the transforma- tion decomposes the signal in weigthed basis functions. In our case these basis functions are the cosine and sine (remember exp(iφ) = cos(φ) + i sin(φ)). The result will be the weight functions of each basis function. When we have a function which is a function of the independent variable t, then we can transform this independent variable to the independent variable frequency f via: +1 A(f) = a(t) exp( 2πift)dt (1.1) −∞ − Z In order to go back to the independent variable t, we define the inverse transform as: +1 a(t) = A(f) exp(2πift)df (1.2) Z−∞ Notice that for the function in the time domain, we use lower-case letters, while for the frequency-domain expression the corresponding uppercase letters are used. A(f) is called the spectrum of a(t).
    [Show full text]
  • Fourier Transform, Convolution Theorem, and Linear Dynamical Systems April 28, 2016
    Mathematical Tools for Neuroscience (NEU 314) Princeton University, Spring 2016 Jonathan Pillow Lecture 23: Fourier Transform, Convolution Theorem, and Linear Dynamical Systems April 28, 2016. Discrete Fourier Transform (DFT) We will focus on the discrete Fourier transform, which applies to discretely sampled signals (i.e., vectors). Linear algebra provides a simple way to think about the Fourier transform: it is simply a change of basis, specifically a mapping from the time domain to a representation in terms of a weighted combination of sinusoids of different frequencies. The discrete Fourier transform is therefore equiv- alent to multiplying by an orthogonal (or \unitary", which is the same concept when the entries are complex-valued) matrix1. For a vector of length N, the matrix that performs the DFT (i.e., that maps it to a basis of sinusoids) is an N × N matrix. The k'th row of this matrix is given by exp(−2πikt), for k 2 [0; :::; N − 1] (where we assume indexing starts at 0 instead of 1), and t is a row vector t=0:N-1;. Recall that exp(iθ) = cos(θ) + i sin(θ), so this gives us a compact way to represent the signal with a linear superposition of sines and cosines. The first row of the DFT matrix is all ones (since exp(0) = 1), and so the first element of the DFT corresponds to the sum of the elements of the signal. It is often known as the \DC component". The next row is a complex sinusoid that completes one cycle over the length of the signal, and each subsequent row has a frequency that is an integer multiple of this \fundamental" frequency.
    [Show full text]
  • Numerical Analysis of Elastic Contact Between Coated Bodies
    Hindawi Advances in Tribology Volume 2018, Article ID 6498503, 13 pages https://doi.org/10.1155/2018/6498503 Research Article Numerical Analysis of Elastic Contact between Coated Bodies Sergiu Spinu 1,2 1 Department of Mechanics and Technologies, Stefan cel Mare University of Suceava, 13th University Street, 720229, Romania 2Integrated Center for Research, Development and Innovation in Advanced Materials, Nanotechnologies, and Distributed Systems for Fabrication and Control (MANSiD), Stefan cel Mare University of Suceava, Romania Correspondence should be addressed to Sergiu Spinu; [email protected] Received 31 May 2018; Revised 28 September 2018; Accepted 11 October 2018; Published 1 November 2018 Academic Editor: Patrick De Baets Copyright © 2018 Sergiu Spinu. Tis is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Substrate protection by means of a hard coating is an efcient way of extending the service life of various mechanical, electrical, or biomedical elements. Te assessment of stresses induced in a layered body under contact load may advance the understanding of the mechanisms underlying coating performance and improve the design of coated systems. Te iterative derivation of contact area and contact tractions requires repeated displacement evaluation; therefore the robustness of a contact solver relies on the efciency of the algorithm for displacement calculation. Te fast Fourier transform coupled with the discrete convolution theorem has been widely used in the contact modelling of homogenous bodies, as an efcient computational tool for the rapid evaluation of convolution products that appear in displacements and stresses calculation.
    [Show full text]
  • Generalized Poisson Summation Formula for Tempered Distributions
    2015 International Conference on Sampling Theory and Applications (SampTA) Generalized Poisson Summation Formula for Tempered Distributions Ha Q. Nguyen and Michael Unser Biomedical Imaging Group, Ecole´ Polytechnique Fed´ erale´ de Lausanne (EPFL) Station 17, CH–1015, Lausanne, Switzerland fha.nguyen, michael.unserg@epfl.ch Abstract—The Poisson summation formula (PSF), which re- In the mathematics literature, the PSF is often stated in a lates the sampling of an analog signal with the periodization of dual form with the sampling occurring in the Fourier domain, its Fourier transform, plays a key role in the classical sampling and under strict conditions on the function f. Various versions theory. In its current forms, the formula is only applicable to a of (PSF) have been proven when both f and f^ are in appro- limited class of signals in L1. However, this assumption on the signals is too strict for many applications in signal processing priate subspaces of L1(R) \C(R). In its most general form, ^ that require sampling of non-decaying signals. In this paper when f; f 2 L1(R)\C(R), the RHS of (PSF) is a well-defined we generalize the PSF for functions living in weighted Sobolev periodic function in L1([0; 1]) whose (possibly divergent) spaces that do not impose any decay on the functions. The Fourier series is the LHS. If f and f^ additionally satisfy only requirement is that the signal to be sampled and its weak ^ −1−" derivatives up to order 1=2 + " for arbitrarily small " > 0, grow jf(x)j+jf(x)j ≤ C(1+jxj) ; 8x 2 R, for some C; " > 0, slower than a polynomial in the L2 sense.
    [Show full text]
  • Lecture 6 Basic Signal Processing
    Lecture 6 Basic Signal Processing Copyright c 1996, 1997 by Pat Hanrahan Motivation Many aspects of computer graphics and computer imagery differ from aspects of conven- tional graphics and imagery because computer representations are digital and discrete, whereas natural representations are continuous. In a previous lecture we discussed the implications of quantizing continuous or high precision intensity values to discrete or lower precision val- ues. In this sequence of lectures we discuss the implications of sampling a continuous image at a discrete set of locations (usually a regular lattice). The implications of the sampling pro- cess are quite subtle, and to understand them fully requires a basic understanding of signal processing. These notes are meant to serve as a concise summary of signal processing for computer graphics. Reconstruction Recall that a framebuffer holds a 2D array of numbers representing intensities. The display creates a continuous light image from these discrete digital values. We say that the discrete image is reconstructed to form a continuous image. Although it is often convenient to think of each 2D pixel as a little square that abuts its neighbors to fill the image plane, this view of reconstruction is not very general. Instead it is better to think of each pixel as a point sample. Imagine an image as a surface whose height at a point is equal to the intensity of the image at that point. A single sample is then a “spike;” the spike is located at the position of the sample and its height is equal to the intensity asso- ciated with that sample.
    [Show full text]
  • Convolution and Filtering
    Convolution and 7 Filtering Lab Objective: The Fourier transform reveals information in the frequency domain about signals and images that might not be apparent in the usual time (sound) or spatial (image) domain. In this lab, we use the discrete Fourier transform to eciently convolve sound signals and lter out some types of unwanted noise from both sounds and images. This lab is a continuation of the Discrete Fourier Transform lab and should be completed in the same Jupyter Notebook. Convolution Mixing two sounds signalsa common procedure in signal processing and analysisis usually done through a discrete convolution. Given two periodic sound sample vectors f and g of length n, the discrete convolution of f and g is a vector of length n where the kth component is given by n−1 X (f ∗ g)k = fk−jgj; k = 0; 1; 2; : : : ; n − 1: (7.1) j=0 Since audio needs to be sampled frequently to create smooth playback, a recording of a song can contain tens of millions of samples; even a one-minute signal has 2; 646; 000 samples if it is recorded at the standard rate of 44; 100 samples per second (44; 100 Hz). The naïve method of using the sum in (7.1) n times is O(n2), which is often too computationally expensive for convolutions of this size. Fortunately, the discrete Fourier transform (DFT) can be used compute convolutions eciently. The nite convolution theorem states that the Fourier transform of a convolution is the element-wise product of Fourier transforms: Fn(f ∗ g) = n(Fnf) (Fng): (7.2) In other words, convolution in the time domain is equivalent to component-wise multiplication in the frequency domain.
    [Show full text]
  • Review of Fourier Analysis
    Review of Fourier Analysis. A. The Fourier Transform. Definition 0.1 The normed linear space L1(R) consists of all functions f, Lebesgue mea- surable on R with the property that the norm Z 1 kfk1 = jf(x)j dx −∞ is finite. Remarks. (a) L1(R) is a linear space under the usual addition and scalar multiplication of functions. 2 1 (b) We saw before that L [0; 1] ⊆ L [0; 1] which follows from the inequality kfk1 ≤ kfk2 (where the norms are on the spaces Lr[0; 1], r = 1; 2). However this is false if [0; 1] is replaced by R. 1 1 (c) The spaces Cc(R) and Cc (R) are both dense in L (R). Definition 0.2 The Fourier transform of f 2 L1(R), denoted fb(γ), is given by, Z 1 fb(γ) = f(x) e−2πiγx dx: −∞ Remarks. (a) Note a superficial resemblance to the definition of the Fourier coefficients of a function on [0; 1], namely Z 1 fb(n) = f(x) e−2πinx dx: 0 We will explore this relationship later. (b) We assume that f 2 L1(R) in order to guarantee that the integral converges. This is only technical as the integral can be interpreted for a variety of function spaces. Definition 0.3 The Fourier inversion formula is the following. If fb(γ) is the Fourier trans- form of f(x) then, Z 1 f(x) = fb(γ) e2πiγx dγ: −∞ Remarks. (a) Fourier inversion is not always valid. Indeed a major focus of a rigorous course in Fourier analysis is to determine conditions under which the inversion formula holds pointwise and in other senses.
    [Show full text]
  • Fourier Transforms of a Few Functions
    – 1 – 6. FOURIER TRANSFORM, CONVOLUTION AND CORRELATION Reference: ”The Fourier Transform and Its Applications”, by R. N. Bracewell. 6.1 Fourier Transform A function f(t) can be expressed as Z ∞ f(t) = F(ν) e−i2πνt dν, (6.1) −∞ where F(ν) is the Fourier Transform of f(t). This expression can be understood to mean that f(t) can be decomposed into a sum of ”Fourier components” – e−i2πνt – over a continuous range of ν. In other words, f(t) is decomposed into its frequency components. To show the frequency decomposition more directly, one can express equation (6.1) as Z ∞ Z ∞ f(t) = F(ν) cos(2πνt) dν + i F(ν) sin(2πνt) dν. −∞ −∞ F(ν) then specifies how much each frequency component, e−i2πνt, is present in the sum. Equation (6.1) is analogous to the decomposition of a vector V into its basis vectors, xi, 3 X V = vixi, i=1 −i2πνt where vi and xi are analogs of F(ν) and e . Just as in the case of vectors where one can specify the vector V by its components vi, one can specify f(t) by its Fourier Transform F(ν). For instance, if t and ν are time and frequency respectively, instead of plotting the entire f(t) = 2cos(2πνot) over the infinite t-axis, one can simply specify the Fourier Transform which consists of two delta functions (see below) at frequencies: ±νo. Given f(t), its Fourier Transform can be found by Z ∞ F(ν) = f(t) e+i2πνt dt. −∞ Table 1 gives a list of useful functions and their Fourier Transform.
    [Show full text]