11.3. Properties of the Fourier Transform

F{af(t)+bg(t)}=aF(ω)+bG(ω)

Linearity allows superposition of transformations.

The linearity property of the Fourier Transform states that if you have two functions, f(t) and g(t), and you apply the Fourier Transform to a linear combination of these functions, you can separately compute the transforms of each function and combine them. This means if 'a' and 'b' are constants, then the Fourier Transform of the sum is simply the sum of their individual transforms multiplied by their respective constants. This property simplifies calculations and is particularly useful in signal analysis, where many signals can be considered as the sum of simpler components.

Think of making a fruit smoothie. If you have a banana and some strawberries, and you blend them together, you can think of that smoothie as a linear mix of the flavors from both fruits. If you wanted to know how the smoothie would taste with just bananas or just strawberries, you could separately analyze the taste of each fruit and then combine that knowledge to predict the overall flavor of the smoothie.

F{f(t−t0)}=e−iωt0F(ω)

Shifting a function in time corresponds to a phase shift in frequency.

The time shifting property indicates that if you delay a function f(t) by an amount t0, the Fourier Transform of this delayed function will introduce a phase shift in the frequency domain. Specifically, if you shift f(t) to the right by t0, the resulting transform will include a term e^{-iωt0}. This affects the phase of the frequencies in the transformed function but not their amplitude. This property is crucial in signal processing for understanding how shifts in timing affect the frequency characteristics of signals.

Imagine you are watching a video and you pause it for 5 seconds before hitting play again. The video continues smoothly; however, if you analyze sound waves from that video, that 5-second delay corresponds to a shift in the sound's frequency. Just like how you might expect to hear the sound from the video slightly delayed, the frequencies captured in the Fourier Transform will show a similar phase shift.

F{eiω0tf(t)}=F(ω−ω0)

Multiplication by a complex exponential in time domain shifts the frequency spectrum.

This property states that if a function f(t) is multiplied by a complex exponential term e^{iω0t}, the result is a shift in the frequency representation of the original function. Specifically, the frequency components of F(ω) will be shifted to ω - ω0. This property is often used in modulation, where signals are shifted to different frequencies for transmission without interference.

Consider tuning a radio. If you want to listen to a different frequency, you change the dial, which effectively shifts the waves you're picking up. This is similar to how multiplying your signal by a complex exponential changes its frequency content, allowing you to capture it at different 'channels.'

F{f(at)}=F(ω/|a|)/|a|

Scaling the time domain compresses or stretches the frequency domain.

The time scaling property explains how modifying the time variable of a function affects its frequency domain representation. If the time variable is scaled by a factor 'a', then in the frequency domain, the frequencies are scaled by 1/|a| and the amplitude of the transformed function changes proportionally. If 'a' is greater than 1, it compresses the time function, which stretches the frequency representation, and if 'a' is less than 1, it stretches the time function, compressing the frequency representation.

Imagine you are watching a movie and you speed it up. The scenes change more quickly (compressed time), and the pitch of the audio might sound higher (stretched frequency). Conversely, slowing down a track makes sounds lower in pitch and stretches the duration between scenes.

F{d^nf(t)/dt^n}=(iω)^nF(ω)

This is particularly useful in solving differential equations.

This property tells us that taking the n-th derivative of a time-domain function f(t) corresponds to multiplying its Fourier Transform by (iω)^n. This means the operation of differentiation in the time domain transforms into a simple multiplication in the frequency domain. This property is invaluable in solving differential equations because it simplifies the mathematics, transforming complex differential equations into algebraic equations in the frequency domain.

Think of this property like tuning up a car engine. When you tune it properly, you enhance the performance—making computations about the system easier. Here differentiating the function is akin to tuning the equation, making it easy to handle in the frequency domain where operations become simpler.

If f(t)*g(t) is the convolution of f and g, then: F{f * g}=F(ω)·G(ω)

This transforms convolution in the time domain into multiplication in the frequency domain.

The Convolution Theorem states that the convolution of two functions in the time domain corresponds to the multiplication of their Fourier Transforms in the frequency domain. Convolution is an operation that combines two signals to form a third signal, and this property significantly simplifies calculations by allowing analysis in the frequency domain rather than the time domain.

Imagine making a layered cake. Each layer represents a signal, and when you combine them (convolution), you get a new cake. If you wanted to analyze how delicious it is (the performance), you might instead study the individual layers (their frequency transforms) and multiply their qualities together, which is easier than analyzing the whole cake at once.

Z ∞ 1 Z ∞ |f(t)|^2 dt = |F(ω)|^2 dω 2π −∞ −∞

It preserves the energy of a signal across time and frequency domains.

Parseval’s theorem provides an important link between a function and its Fourier Transform by stating that the integral of the square of a function over time is equal to the integral of the square of its transform over frequency. This theorem shows that energy (or power) in a signal is preserved when switching from the time domain to the frequency domain, emphasizing that both domains provide valuable representation of the same signal.

Consider a flashlight shining on a wall. The intensity and area of light on the wall represent the energy of the light signal over time, and if you switch to a spectrum of colors (frequency domain), the total brightness remains constant. Parseval’s theorem showcases that whether you analyze the light intensity or its frequency color distribution, you get the same measure of energy.

--