Fourier transform
≈ 55 min read
Names in other languages
No names in other languages are held for this entry yet.
Article bodies are English and are not machine-translated. A name in another language is not a translation of the article.
piano chord (with logarithmic horizontal (frequency) axis). The first three peaks on the left correspond to the fundamental frequencies of the chord (C, E, G). The remaining smaller peaks are higher-frequency overtones of the fundamental pitches. In mathematics, the Fourier transform (FT) is an integral transform that takes a function as input and outputs another function that describes the extent to which various frequencies are present in the original function. The output of the transform is a complex valued function of frequency. The term Fourier transform refers to both the mathematical operation and to this complex-valued function. When a distinction needs to be made, the output of the operation is sometimes called the frequency domain representation of the original function. The Fourier transform is analogous to decomposing the sound of a musical chord into the intensities of its constituent pitches. Functions that are localized in the time domain have Fourier transforms that are spread out across the frequency domain and vice versa, a phenomenon known as the uncertainty principle. The critical case for this principle is the Gaussian function, of substantial importance in probability theory and statistics as well as in the study of physical phenomena exhibiting normal distribution (e.g., diffusion).…
piano chord (with logarithmic horizontal (frequency) axis). The first three peaks on the left correspond to the fundamental frequencies of the chord (C, E, G). The remaining smaller peaks are higher-frequency overtones of the fundamental pitches. In mathematics, the Fourier transform (FT) is an integral transform that takes a function as input and outputs another function that describes the extent to which various frequencies are present in the original function. The output of the transform is a complex valued function of frequency. The term Fourier transform refers to both the mathematical operation and to this complex-valued function. When a distinction needs to be made, the output of the operation is sometimes called the frequency domain representation of the original function. The Fourier transform is analogous to decomposing the sound of a musical chord into the intensities of its constituent pitches. Functions that are localized in the time domain have Fourier transforms that are spread out across the frequency domain and vice versa, a phenomenon known as the uncertainty principle. The critical case for this principle is the Gaussian function, of substantial importance in probability theory and statistics as well as in the study of physical phenomena exhibiting normal distribution (e.g., diffusion). The Fourier transform of a Gaussian function is another Gaussian function. Joseph Fourier introduced sine and cosine transforms (which correspond to the imaginary and real components of the modern Fourier transform) in his study of heat transfer, where Gaussian functions appear as solutions of the heat equation. The Fourier transform can be formally defined as an improper Riemann integral, making it an integral transform, although this definition is not suitable for many applications requiring a more sophisticated integration theory. For example, many relatively simple applications use the Dirac delta function, which can be treated formally as if it were a function, but the justification requires a mathematically more sophisticated viewpoint. The Fourier transform can also be generalized to functions of several variables on Euclidean space, sending a function of "position space" to a function of momentum (or a function of space and time to a function of 4-momentum). This idea makes the spatial Fourier transform very natural in the study of waves, as well as in quantum mechanics, where it is important to be able to represent wave solutions as functions of either position or momentum and sometimes both. In general, functions to which Fourier methods are applicable are complex-valued, and possibly vector-valued. Still further generalization is possible to functions on groups, which, besides the original Fourier transform on or , notably includes the discrete-time Fourier transform (DTFT, group = ), the discrete Fourier transform (DFT, group = ) and the Fourier series or circular Fourier transform (group = , the unit circle ≈ closed finite interval with endpoints identified). The latter is routinely employed to handle periodic functions. The fast Fourier transform (FFT) is an algorithm for computing the DFT. f(x)\ \stackrel{\mathcal{F}}{\longleftrightarrow}\ \widehat f(\xi). For example, the Fourier transform of the delta function is the constant function : \delta(x)\ \stackrel{\mathcal{F}}{\longleftrightarrow}\ 1. The substitution \xi = \tfrac{\omega}{2 \pi} into produces this convention, where function \widehat f is relabeled : \begin{align} \widehat f_3(\omega) &\triangleq \int_{-\infty}^{\infty} f(x)\cdot e^{-i\omega x}\, dx = \widehat f_1\left(\tfrac{\omega}{2\pi}\right),\\ f(x) &= \frac{1}{2\pi} \int_{-\infty}^{\infty} \widehat f_3(\omega)\cdot e^{i\omega x}\, d\omega. \end{align} Unlike the definition, the Fourier transform is no longer a unitary transformation, and there is less symmetry between the formulas for the transform and its inverse. Those properties are restored by splitting the 2 \pi factor evenly between the transform and its inverse, which leads to another convention: \begin{align} \widehat f_2 (\omega) &\triangleq \frac{1}{\sqrt{2\pi}} \int_{-\infty}^{\infty} f(x)\cdot e^{- i\omega x}\, dx = \frac{1}{\sqrt{2\pi}}\ \ \widehat f_1 \left(\tfrac{\omega}{2\pi}\right), \\ f(x) &= \frac{1}{\sqrt{2\pi}} \int_{-\infty}^{\infty} \widehat f_2 (\omega)\cdot e^{ i\omega x}\, d\omega. \end{align} Variations of all three conventions can be created by conjugating the complex-exponential kernel of both the forward and the reverse transform. The signs must be opposites. \|f\|_1 = \int_{\mathbb R}|f(x)|\,dx If f is Lebesgue integrable then the Fourier transform, given by , is well-defined for all . Furthermore, \widehat f\in L^\infty\cap C_0(\mathbb R) is bounded, uniformly continuous and (by the Riemann–Lebesgue lemma) vanishing at infinity. Here C_0(\mathbb R) denotes the space of continuous functions on \mathbb R that approach 0 as x approaches positive or negative infinity. The space L^1(\mathbb R) is the space of measurable functions for which the norm \|f\|_1 is finite, modulo the equivalence relation of equality almost everywhere. The Fourier transform on L^1(\mathbb R) is one-to-one. However, there is no easy characterization of the image, and thus no easy characterization of the inverse transform. In particular, is no longer valid, as it was stated only under the hypothesis that f(x) was "sufficiently nice" (e.g., f(x) decays with all derivatives). While defines the Fourier transform for (complex-valued) functions in , it is not well-defined for other integrability classes, most importantly the space of square-integrable functions . For example, the function f(x)=(1+x^2)^{-1/2} is in L^2 but not L^1 and therefore the Lebesgue integral does not exist. However, the Fourier transform on the dense subspace L^1\cap L^2(\mathbb R) \subset L^2(\mathbb R) admits a unique continuous extension to a unitary operator on . This extension is important in part because, unlike the case of , the Fourier transform is an automorphism of the space . In such cases, the Fourier transform can be obtained explicitly by regularizing the integral, and then passing to a limit. In practice, the integral is often regarded as an improper integral instead of a proper Lebesgue integral, but sometimes for convergence one needs to use weak limit or principal value instead of the (pointwise) limits implicit in an improper integral. and each gives three rigorous ways of extending the Fourier transform to square integrable functions using this procedure. A general principle in working with the L^2 Fourier transform is that finite linear combinations of Gaussians are dense in , and the various features of the Fourier transform, such as its unitarity, are easily inferred for Gaussians. Many of the properties of the Fourier transform can then be proven from two facts about Gaussians: that e^{-\pi x^2} is its own Fourier transform; and that the Gaussian integral . A feature of the L^1 Fourier transform is that it is a homomorphism of Banach algebras from L^1 equipped with the convolution operation to the Banach algebra of continuous functions under the L^\infty (supremum) norm. The conventions chosen in this article are those of harmonic analysis, such that the Fourier transform is both unitary on and an algebra homomorphism from to , without renormalizing the Lebesgue measure. In general, the coefficients \widehat f(\xi) are complex numbers, which have two equivalent forms (see Euler's formula): \widehat f(\xi) = \underbrace{A e^{i \theta}}_{\text{polar coordinate form}} = \underbrace{A \cos(\theta) + i A \sin(\theta)}_{\text{rectangular coordinate form}}. The product with e^{i 2 \pi \xi x} ( ) has these forms: \begin{aligned}\widehat f(\xi)\cdot e^{i 2 \pi \xi x} &= A e^{i \theta} \cdot e^{i 2 \pi \xi x}\\[6pt] &= \underbrace{A e^{i (2 \pi \xi x+\theta)}}_{\text{polar coordinate form}}\\[6pt] &= \underbrace{A\cos(2\pi \xi x +\theta) + i A\sin(2\pi \xi x +\theta)}_{\text{rectangular coordinate form}},\end{aligned} which conveys both amplitude and phase of frequency . Likewise, the intuitive interpretation of is that multiplying f(x) by e^{-i 2\pi \xi x} has the effect of subtracting \xi from every frequency component of function . Only the component that was at frequency \xi can produce a non-zero value of the infinite integral, because (at least formally) all the other shifted components are oscillatory and integrate to zero (see ). It is noteworthy how easily the product was simplified using the polar form, and how easily the rectangular form was deduced by an application of Euler's formula. For a real-valued , has the symmetry property \widehat f(-\xi) = \widehat {f}^* (\xi) (see below). This redundancy enables to distinguish f(x) = \cos(2 \pi \xi_0 x) from . But it cannot determine the actual sign of , because \cos(2 \pi \xi_0 x) and \cos(2 \pi (-\xi_0) x) are indistinguishable on just the real numbers line. \widehat{f}(\xi) = \sum_{n=-\infty}^\infty c_n \cdot \delta \left(\xi - \tfrac{n}{P}\right), where c_n are the Fourier series coefficients of , and \delta is the Dirac delta function. In other words, the Fourier transform is a Dirac comb function whose teeth are multiplied by the Fourier series coefficients. f_P(x) \triangleq \sum_{n=-\infty}^{\infty} f(x+nP) = \frac{1}{P}\sum_{k=-\infty}^{\infty} \widehat f\left(\tfrac{k}{P}\right) e^{i2\pi \frac{k}{P} x}, \quad \forall k \in \mathbb{Z} . The integrability of f ensures the periodic summation converges. Therefore, the samples \widehat f(\tfrac{k}{P}) can be determined by Fourier series analysis: \widehat f\left(\tfrac{k}{P}\right) = \int_{P} f_P(x) \cdot e^{-i2\pi \frac{k}{P} x} \,dx. When f(x) has compact support, f_P(x) has a finite number of terms within the interval of integration. When f(x) does not have compact support, numerical evaluation of f_P(x) requires an approximation, such as tapering f(x) or truncating the number of terms. \varphi (\lambda) = \int_{-\infty}^\infty f(x) e^{i\lambda x} \,dx. In probability theory and mathematical statistics, the use of the Fourier—Stieltjes transform is preferred, because many random variables are not of continuous type, and do not possess a density function, and one must treat not functions but distributions, i.e., measures that possess "atoms". From the higher point of view of group characters, which is much more abstract, all these arbitrary choices disappear, as will be explained in the later section of this article, which treats the notion of the Fourier transform of a function on a locally compact abelian group. \int_{-\infty}^\infty |f(x)| \, dx We denote the Fourier transforms of these functions as \widehat f(\xi) and \widehat h(\xi) respectively. a\ f(x) + b\ h(x)\ \ \stackrel{\mathcal{F}}{\Longleftrightarrow}\ \ a\ \widehat f(\xi) + b\ \widehat h(\xi);\quad \ a,b \in \mathbb C f(x-x_0)\ \ \stackrel{\mathcal{F}}{\Longleftrightarrow}\ \ e^{-i 2\pi x_0 \xi}\ \widehat f(\xi);\quad \ x_0 \in \mathbb R e^{i 2\pi \xi_0 x} f(x)\ \ \stackrel{\mathcal{F}}{\Longleftrightarrow}\ \ \widehat f(\xi - \xi_0);\quad \ \xi_0 \in \mathbb R f(ax)\ \ \stackrel{\mathcal{F}}{\Longleftrightarrow}\ \ \frac{1} \widehat{f}\left(\frac{\xi}{a}\right);\quad \ a \ne 0 The case a=-1 leads to the time-reversal property: f(-x)\ \ \stackrel{\mathcal{F}}{\Longleftrightarrow}\ \ \widehat f (-\xi) \mathsf{Time\ domain} & f & = & f_{_{\text{RE}}} & + & f_{_{\text{RO}}} & + & i\ f_{_{\text{IE}}} & + & \underbrace{i\ f_{_{\text{IO}}}} \\ &\Bigg\Updownarrow\mathcal{F} & &\Bigg\Updownarrow\mathcal{F} & &\ \ \Bigg\Updownarrow\mathcal{F} & &\ \ \Bigg\Updownarrow\mathcal{F} & &\ \ \Bigg\Updownarrow\mathcal{F}\\ \mathsf{Frequency\ domain} & \widehat f & = & \widehat f\!_{_{\text{RE}}} & + & \overbrace{i\ \widehat f\!_{_{\text{IO}}}} & + & i\ \widehat f\!_{_{\text{IE}}} & + & \widehat f\!_{_{\text{RO}}} \end{array} From this, various relationships are apparent, for example: The transform of a real-valued function ({{tmath| f_{_{\text{RE} } }+f_{_{\text{RO} } } }}) is the conjugate symmetric function {{tmath| \widehat f\!_{_{\text{RE} } }+i\ \widehat f\!_{_{\text{IO} } } }}. Conversely, a conjugate symmetric transform implies a real-valued time-domain. The transform of an imaginary-valued function ({{tmath| i\ f_{_{\text{IE} } }+i\ f_{_{\text{IO} } } }}) is the conjugate antisymmetric function {{tmath| \widehat f\!_{_{\text{RO} } }+i\ \widehat f\!_{_{\text{IE} } } }}, and the converse is true. The transform of a conjugate symmetric function (f_{_{\text{RE}}}+i\ f_{_{\text{IO}}}) is the real-valued function {{tmath| \widehat f\!_{_{\text{RE} } }+\widehat f\!_{_{\text{RO} } } }}, and the converse is true. The transform of a conjugate antisymmetric function (f_{_{\text{RO}}}+i\ f_{_{\text{IE}}}) is the imaginary-valued function {{tmath| i\ \widehat f\!_{_{\text{IE} } }+i\ \widehat f\!_{_{\text{IO} } } }}, and the converse is true. \bigl(f(x)\bigr)^*\ \ \stackrel{\mathcal{F}}{\Longleftrightarrow}\ \ \left(\widehat{f}(-\xi)\right)^* (Note: the denotes complex conjugation.) In particular, if f is real, then \widehat f is conjugate symmetric ( Hermitian function): \widehat{f}(-\xi)=\bigl(\widehat f(\xi)\bigr)^*. If f is purely imaginary, then \widehat f is odd symmetric: \widehat f(-\xi) = -(\widehat f(\xi))^*. \operatorname{Re}\{f(x)\}\ \ \stackrel{\mathcal{F}}{\Longleftrightarrow}\ \ \tfrac{1}{2} \left( \widehat f(\xi) + \bigl(\widehat f (-\xi) \bigr)^* \right) \operatorname{Im}\{f(x)\}\ \ \stackrel{\mathcal{F}}{\Longleftrightarrow}\ \ \tfrac{1}{2i} \left( \widehat f(\xi) - \bigl(\widehat f (-\xi) \bigr)^* \right) Substituting \xi = 0 in the definition, we obtain: \widehat{f}(0) = \int_{-\infty}^{\infty} f(x)\,dx. The integral of f over its domain is the total mass or DC bias of the function. \left\|\widehat{f}\right\|_\infty \leq \left\|f\right\|_1 By the Riemann–Lebesgue lemma, \widehat{f}(\xi) \to 0\text{ as }|\xi| \to \infty. However, \widehat{f} need not be integrable. For example, the Fourier transform of the rectangular function, which is integrable, is the sinc function, which is not Lebesgue integrable, because its improper integrals behave analogously to the alternating harmonic series, in converging to a sum without being absolutely convergent. It is not generally possible to write the inverse transform as a Lebesgue integral. However, when both f and \widehat{f} are integrable, the inverse equality f(x) = \int_{-\infty}^\infty \widehat f(\xi) e^{i 2\pi x \xi} \, d\xi holds for almost every . As a result, the Fourier transform is injective on . \langle f, g\rangle_{L^{2}} = \int_{-\infty}^{\infty} f(x) \overline{g(x)} \,dx = \int_{-\infty}^\infty \widehat{f}(\xi) \overline{\widehat{g}(\xi)} \,d\xi, where the bar denotes complex conjugation. The Plancherel theorem, which follows from the above, states that \|f\|^2_{L^{2}} = \int_{-\infty}^\infty \left| f(x) \right|^2\,dx = \int_{-\infty}^\infty \left| \widehat{f}(\xi) \right|^2\,d\xi. Plancherel's theorem makes it possible to extend the Fourier transform, by a continuity argument, to a unitary operator on . On , this extension agrees with original Fourier transform defined on , thus enlarging the domain of the Fourier transform to (and consequently to for ). Plancherel's theorem has the interpretation in the sciences that the Fourier transform preserves the energy of the original quantity. The terminology of these formulas is not quite standardised. Parseval's theorem was proved only for Fourier series, and was first proved by Lyapunov. But Parseval's formula makes sense for the Fourier transform as well, and so even though in the context of the Fourier transform it was proved by Plancherel, it is still often referred to as Parseval's formula, or Parseval's relation, or even Parseval's theorem. See Pontryagin duality for a general formulation of this concept in the context of locally compact abelian groups. h(x) = (f*g)(x) = \int_{-\infty}^\infty f(y)g(x - y)\,dy, where denotes the convolution operation, then: \widehat{h}(\xi) = \widehat{f}(\xi)\, \widehat{g}(\xi). In linear time invariant (LTI) system theory, it is common to interpret as the impulse response of an LTI system with input and output , since substituting the unit impulse for yields . In this case, represents the frequency response of the system. Conversely, if can be decomposed as the product of two square integrable functions and , then the Fourier transform of is given by the convolution of the respective Fourier transforms and . h(x) = (f \star g)(x) = \int_{-\infty}^\infty \overline{f(y)}g(x + y)\,dy then the Fourier transform of is: \widehat{h}(\xi) = \overline{\widehat{f}(\xi)} \, \widehat{g}(\xi). As a special case, the autocorrelation of function is: h(x) = (f \star f)(x) = \int_{-\infty}^\infty \overline{f(y)}f(x + y)\,dy for which \widehat{h}(\xi) = \overline{\widehat{f}(\xi)}\widehat{f}(\xi) = \left|\widehat{f}(\xi)\right|^2. \widehat{f'}(\xi) = \mathcal{F}\left\{ \frac{d}{dx} f(x)\right\} = i 2\pi \xi\widehat{f}(\xi). More generally, the Fourier transformation of the th derivative is given by \widehat{f^{(n)}}(\xi) = \mathcal{F}\left\{ \frac{d^n}{dx^n} f(x) \right\} = (i 2\pi \xi)^n\widehat{f}(\xi). Analogously, {{tmath|1= \mathcal{F}^{-1}\left\{ \frac{d^n}{d\xi^n} \widehat{f}(\xi)\right\} = (-i 2\pi x)^n f(x) }}, so . By applying the Fourier transform and using these formulas, some ordinary differential equations can be transformed into algebraic equations, which are much easier to solve. These formulas also give rise to the rule of thumb " is smooth if and only if quickly falls to for ". By using the analogous rules for the inverse Fourier transform, one can also say " quickly falls to for if and only if is smooth". \left[ U\left( \frac{1}{2\pi}\frac{d}{dx} \right) + U( x ) \right] \psi(x) = 0 leads to eigenfunctions \psi(x) of the Fourier transform \mathcal{F} as long as the form of the equation remains invariant under Fourier transform. In other words, every solution \psi(x) and its Fourier transform \widehat\psi(\xi) obey the same equation. Assuming uniqueness of the solutions, every solution \psi(x) must therefore be an eigenfunction of the Fourier transform. The form of the equation remains unchanged under Fourier transform if U(x) can be expanded in a power series in which for all terms the same factor of either one of , arises from the factors i^n introduced by the differentiation rules upon Fourier transforming the homogeneous differential equation because this factor may then be cancelled. The simplest allowable U(x)=x leads to the standard normal distribution. More generally, a set of eigenfunctions is also found by noting that the differentiation rules imply that the ordinary differential equation \left[ W\left( \frac{i}{2\pi}\frac{d}{dx} \right) + W(x) \right] \psi(x) = C \psi(x) with C constant and W(x) being a non-constant even function remains invariant in form when applying the Fourier transform \mathcal{F} to both sides of the equation. The simplest example is provided by , which is equivalent to considering the Schrödinger equation for the quantum harmonic oscillator. The corresponding solutions provide an important choice of an orthonormal basis for and are given by the "physicist's" Hermite functions. Equivalently one may use \psi_n(x) = \frac{\sqrt[4]{2}}{\sqrt{n!}} e^{-\pi x^2}\mathrm{He}_n\left(2x\sqrt{\pi}\right), where are the "probabilist's" Hermite polynomials, defined as \mathrm{He}_n(x) = (-1)^n e^{\frac{1}{2}x^2}\left(\frac{d}{dx}\right)^n e^{-\frac{1}{2}x^2}. Under this convention for the Fourier transform, we have that \widehat\psi_n(\xi) = (-i)^n \psi_n(\xi). In other words, the Hermite functions form a complete orthonormal system of eigenfunctions for the Fourier transform on . However, this choice of eigenfunctions is not unique. Because of \mathcal{F}^4 = \mathrm{id} there are only four different eigenvalues of the Fourier transform (the fourth roots of unity and ) and any linear combination of eigenfunctions with the same eigenvalue gives another eigenfunction. As a consequence of this, it is possible to decompose as a direct sum of four spaces , , , and where the Fourier transform acts on simply by multiplication by . Since the complete set of Hermite functions provides a resolution of the identity they diagonalize the Fourier operator, i.e. the Fourier transform can be represented by such a sum of terms weighted by the above eigenvalues, and these sums can be explicitly summed: \mathcal{F}[f](\xi) = \int dx f(x) \sum_{n \geq 0} (-i)^n \psi_n(x) \psi_n(\xi) ~. This approach to define the Fourier transform was first proposed by Norbert Wiener. Among other properties, Hermite functions decrease exponentially fast in both frequency and time domains, and they are thus used to define a generalization of the Fourier transform, namely the fractional Fourier transform used in time–frequency analysis. In physics, this transform was introduced by Edward Condon. This change of basis becomes possible because the Fourier transform is a unitary transform when using the right conventions. Consequently, under the proper conditions it may be expected to result from a self-adjoint generator N via \mathcal{F}[\psi] = e^{-i t N} \psi. The operator N is the number operator of the quantum harmonic oscillator written as N \equiv \frac{1}{4\pi}\left(2\pi x - \frac{\partial}{\partial x}\right)\left(2\pi x + \frac{\partial}{\partial x}\right) = -\frac{1}{4\pi}\frac{\partial^2}{\partial x^2} + \pi x^2 - \frac{1}{2}. It can be interpreted as the generator of fractional Fourier transforms for arbitrary values of , and of the conventional continuous Fourier transform \mathcal{F} for the particular value , with the Mehler kernel implementing the corresponding active transform. The eigenfunctions of N are the Hermite functions , which are therefore also eigenfunctions of . Upon extending the Fourier transform to distributions the Dirac comb is also an eigenfunction of the Fourier transform. \begin{align} \mathcal{F}^0 &= \mathrm{id}, \\ \mathcal{F}^1 &= \mathcal{F}, \\ \mathcal{F}^2 &= \mathcal{P}, \\ \mathcal{F}^3 &= \mathcal{F}^{-1} = \mathcal{P} \circ \mathcal{F} = \mathcal{F} \circ \mathcal{P}, \\ \mathcal{F}^4 &= \mathrm{id} \end{align} These equalities of operators require careful definition of the space of functions in question, defining equality of functions (equality at every point? equality almost everywhere?) and defining equality of operators – that is, defining the topology on the function space and operator space in question. These are not true for all functions, but are true under various conditions, which are the content of the various forms of the Fourier inversion theorem. This fourfold periodicity of the Fourier transform is similar to a rotation of the plane by 90°, particularly as the two-fold iteration yields a reversal, and in fact this analogy can be made precise. While the Fourier transform can simply be interpreted as switching the time domain and the frequency domain, with the inverse Fourier transform switching them back, more geometrically it can be interpreted as a rotation by 90° in the time–frequency domain (considering time as the -axis and frequency as the -axis), and the Fourier transform can be generalized to the fractional Fourier transform, which involves rotations by other angles. This can be further generalized to linear canonical transformations, which can be visualized as the action of the special linear group on the time–frequency plane, with the preserved symplectic form corresponding to the uncertainty principle, below. This approach is particularly studied in signal processing, under time–frequency analysis. \left(M^{-1}_\xi T^{-1}_y M_\xi T_yf\right)(x) = e^{i 2\pi\xi y}f(x) , which is multiplication by the constant (independent of ) (the circle group of unit modulus complex numbers). As an abstract group, the Heisenberg group is the three-dimensional Lie group of triples , with the group law \left(x_1, \xi_1, t_1\right) \cdot \left(x_2, \xi_2, t_2\right) = \left(x_1 + x_2, \xi_1 + \xi_2, t_1 t_2 e^{-2 i \pi x_1 \xi_2}\right). Denote the Heisenberg group by . The above procedure describes not only the group structure, but also a standard unitary representation of on a Hilbert space, which we denote by . Define the linear automorphism of by J \begin{pmatrix} x \\ \xi \end{pmatrix} = \begin{pmatrix} -\xi \\ x \end{pmatrix} so that . This can be extended to a unique automorphism of : j\left(x, \xi, t\right) = \left(-\xi, x, te^{-i 2\pi\xi x}\right). According to the Stone–von Neumann theorem, the unitary representations and are unitarily equivalent, so there is a unique intertwiner such that \rho \circ j = W \rho W^*. This operator is the Fourier transform. Many of the standard properties of the Fourier transform are immediate consequences of this more general framework. For example, the square of the Fourier transform, , is an intertwiner associated with , and so we have is the reflection of the original function . \widehat f (\xi) = \int _{-\infty}^\infty e^{-2i \pi t \xi} f(t) \, dt can be studied for complex values of its argument . Depending on the properties of , this might not converge off the real axis at all, or it might converge to a complex analytic function for all values of , or something in between. The Paley–Wiener theorem says that is smooth (i.e., -times differentiable for all positive integers ) and compactly supported if and only if is a holomorphic function for which there exists a constant such that for any integer , \left\vert \xi ^n \widehat f(\xi) \right\vert \leq C_n e^{2\pi a\vert\tau\vert} for some constant . (In this case, is supported on .) This can be expressed by saying that is an entire function that is rapidly decreasing in (for fixed ) and of exponential growth in (uniformly in ). (If is not smooth, but only , a corresponding version holds with the rapid-decrease condition replaced by the appropriate condition. ) The space of such functions of a complex variable is called the Paley–Wiener space. This theorem has been generalised to semisimple Lie groups. If is supported on the half-line , then is said to be "causal" because the impulse response function of a physically realisable filter must have this property, as no effect can precede its cause. Paley and Wiener showed that then, under suitable integrability hypotheses, extends to a holomorphic function on the complex lower half-plane that tends to zero as goes to . A simple converse in this form is false; precise converses require additional growth or Hardy-space hypotheses. f(t)=0\quad (t for some constants , then \widehat f (i\tau) = \int _0^\infty e^{ 2\pi \tau t} f(t) \, dt, convergent for all , is a one-sided Laplace transform of . The usual one-sided version of the Laplace transform is F(s) = \int_0^\infty f(t) e^{-st} \, dt. If is causal and the integrals converge, then . Thus, extending the Fourier transform to the complex domain means it includes the Laplace transform as a special case in the case of causal functions—but with the change of variable . From another, perhaps more classical viewpoint, the Laplace transform by its form involves an additional exponential regulating term that lets it converge outside of the imaginary line where the Fourier transform is defined. As such it can converge for functions and integrals with at most exponential growth in the regulated direction, whereas the original Fourier decomposition cannot, enabling analysis of systems with divergent or critical elements. Two particular examples from linear signal processing are the construction of allpass filter networks from critical comb and mitigating filters via exact pole-zero cancellation on the unit circle. Such designs are common in audio processing, where highly nonlinear phase response is sought for, as in reverb. Furthermore, when extended pulselike impulse responses are sought for signal processing work, the easiest way to produce them is to have one circuit that produces a divergent time response, and then to cancel its divergence through a delayed opposite and compensatory response. There, only the delay circuit in-between admits a classical Fourier description, which is critical. Both the circuits to the side are unstable, and do not admit a convergent Fourier decomposition. However, they do admit a Laplace domain description, with compatible half-planes of convergence in the complex plane (or in the discrete case, the Z-plane), wherein their effects cancel. In modern mathematics the Laplace transform is conventionally subsumed under the aegis of Fourier methods. Both of them are subsumed by the far more general, and more abstract, idea of harmonic analysis. \int _{-\infty}^\infty \widehat f (\sigma + ia) e^{ i 2\pi (\sigma+ia) t} \, d\sigma = \int _{-\infty}^\infty \widehat f (\sigma + ib) e^{ i 2\pi (\sigma+ib) t} \, d\sigma by Cauchy's integral theorem. Therefore, the Fourier inversion formula can use integration along different lines, parallel to the real axis. Theorem: If for , and for some constants and , then f(t) = \int_{-\infty}^\infty \widehat f(\sigma + i\tau) e^{i 2 \pi (\sigma+i\tau) t} \, d\sigma, for any , under the usual hypotheses for Fourier inversion. This theorem implies the Mellin inversion formula for the Laplace transformation, f(t) = \frac 1 {i 2\pi} \int_{b-i\infty}^{b+i\infty} F(s) e^{st}\, ds for any , where is the Laplace transform of . The hypotheses can be weakened, as in standard Fourier inversion results, to being , provided that be of bounded variation in a closed neighborhood of (cf. Dini test), the value of at be taken to be the arithmetic mean of the left and right limits, and that the integrals be taken in the sense of Cauchy principal values. versions of these inversion formulas are also available. \widehat{f}(\boldsymbol{\xi}) = \mathcal{F}(f)(\boldsymbol{\xi}) = \int_{\R^n} f(\mathbf{x}) e^{-i 2\pi \boldsymbol{\xi}\cdot\mathbf{x}} \, d\mathbf{x} where and are -dimensional vectors, and is the dot product of the vectors. Alternatively, can be viewed as belonging to the dual vector space , in which case the dot product becomes the contraction of and , usually written as . All of the basic properties listed above hold for the -dimensional Fourier transform, as do Plancherel's and Parseval's theorem. When the function is integrable, the Fourier transform is still uniformly continuous and the Riemann–Lebesgue lemma holds. \int_{-\infty}^\infty |f(x)|^2 \,dx=1. It follows from the Plancherel theorem that is also normalized. The spread around may be measured by the dispersion about zero defined by D_0(f)=\int_{-\infty}^\infty x^2|f(x)|^2\,dx. In probability terms, this is the second moment of about zero. The uncertainty principle states that, if is absolutely continuous and the functions and are square integrable, then D_0(f)D_0(\widehat{f}) \geq \frac{1}{16\pi^2}. The equality is attained only in the case \begin{align} f(x) &= C_1 \, e^{-\pi \frac{x^2}{\sigma^2} }\\ \therefore \widehat{f}(\xi) &= \sigma C_1 \, e^{-\pi\sigma^2\xi^2} \end{align} where is arbitrary and so that is -normalized. In other words, where is a (normalized) Gaussian function with variance , centered at zero, and its Fourier transform is a Gaussian function with variance . Gaussian functions are examples of Schwartz functions (see the discussion on tempered distributions below). In fact, this inequality implies that: \left(\int_{-\infty}^\infty (x-x_0)^2|f(x)|^2\,dx\right)\left(\int_{-\infty}^\infty(\xi-\xi_0)^2\left|\widehat{f}(\xi)\right|^2\,d\xi\right)\geq \frac{1}{16\pi^2}, \quad \forall x_0, \xi_0 \in \mathbb{R}. In quantum mechanics, the momentum and position wave functions are Fourier transform pairs, up to a factor of the Planck constant. With this constant properly taken into account, the inequality above becomes the statement of the Heisenberg uncertainty principle. A stronger uncertainty principle is the Hirschman uncertainty principle, which is expressed as: H\left(\left|f\right|^2\right)+H\left(\left|\widehat{f}\right|^2\right)\ge \log\left(\frac{e}{2}\right) where is the differential entropy of the probability density function : H(p) = -\int_{-\infty}^\infty p(x)\log\bigl(p(x)\bigr) \, dx where the logarithms may be in any base that is consistent. The equality is attained for a Gaussian, as in the previous case. f(t) = \int_0^\infty \bigl( a(\lambda ) \cos( 2\pi \lambda t) + b(\lambda ) \sin( 2\pi \lambda t)\bigr) \, d\lambda. This is called an expansion as a trigonometric integral, or a Fourier integral expansion. The coefficient functions and can be found by using variants of the Fourier cosine transform and the Fourier sine transform (the normalisations are, again, not standardised): a (\lambda) = 2\int_{-\infty}^\infty f(t) \cos(2\pi\lambda t) \, dt and b (\lambda) = 2\int_{-\infty}^\infty f(t) \sin(2\pi\lambda t) \, dt. Older literature refers to the two transform functions, the Fourier cosine transform, , and the Fourier sine transform, . The function can be recovered from the sine and cosine transform using f(t) = 2\int_0 ^{\infty} \int_{-\infty}^{\infty} f(\tau) \cos\bigl( 2\pi \lambda(\tau-t)\bigr) \, d\tau \, d\lambda. together with trigonometric identities. This is referred to as Fourier's integral formula. \widehat{f}(\xi)=F_0(|\xi|)P(\xi) where F_0(r) = 2\pi i^{-k}r^{-\frac{n+2k-2}{2}} \int_0^\infty f_0(s)J_\frac{n+2k-2}{2}(2\pi rs)s^\frac{n+2k}{2}\,ds. Here denotes the Bessel function of the first kind with order . When this gives a useful formula for the Fourier transform of a radial function. This is essentially the Hankel transform. Moreover, there is a simple recursion relating the cases and allowing to compute, e.g., the three-dimensional Fourier transform of a radial function from the one-dimensional one. f_R(x) = \int_{E_R}\widehat{f}(\xi) e^{i 2\pi x\cdot\xi}\, d\xi, \quad x \in \mathbb{R}^n. Suppose in addition that . For and , if one takes , then converges to in as tends to infinity, by the boundedness of the Hilbert transform. Naively one may hope the same holds true for . In the case that is taken to be a cube with side length , then convergence still holds. Another natural candidate is the Euclidean ball {{math|E R {ξ : \mathcal{F}:L^1(\mathbb{R}^n) \to L^\infty(\mathbb{R}^n) is given by f(x)\mapsto \widehat{f}(\xi) = \int_{\mathbb{R}^n} f(x)e^{-i 2\pi \xi\cdot x}\,dx, \quad \forall \xi \in \mathbb{R}^n. This operator is bounded as \sup_{\xi \in \mathbb{R}^n}\left\vert\widehat{f}(\xi)\right\vert \leq \int_{\mathbb{R}^n} \vert f(x)\vert \,dx, which shows that its operator norm is bounded by . The Riemann–Lebesgue lemma shows that if f\in L^1(\mathbb{R}^n) then its Fourier transform actually belongs to the space of continuous functions that vanish at infinity, i.e., . Furthermore, the image of L^1 under \mathcal{F} is a strict subset of . Similarly to the case of one variable, the Fourier transform can be defined on . The Fourier transform in L^2(\mathbb R^n) is no longer given by an ordinary Lebesgue integral, although it can be computed by an improper integral, i.e., \widehat{f}(\xi) = \lim_{R\to\infty}\int_ \int_{[0,1)}f(y)\overline{g}(y)d\mu(y) with the normalizing factor . The sequence \{e_{k}\mid k\in Z\} is an orthonormal basis of the space of class functions . For any representation of a finite group , \chi_{v} can be expressed as the span \sum_{i} \left\langle \chi_{v},\chi_{v_{i}} \right\rangle \chi_{v_{i}} ( V_{i} are the irreducible representations of ), such that {{tmath|1= \textstyle \left\langle \chi_{v}, \chi_{v_{i} } \right\rangle = \frac{1}{\vert G \vert}\sum_{g\in G}\chi_{v}(g)\overline{\chi}_{v_{i} }(g) }}. Similarly for G = T and , . The Pontriagin dual \widehat{T} is \{e_{k}\}(k\in Z) and for , \widehat{f}(k) = \frac{1} \int_{[0,1)}f(y)e^{-i 2\pi ky}dy is its Fourier transform for . f^*(g) = \overline{f\left(g^{-1}\right)}. Taking the completion with respect to the largest possible -norm gives its enveloping -algebra, called the group -algebra of . (Any -norm on is bounded by the norm, therefore their supremum exists.) Given any abelian -algebra , the Gelfand transform gives an isomorphism between and , where is the multiplicative linear functionals, i.e. one-dimensional representations, on with the weak-* topology. The map is simply given by a \mapsto \bigl( \varphi \mapsto \varphi(a) \bigr). It turns out that the multiplicative linear functionals of , after suitable identification, are exactly the characters of , and the Gelfand transform, when restricted to the dense subset , is the Fourier–Pontryagin transform. \widehat f(\sigma)=\int_G f(g) U^{(\sigma)}_{g^{-1}}\,d\lambda(g). Equivalently, \left\langle \widehat f(\sigma)\xi,\eta\right\rangle_{H_\sigma} = \int_G f(g)\left\langle U^{(\sigma)}_{g^{-1}}\xi,\eta\right\rangle\,d\lambda(g). Since is unitary, this may also be written using the adjoint {U^{(\sigma)}}_g^* . If is a finite complex Borel measure on , then the Fourier–Stieltjes transform of is the operator on defined by \widehat\mu(\sigma)=\int_G U^{(\sigma)}_{g^{-1}}\,d\mu(g), or, weakly, \left\langle \widehat{\mu}(\sigma)\xi,\eta\right\rangle_{H_\sigma} = \int_G \left\langle U^{(\sigma)}_{g^{-1}}\xi,\eta\right\rangle\,d\mu(g). If is absolutely continuous with respect to , represented as d\mu = f\,d\lambda for some , one identifies the Fourier transform of with the Fourier–Stieltjes transform of . The mapping \mu\mapsto\widehat{\mu} is injective and sends finite measures to bounded fields of operators , with \sup_{\sigma\in\Sigma}\|\widehat\mu(\sigma)\|\leq \|\mu\|. Thus it may be viewed as a representation of the Banach algebra of finite Borel measures, with multiplication given by convolution of measures. With the convention above, convolution corresponds to operator multiplication with the order reversed: \widehat{\mu * \nu}(\sigma)=\widehat\nu(\sigma)\widehat\mu(\sigma). Using the alternative convention reverses this order. The involution on is given, for absolutely continuous measures, by f^*(g)=\overline{f(g^{-1})}, since compact groups are unimodular. The Peter–Weyl theorem holds, and a version of the Fourier inversion formula follows: if , then f(g)=\sum_{\sigma\in\Sigma} d_\sigma \operatorname{tr}\left(\widehat f(\sigma)U^{(\sigma)}_g\right), where the summation is understood as convergent in the sense. The corresponding Plancherel formula is \|f\|_2^2 = \sum_{\sigma\in\Sigma} d_\sigma \|\widehat f(\sigma)\|_{\mathrm{HS}}^2, where denotes the Hilbert–Schmidt norm. The generalization of the Fourier transform to the noncommutative situation has also in part contributed to the development of noncommutative geometry. In this context, a categorical generalization of the Fourier transform to noncommutative groups is Tannaka–Krein duality, which replaces the group of characters with the category of representations. However, this is no longer simply a transform of scalar-valued functions into scalar-valued functions. However, when you try to measure a frequency that is not present, both the real and imaginary component of the integral vary rapidly between positive and negative values. For instance, the red curve is looking for 5 Hz. The absolute value of its integral is nearly zero, indicating that almost no 5 Hz component was in the signal. The general situation is usually more complicated than this, but heuristically this is how the Fourier transform measures how much of an individual frequency is present in a function . File:Offfreq i2p.svg| Real and imaginary parts of the integrand for its Fourier transform at +5 Hz. File:Fourier transform of oscillating function.svg| Magnitude of its Fourier transform, with +3 and +5 Hz labeled. To re-enforce an earlier point, the reason for the response at \xi=-3 Hz is because \cos(2\pi 3t) and \cos(2\pi(-3)t) are indistinguishable. The transform of e^{i2\pi 3t}\cdot e^{-\pi t^2} would have just one response, whose amplitude is the integral of the smooth envelope: , whereas \operatorname{Re}(f(t)\cdot e^{-i2\pi 3t}) is . \frac{\partial^2 y(x, t)}{\partial^2 x} = \frac{\partial y(x, t)}{\partial t}. The example we will give, a slightly more difficult one, is the wave equation in one dimension, \frac{\partial^2y(x, t)}{\partial^2 x} = \frac{\partial^2y(x, t)}{\partial^2t}. As usual, the problem is not to find a solution: there are infinitely many. The problem is that of the so-called "boundary problem": find a solution that satisfies the 'boundary conditions' y(x, 0) = f(x),\qquad \frac{\partial y(x, 0)}{\partial t} = g(x). Here, and are given functions. For the heat equation, only one boundary condition can be required (usually the first one). But for the wave equation, there are still infinitely many solutions that satisfy the first boundary condition. But when one imposes both conditions, there is only one possible solution. It is easier to find the Fourier transform of the solution than to find the solution directly. This is because the Fourier transformation takes differentiation into multiplication by the Fourier-dual variable, and so a partial differential equation applied to the original function is transformed into multiplication by polynomial functions of the dual variables applied to the transformed function. After is determined, we can apply the inverse Fourier transformation to find . Fourier's method is as follows. First, note that any function of the forms \cos\bigl(2\pi\xi(x\pm t)\bigr) \text{ or } \sin\bigl(2\pi\xi(x \pm t)\bigr) satisfies the wave equation. These are called the elementary solutions. Second, note that therefore any integral \begin{align} y(x, t) = \int_{0}^{\infty} d\xi \Bigl[ &a_+(\xi)\cos\bigl(2\pi\xi(x + t)\bigr) + a_-(\xi)\cos\bigl(2\pi\xi(x - t)\bigr) +{} \\ &b_+(\xi)\sin\bigl(2\pi\xi(x + t)\bigr) + b_-(\xi)\sin\left(2\pi\xi(x - t)\right) \Bigr] \end{align} satisfies the wave equation for arbitrary . This integral may be interpreted as a continuous linear combination of solutions for the linear equation. Now this resembles the formula for the Fourier synthesis of a function. In fact, this is the real inverse Fourier transform of and in the variable . The third step is to examine how to find the specific unknown coefficient functions and that will lead to satisfying the boundary conditions. We are interested in the values of these solutions at . So we will set . Assuming that the conditions needed for Fourier inversion are satisfied, we can then find the Fourier sine and cosine transforms (in the variable ) of both sides and obtain 2\int_{-\infty}^\infty y(x,0) \cos(2\pi\xi x) \, dx = a_+ + a_- and 2\int_{-\infty}^\infty y(x,0) \sin(2\pi\xi x) \, dx = b_+ + b_-. Similarly, taking the derivative of with respect to and then applying the Fourier sine and cosine transformations yields 2\int_{-\infty}^\infty \frac{\partial y(u,0)}{\partial t} \sin (2\pi\xi x) \, dx = (2\pi\xi)\left(-a_+ + a_-\right) and 2\int_{-\infty}^\infty \frac{\partial y(u,0)}{\partial t} \cos (2\pi\xi x) \, dx = (2\pi\xi)\left(b_+ - b_-\right). These are four linear equations for the four unknowns and , in terms of the Fourier sine and cosine transforms of the boundary conditions, which are easily solved by elementary algebra, provided that these transforms can be found. In summary, we chose a set of elementary solutions, parametrized by , of which the general solution would be a (continuous) linear combination in the form of an integral over the parameter . But this integral was in the form of a Fourier integral. The next step was to express the boundary conditions in terms of these integrals, and set them equal to the given functions and . But these expressions also took the form of a Fourier integral because of the properties of the Fourier transform of a derivative. The last step was to exploit Fourier inversion by applying the Fourier transformation to both sides, thus obtaining expressions for the coefficient functions and in terms of the given boundary conditions and . From a higher point of view, Fourier's procedure can be reformulated more conceptually. Since there are two variables, we will use the Fourier transformation in both and rather than operate as Fourier did, who only transformed in the spatial variables. Note that must be considered in the sense of a distribution since is not going to be : as a wave, it will persist through time and thus is not a transient phenomenon. But it will be bounded and so its Fourier transform can be defined as a distribution. The operational properties of the Fourier transformation that are relevant to this equation are that it takes differentiation in to multiplication by and differentiation with respect to to multiplication by where is the frequency. Then the wave equation becomes an algebraic equation in : \xi^2 \widehat y (\xi, f) = f^2 \widehat y (\xi, f). This is equivalent to requiring unless . Right away, this explains why the choice of elementary solutions we made earlier worked so well: obviously will be solutions. Applying Fourier inversion to these delta functions, we obtain the elementary solutions we picked earlier. But from the higher point of view, one does not pick elementary solutions, but rather considers the space of all distributions that are supported on the (degenerate) conic . We may as well consider the distributions supported on the conic that are given by distributions of one variable on the line plus distributions on the line as follows: if is any test function, \iint \widehat y \varphi(\xi,f) \, d\xi \, df = \int s_+ \varphi(\xi,\xi) \, d\xi + \int s_- \varphi(\xi,-\xi) \, d\xi, where , and , are distributions of one variable. Then Fourier inversion gives, for the boundary conditions, something very similar to what we had more concretely above (put , which is clearly of polynomial growth): y(x,0) = \int\bigl\{s_+(\xi) + s_-(\xi)\bigr\} e^{i 2\pi \xi x+0} \, d\xi and \frac{\partial y(x,0)}{\partial t} = \int\bigl\{s_+(\xi) - s_-(\xi)\bigr\} i 2\pi \xi e^{i 2\pi\xi x+0} \, d\xi. Now, as before, applying the one-variable Fourier transformation in the variable to these functions of yields two equations in the two unknown distributions (which can be taken to be ordinary functions if the boundary conditions are or ). From a calculational point of view, the drawback of course is that one must first calculate the Fourier transforms of the boundary conditions, then assemble the solution from these, and then calculate an inverse Fourier transform. Closed form formulas are rare, except when there is some geometric symmetry that can be exploited, and the numerical calculations are difficult because of the oscillatory nature of the integrals, which makes convergence slow and hard to estimate. For practical calculations, other methods are often used. \varphi(p) = \int dq\, \psi (q) e^{-i pq/h} , or, equivalently, \psi(q) = \int dp \, \varphi (p) e^{i pq/h}. Physically realisable states are , and so by the Plancherel theorem, their Fourier transforms are also . (Note that since is in units of distance and is in units of momentum, the presence of the Planck constant in the exponent makes the exponent dimensionless, as it should be.) Therefore, the Fourier transform can be used to pass from one way of representing the state of the particle, by a wave function of position, to another way of representing the state of the particle: by a wave function of momentum. Infinitely many different polarisations are possible, and all are equally valid. Being able to transform states from one representation to another by the Fourier transform is not only convenient but also the underlying reason of the Heisenberg uncertainty principle. The other use of the Fourier transform in both quantum mechanics and quantum field theory is to solve the applicable wave equation. In non-relativistic quantum mechanics, the Schrödinger equation for a time-varying wave function in one-dimension, not subject to external forces, is -\frac{\partial^2}{\partial x^2} \psi(x,t) = i \frac h{2\pi} \frac{\partial}{\partial t} \psi(x,t). This is the same as the heat equation except for the presence of the imaginary unit . Fourier methods can be used to solve this equation. In the presence of a potential, given by the potential energy function , the equation becomes -\frac{\partial^2}{\partial x^2} \psi(x,t) + V(x)\psi(x,t) = i \frac h{2\pi} \frac{\partial}{\partial t} \psi(x,t). The "elementary solutions", as we referred to them above, are the so-called "stationary states" of the particle, and Fourier's algorithm, as described above, can still be used to solve the boundary value problem of the future evolution of given its values for . Neither of these approaches is of much practical use in quantum mechanics. Boundary value problems and the time-evolution of the wave function is not of much practical interest: it is the stationary states that are most important. In relativistic quantum mechanics, the Schrödinger equation becomes a wave equation as was usual in classical physics, except that complex-valued waves are considered. A simple example, in the absence of interactions with other particles or fields, is the free one-dimensional Klein–Gordon–Schrödinger–Fock equation, this time in dimensionless units, \left (\frac{\partial^2}{\partial x^2} +1 \right) \psi(x,t) = \frac{\partial^2}{\partial t^2} \psi(x,t). This is, from the mathematical point of view, the same as the wave equation of classical physics solved above (but with a complex-valued wave, which makes no difference in the methods). This is of great use in quantum field theory: each separate Fourier component of a wave can be treated as a separate harmonic oscillator and then quantized, a procedure known as "second quantization". Fourier methods have been adapted to also deal with non-trivial interactions. Finally, the number operator of the quantum harmonic oscillator can be interpreted, for example via the Mehler kernel, as the generator of the Fourier transform . R_f (\tau) = \lim_{T\rightarrow \infty} \frac{1}{2T} \int_{-T}^T f(t) f(t+\tau) \, dt. This function is a function of the time-lag elapsing between the values of to be correlated. For most functions that occur in practice, is a bounded even function of the time-lag and for typical noisy signals it turns out to be uniformly continuous with a maximum at . The autocorrelation function, more properly called the autocovariance function unless it is normalized in some appropriate fashion, measures the strength of the correlation between the values of separated by a time lag. This is a way of searching for the correlation of with its own past. It is useful even for other statistical tasks besides the analysis of signals. For example, if represents the temperature at time , one expects a strong correlation with the temperature at a time lag of 24 hours. It possesses a Fourier transform, P_f(\xi) = \int_{-\infty}^\infty R_f (\tau) e^{-i 2\pi \xi\tau} \, d\tau. This Fourier transform is called the power spectral density function of . (Unless all periodic components are first filtered out from , this integral will diverge, but it is easy to filter out such periodicities.) The power spectrum, as indicated by this density function , measures the amount of variance contributed to the data by the frequency . In electrical signals, the variance is proportional to the average power (energy per unit time), and so the power spectrum describes how much the different frequencies contribute to the average power of the signal. This process is called the spectral analysis of time-series and is analogous to the usual analysis of variance of data that is not a time-series (ANOVA). Knowledge of which frequencies are "important" in this sense is crucial for the proper design of filters and for the proper evaluation of measuring apparatuses. It can also be useful for the scientific analysis of the phenomena responsible for producing the data. The power spectrum of a signal can also be approximately measured directly by measuring the average power that remains in a signal after all the frequencies outside a narrow band have been filtered out. Spectral analysis is carried out for visual signals as well. The power spectrum ignores all phase relations, which is good enough for many purposes, but for video signals other types of spectral analysis must also be employed, still using the Fourier transform as a tool. \tilde{f}(\xi),\ F(\xi),\ \mathcal{F}\left(f\right)(\xi),\ \left(\mathcal{F}f\right)(\xi),\ \mathcal{F}(f),\ \mathcal{F}\{f\},\ \mathcal{F} \bigl(f(t)\bigr),\ \mathcal{F} \bigl\{f(t)\bigr\}. In the sciences and engineering it is also common to make substitutions like these: \xi \rightarrow f, \quad x \rightarrow t, \quad f \rightarrow x,\quad \widehat f \rightarrow X. So the transform pair f(x)\ \stackrel{\mathcal{F}}{\Longleftrightarrow}\ \widehat{f}(\xi) can become x(t)\ \stackrel{\mathcal{F}}{\Longleftrightarrow}\ X(f) A disadvantage of the capital letter notation is when expressing a transform such as \widehat f\cdot g or , which become the more awkward \mathcal{F}\{f\cdot g\} and . In some contexts such as particle physics, the same symbol f may be used for both for a function as well as it Fourier transform, with the two only distinguished by their argument I.e. f(k_1 + k_2) would refer to the Fourier transform because of the momentum argument, while f(x_0 + \pi \vec r) would refer to the original function because of the positional argument. Although tildes may be used as in \tilde{f} to indicate Fourier transforms, tildes may also be used to indicate a modification of a quantity with a more Lorentz invariant form, such as , so care must be taken. Similarly, \widehat f often denotes the Hilbert transform of . The interpretation of the complex function may be aided by expressing it in polar coordinate form \widehat f(\xi) = A(\xi) e^{i\varphi(\xi)} in terms of the two real functions and where: A(\xi) = \left|\widehat f(\xi)\right|, is the amplitude and \varphi (\xi) = \arg \left( \widehat f(\xi) \right), is the phase (see Arg). Then the inverse transform can be written: f(x) = \int _{-\infty}^\infty A(\xi)\ e^{ i\bigl(2\pi \xi x +\varphi (\xi)\bigr)}\,d\xi, which is a recombination of all the frequency components of . Each component is a complex sinusoid of the form whose amplitude is and whose initial phase angle (at ) is . The Fourier transform may be thought of as a mapping on function spaces. This mapping is here denoted and is used to denote the Fourier transform of the function . This mapping is linear, which means that can also be seen as a linear transformation on the function space and implies that the standard notation in linear algebra of applying a linear transformation to a vector (here the function ) can be used to write instead of . Since the result of applying the Fourier transform is again a function, we can be interested in the value of this function evaluated at the value for its variable, and this is denoted either as or as . Notice that in the former case, it is implicitly understood that is applied first to and then the resulting function is evaluated at , not the other way around. In mathematics and various applied sciences, it is often necessary to distinguish between a function and the value of when its variable equals , denoted . This means that a notation like formally can be interpreted as the Fourier transform of the values of at . Despite this flaw, the previous notation appears frequently, often when a particular function or a function of a particular variable is to be transformed. For example, \mathcal F\bigl( \operatorname{rect}(x) \bigr) = \operatorname{sinc}(\xi) is sometimes used to express that the Fourier transform of a rectangular function is a sinc function, or \mathcal F\bigl(f(x + x_0)\bigr) = \mathcal F\bigl(f(x)\bigr)\, e^{i 2\pi x_0 \xi} is used to express the shift property of the Fourier transform. Notice, that the last example is only correct under the assumption that the transformed function is a function of , not of . As discussed above, the characteristic function of a random variable is the same as the Fourier–Stieltjes transform of its distribution measure, but in this context it is typical to take a different convention for the constants. Typically characteristic function is defined E\left(e^{it\cdot X}\right)=\int e^{it\cdot x} \, d\mu_X(x). As in the case of the "non-unitary angular frequency" convention above, the factor of 2 appears in neither the normalizing constant nor the exponent. Unlike any of the conventions appearing above, this convention takes the opposite sign in the exponent. |Scaling in the time domain. If is large, then is concentrated around and \frac{1} \widehat{f} \left( \frac{\omega}{a} \right) spreads out and flattens. |- | 105 | \widehat f_n(x) | \widehat f_1(x) \ \stackrel{\mathcal{F}_1}{\longleftrightarrow}\ f(-\xi) | \widehat f_2(x) \ \stackrel{\mathcal{F}_2}{\longleftrightarrow}\ f(-\omega) | \widehat f_3(x) \ \stackrel{\mathcal{F}_3}{\longleftrightarrow}\ 2\pi f(-\omega) |The same transform is applied twice, but replaces the frequency variable ( or ) after the first transform. |- | 106 | \frac{d^n f(x)}{dx^n} | (i 2\pi \xi)^n \widehat{f}(\xi) | (i\omega)^n \widehat{f}(\omega) | (i\omega)^n \widehat{f}(\omega) | th-order derivative. As is a Schwartz function |- |106.5 | \int_{-\infty}^{x} f(\tau) d \tau | \frac{\widehat{f}(\xi)}{i 2 \pi \xi} + C \, \delta(\xi) | \frac{\widehat{f} (\omega)}{i\omega} + \sqrt{2 \pi} C \delta(\omega) | \frac{\widehat{f} (\omega)}{i\omega} + 2 \pi C \delta(\omega) |Integration. Note: \delta is the Dirac delta function and C is the average (DC) value of f(x) such that \int_{-\infty}^\infty (f(x) - C) \, dx = 0 |- | 107 | x^n f(x) | \left (\frac{i}{2\pi}\right)^n \frac{d^n \widehat{f}(\xi)}{d\xi^n} | i^n \frac{d^n \widehat{f}(\omega)}{d\omega^n} | i^n \frac{d^n \widehat{f}(\omega)}{d\omega^n} |This is the dual of 106 |- | 108 | (f * g)(x) | \widehat{f}(\xi) \widehat{g}(\xi) | \sqrt{2\pi}\ \widehat{f}(\omega) \widehat{g}(\omega) | \widehat{f}(\omega) \widehat{g}(\omega) |The notation denotes the convolution of and – this rule is the convolution theorem |- | 109 | f(x) g(x) | \left(\widehat{f} * \widehat{g}\right)(\xi) | \frac{1}\sqrt{2\pi}\left(\widehat{f} * \widehat{g}\right)(\omega) | \frac{1}{2\pi}\left(\widehat{f} * \widehat{g}\right)(\omega) |This is the dual of 108 |- | 110 |For purely real | \widehat{f}(-\xi) = \overline{\widehat{f}(\xi)} | \widehat{f}(-\omega) = \overline{\widehat{f}(\omega)} | \widehat{f}(-\omega) = \overline{\widehat{f}(\omega)} |Hermitian symmetry. indicates the complex conjugate. |- | 113 |For purely imaginary | \widehat{f}(-\xi) = -\overline{\widehat{f}(\xi)} | \widehat{f}(-\omega) = -\overline{\widehat{f}(\omega)} | \widehat{f}(-\omega) = -\overline{\widehat{f}(\omega)} | indicates the complex conjugate. |- | 114 | \overline{f(x)} || \overline{\widehat{f}(-\xi)} || \overline{\widehat{f}(-\omega)} || \overline{\widehat{f}(-\omega)} || Complex conjugation, generalization of 110 and 113 |- |115 | f(x) \cos (a x) | \frac{ \widehat{f}{\left(\xi - \frac{a}{2\pi}\right)} + \widehat{f}{\left(\xi+\frac{a}{2\pi}\right)}}{2} | \frac{\widehat{f}(\omega-a)+\widehat{f}(\omega+a)}{2} | \frac{\widehat{f}(\omega-a)+\widehat{f}(\omega+a)}{2} |This follows from rules 101 and 103 using Euler's formula: {{tmath|1= \cos(a x) = \tfrac{e^{i a x} + e^{-i a x} }{2} }}. |- |116 | f(x)\sin( ax) | \frac{\widehat{f}{\left(\xi-\frac{a}{2\pi}\right)} - \widehat{f}{\left(\xi+\frac{a}{2\pi}\right)}}{2i} | \frac{\widehat{f}(\omega-a)-\widehat{f}(\omega+a)}{2i} | \frac{\widehat{f}(\omega-a)-\widehat{f}(\omega+a)}{2i} |This follows from 101 and 103 using Euler's formula: {{tmath|1= \sin(a x) = \tfrac{e^{i a x} - e^{-i a x} }{2i} }}. |} | \sum_{n=-\infty}^{\infty} \delta (x - n T) | \frac{1}{T} \sum_{k=-\infty}^{\infty} \delta{\left( \xi -\frac{k }{T}\right)} | \frac{\sqrt{2\pi }}{T}\sum_{k=-\infty}^{\infty} \delta{\left( \omega -\frac{2\pi k}{T}\right)} | \frac{2\pi}{T}\sum_{k=-\infty}^{\infty} \delta{\left( \omega -\frac{2\pi k}{T}\right)} |This function is known as the Dirac comb function. This result can be derived from 302 and 102, together with the fact that \sum_{n=-\infty}^{\infty} e^{inx} = 2\pi\sum_{k=-\infty}^{\infty} \delta(x+2\pi k) as distributions. |- | 317 | J_0 (x) | \frac{2\, \operatorname{rect}(\pi\xi)}{\sqrt{1 - 4 \pi^2 \xi^2}} | \sqrt{\frac{2}{\pi}} \, \frac{\operatorname{rect}\left( \frac{\omega}{2} \right)}{\sqrt{1 - \omega^2}} | \frac{2\,\operatorname{rect}\left(\frac{\omega}{2} \right)}{\sqrt{1 - \omega^2}} | The function is the zeroth order Bessel function of first kind. |- | 318 | J_n (x) | \frac{2 (-i)^n T_n (2 \pi \xi) \operatorname{rect}(\pi \xi)}{\sqrt{1 - 4 \pi^2 \xi^2}} | \sqrt{\frac{2}{\pi}} \frac{ (-i)^n T_n (\omega) \operatorname{rect} \left( \frac{\omega}{2} \right)}{\sqrt{1 - \omega^2}} | \frac{2(-i)^n T_n (\omega) \operatorname{rect} \left( \frac{\omega}{2} \right)}{\sqrt{1 - \omega^2}} | This is a generalization of 317. The function is the th order Bessel function of first kind. The function is the Chebyshev polynomial of the first kind. |- | 319 | \log \left| x \right| | -\frac{1}{2} \frac{1}{\left| \xi \right|} - \gamma \delta \left( \xi \right) | -\frac{\sqrt\frac{\pi}{2}}{\left| \omega \right|} - \sqrt{2 \pi} \gamma \delta \left( \omega \right) | -\frac{\pi}{\left| \omega \right|} - 2 \pi \gamma \delta \left( \omega \right) | is the Euler–Mascheroni constant. It is necessary to use a finite part integral when testing or against Schwartz functions. The details of this might change the coefficient of the delta function. |- | 320 | \left( \mp ix \right)^{-\alpha} | \frac{\left(2\pi\right)^\alpha}{\Gamma\left(\alpha\right)}u\left(\pm \xi \right)\left(\pm \xi \right)^{\alpha-1} | \frac{\sqrt{2\pi}}{\Gamma\left(\alpha\right)}u\left(\pm\omega\right)\left(\pm\omega\right)^{\alpha-1} | \frac{2\pi}{\Gamma\left(\alpha\right)}u\left(\pm\omega\right)\left(\pm\omega\right)^{\alpha-1} |This formula is valid for . Use differentiation to derive formula for higher exponents. is the Heaviside function. |} |Here {{tmath|1= \textstyle c_n={\Gamma\left(\frac{n+1}{2}\right)}/{\pi^\frac{n+1}{2} } }}, |}
Text derived from https://en.wikipedia.org/w/index.php?oldid=1377137736 under CC BY-SA 4.0; modifications have been made.