Foundations — Fourier series — motivation from periodic functions

Before we can find "how much of each wave," we must be fluent in every symbol the parent note throws at you. Below, each idea is built from nothing, drawn as a picture, and justified — why does the topic even need this? Read top to bottom; each rung of the ladder rests on the one before.

0. The plain-language starting point

Imagine a bell that rings and a graph that draws that ringing as a wiggle going left-to-right. If the wiggle keeps repeating the same pattern forever, we call it periodic. Fourier's dream is: take that repeating wiggle and rebuild it out of the simplest possible wiggles — pure smooth waves. To do that we need vocabulary. Let's earn it.

1. The variable $x$ and a function $f(x)$

Picture: the horizontal line is $x$ (where you are); the vertical height is $f(x)$ (how tall the shape is there). Nothing more.

Why the topic needs it: the "shape" we want to rebuild — an initial temperature, a plucked string — is a function $f(x)$. It is the target of the whole construction.

2. Periodic, period $T$, and repeating

Picture (figure above): one coloured tile of width $T$ is stamped over and over. The teal bracket marks exactly one period. Slide the whole graph right by $T$ and it lands perfectly on top of itself.

Why the topic needs it: waves like $\sin$ and $\cos$ repeat forever. So they can only ever build things that also repeat forever. Periodicity is the entry ticket — it's the class of shapes Fourier series is allowed to talk about. See the parent topic for the motivation in Hinglish.

3. Half-period $L$ — why the topic writes $2L$

Picture: put the origin in the middle of one tile. The tile stretches from $-L$ on the left to $+L$ on the right — total width $2L$.

Why the topic needs it: integrating from $-L$ to $L$ (a symmetric window) is what makes the even/odd shortcuts work later. Writing $2L$ instead of $T$ is just bookkeeping so the interval is centred.

4. Angle-speak: radians, $\pi$, and one full turn

Picture: walk once around a circle of radius $1$; the distance you travel is $2\pi$. That travelled distance is the angle in radians.

Why the topic needs it: sine and cosine take an angle as input. To make them repeat every $2L$ instead of every $2\pi$, we must convert "how far along $x$" into "how much angle." That conversion is the next symbol.

5. Sine and cosine as circular motion

Picture (figure above): left panel = the spinning point on the unit circle with its two shadows marked (plum = cosine shadow, orange = sine shadow). Right panel = those shadows unrolled into the familiar wavy curves. Cosine starts high at $+1$; sine starts at $0$ climbing up.

Why the topic needs it: these two curves are the only building blocks Fourier uses. Everything else is amplitudes telling us how tall each block is.

6. Frequency $n$ and angular frequency $\omega$

Picture (figure above): $n=1$ fits one hump-and-dip in the window; $n=2$ fits two; $n=3$ fits three. Every one still repeats exactly at the same tile width $2L$ — that's why they can share a period and be added.

Why the topic needs it: a single wave can't reproduce a sharp corner. We need a whole family of faster and faster waves ($n=1,2,3,\dots$) to add up into jagged shapes. The symbol $\frac{n\pi x}{L}$ is exactly $n\omega x$ written out with $\omega =\pi /L$.

7. The summation sign $\sum$

Picture: a conveyor belt dropping term $n=1$, then $n=2$, then $n=3$ into a bucket, running forever; the bucket's total is the sum.

Why the topic needs it: the Fourier series is an infinite sum of waves. Without $\sum$ we'd have to write "$+\cdots$" by hand every time.

8. The integral sign $\int$ — "area, hence total"

Picture (figure above): the region under one wave is shaded — orange strips above the axis add, teal strips below subtract. For a full number of wave cycles the ups and downs cancel to exactly zero.

Why the topic needs it: the coefficient formulas $a_n=\frac{1}{L}\int f\cos \frac{n\pi x}{L}dx$ are integrals. The "cancels to zero over a full cycle" fact is the seed of orthogonality — the trick that lets us pull out one wave's amplitude at a time. See Orthogonality of Functions.

9. Amplitudes $a_n$ and $b_n$ — the "how much of each wave"

Picture: think of a graphic-equaliser with sliders. Each slider is one $a_n$ or $b_n$; setting all sliders correctly reproduces $f$.

Why the topic needs it: these numbers are the answer Fourier series computes. Everything above exists so we can find them.

10. Even and odd — the symmetry shortcut

Picture: even = butterfly wings mirrored across the $y$-axis. Odd = a pinwheel that looks the same after a half-turn about the origin.

Why the topic needs it: an even shape needs only cosine sliders (all $b_n=0$); an odd shape needs only sine sliders (all $a_n=0$). Half the work vanishes. Full detail in Even and Odd Functions.

How the foundations feed the topic

This map plugs into Separation of Variables, which produces sine/cosine modes for the Heat Equation and Wave Equation; matching a starting shape to those modes is exactly what the coefficients $a_n,b_n$ do. Later refinements live in Gibbs Phenomenon and Fourier Transform.

Equipment checklist

Self-test: can you answer each before reading on? If yes to all, you're ready for the derivation.