Fourier Transform Explained Simply: A Visual Guide

Quick answer: The Fourier transform breaks a signal, such as a sound wave, into the pure sine waves (frequencies) it is made of, and tells you how strong each one is. For example, an ideal three-note chord becomes three spikes at the frequencies of its notes. It is based on Joseph Fourier’s 1807 idea that complicated waves are sums of simple ones. For sampled data, the fast Fourier transform (FFT) algorithm computes the discrete version quickly, which makes it practical in Wi-Fi, MRI scans, audio tools and much more.

When you hear a chord, your ear somehow separates it into individual notes. The Fourier transform does the same thing mathematically: it takes a signal and finds the frequencies hidden inside it. It is one of the most useful tools ever invented, quietly at work in your phone, your music and your doctor’s scanner.

How Simple Waves Represent Complicated Signals

In 1807, Joseph Fourier claimed that any reasonable repeating signal can be written as a sum of sine and cosine waves of different frequencies:

\[ f(t) = a_0 + \sum_{n=1}^{\infty} \left( a_n \cos(n\omega t) + b_n \sin(n\omega t) \right). \]

This is a Fourier series (try our Fourier series calculator). Even a square wave (switching between \(+1\) and \(-1\)), with sharp corners, can be built from smooth sines:

\[ \text{square wave} = \frac{4}{\pi}\left( \sin t + \frac{1}{3}\sin 3t + \frac{1}{5}\sin 5t + \cdots \right). \]

From Time to Frequency

The Fourier transform extends this to many non-repeating signals as well. It converts a signal in time, \(f(t)\), into a description by frequency, \(\hat{f}(\xi)\):

\[ \hat{f}(\xi) = \int_{-\infty}^{\infty} f(t)\, e^{-2\pi i \xi t}\, dt. \]

The factor \(e^{-2\pi i \xi t}\) is a spinning arrow, by Euler’s formula. The integral “winds” the signal around a circle at each frequency \(\xi\) and checks whether it lines up; if the signal contains that frequency, the result is large.

A Chord Example

NoteFrequency (Hz)
C4261.6
E4329.6
G4392.0

Record a C major chord and the sound wave looks messy. An ideal mix of three pure tones would show exactly three peaks at about 262, 330 and 392 Hz (equal temperament, with A4 = 440 Hz). A real piano recording also shows overtones, peaks at multiples of each note’s frequency, which give the piano its sound. Tuning apps and music software read these peaks to identify notes.

The Fast Fourier Transform

Computers work with sampled data. For \(N\) samples they use the discrete Fourier transform (DFT). Computing it directly takes about \(N^2\) steps. In 1965, James Cooley and John Tukey published the fast Fourier transform (FFT), an algorithm that computes exactly the same DFT in about \(N \log N\) steps. It is not a different transform or an approximation; it is a faster way to calculate it. For a million samples, that is the difference between about a trillion and about 20 million operations. (Gauss had found a version around 1805, unpublished in his lifetime.)

Where It Is Used

  • Wi-Fi and 4G/5G (OFDM): an inverse FFT combines many frequency subcarriers at the transmitter, and an FFT separates them at the receiver.
  • MRI: the scanner measures spatial-frequency data, and inverse Fourier methods turn it into an image.
  • Compression: JPEG images use a close cousin, the discrete cosine transform (DCT), and MP3 audio uses a modified DCT inside its filter bank.
  • Audio and science: equalizers, spectrum analyzers, speech processing, X-ray crystallography and earthquake analysis.

Common Misconceptions

“It only works for sound.” No. It applies to many kinds of signals: images, radio waves, measurements over time, even quantum wave functions.

Fourier Series vs Fourier Transform vs FFT

ToolInputOutput
Fourier seriesA repeating signalA list of sine and cosine amounts at whole-number multiples of one frequency
Fourier transformA signal on the whole time lineA continuous spectrum over all frequencies
DFT and FFT\(N\) samples\(N\) frequency values; the FFT is the fast algorithm for the DFT

“Fourier’s claim was accepted immediately.” No. Lagrange and others doubted it, and making it rigorous took decades of work by Dirichlet, Riemann and others.

Watch: The Fourier Transform in Video

If you learn better by watching, these videos are worth your time.

But what is the Fourier Transform? A visual introduction. (3Blue1Brown). The famous winding-machine explanation.
The Most Underrated Algorithm (Veritasium). The story of the fast Fourier transform.

Check Your Understanding

Five quick questions to see how much of this has stuck. Pick an answer and press Submit; you will see a short explanation either way.

Answer each question and press Submit to see your result right away.

1. What does the Fourier transform find in a signal?
Found a bug? Let us know!
2. Who proposed that signals are sums of sine waves, in 1807?
Found a bug? Let us know!
3. About how many steps does the FFT need for N samples?
Found a bug? Let us know!
4. A C major chord shows peaks near which frequencies?
Found a bug? Let us know!
5. Which technology uses a cousin of the Fourier transform for compression?
Found a bug? Let us know!

Frequently Asked Questions

What is the Fourier transform in simple terms?

A mathematical tool that breaks a signal into the pure frequencies it contains and measures how much of each is present.

What is the difference between a Fourier series and a Fourier transform?

A Fourier series writes a repeating signal as a sum of sine waves; the Fourier transform extends the idea to any signal, giving a continuous spectrum of frequencies.

What is the FFT?

The fast Fourier transform, an algorithm published by Cooley and Tukey in 1965. It computes the discrete Fourier transform (DFT) of N samples exactly, in about N log N steps instead of N².

Where is the Fourier transform used?

In audio and image compression, Wi-Fi and mobile networks, MRI scans, music software and many scientific fields.

Keep Exploring

Sources and Further Reading

Related to This Article

What people say about "Fourier Transform Explained Simply: A Visual Guide - Effortless Math"?

No one replied yet.

Leave a Reply