Mastering Fourier Series and Signal Decomposition

Introduction to Fourier Series

For any undergraduate mathematician, the ability to decompose a complex signal into simpler components is a fundamental skill. A Fourier series allows us to represent a periodic function as an infinite sum of sines and cosines. This process, known as signal decomposition, is the cornerstone of harmonic analysis, signal processing, and solving partial differential equations.

Understanding this topic is vital for your exams because it bridges the gap between pure calculus and applied physics. By learning how to calculate Fourier coefficients, you are essentially learning how to identify the 'frequency fingerprint' of a periodic function. This article will guide you through the theory and provide the practical tools needed to solve these problems with confidence.

The Mathematical Definition

A periodic function $f(t)$ with period $T$ can be expressed as a sum of harmonics. The fundamental frequency is defined as $\omega_0 = \frac{2\pi}{T}$. The Fourier series representation is given by:

$$f(t) = a_0 + \sum_{n=1}^{\infty} \left[ a_n \cos(n\omega_0 t) + b_n \sin(n\omega_0 t) \right]$$

Here, $a_0$ represents the average value (DC component) of the signal over one period. The coefficients $a_n$ and $b_n$ determine the amplitude of each harmonic component. These are calculated using the orthogonality of sine and cosine functions over the interval $[0, T]$:

$$a_0 = \frac{1}{T} \int_0^T f(t) dt$$ $$a_n = \frac{2}{T} \int_0^T f(t) \cos(n\omega_0 t) dt$$ $$b_n = \frac{2}{T} \int_0^T f(t) \sin(n\omega_0 t) dt$$

Orthogonal Functions

The power of the Fourier series lies in the concept of orthogonal functions. Two functions $g(t)$ and $h(t)$ are orthogonal on an interval $[a, b]$ if their inner product is zero: $\int_a^b g(t)h(t) dt = 0$.

In the context of Fourier series, the set of functions ${1, \cos(n\omega_0 t), \sin(n\omega_0 t)}$ forms an orthogonal basis. This orthogonality is what allows us to isolate individual coefficients. When we multiply $f(t)$ by $\cos(m\omega_0 t)$ and integrate over the period, all terms in the infinite sum vanish except for the term where $n=m$, effectively 'filtering' out the specific harmonic.

Worked Example 1: Simple Periodic Function

Consider the function $f(t) = 5\sin(t) - 2\cos(3t) + 8\cos(17t)$.

This function is already in the form of a Fourier series. Because the sine and cosine functions are linearly independent, the Fourier series of this function is simply the function itself.

• The fundamental frequency $\omega_0 = 1$.
• The coefficients are: $b_1 = 5$, $a_3 = -2$, $a_{17} = 8$, and all other $a_n, b_n = 0$.

This example demonstrates that if a signal is already composed of pure harmonics, the decomposition is trivial. The challenge in exams usually involves functions that are not already expressed as sums of sinusoids, such as square or triangular waves.

Worked Example 2: Integrating a Square Wave

Let $f(t)$ be a square wave defined as $f(t) = 1$ for $0 < t < \pi$ and $f(t) = -1$ for $-\pi < t < 0$, with period $2\pi$.

1. Find $a_0$: The average value of this symmetric wave is $\frac{1}{2\pi} (\int_{-\pi}^0 -1 dt + \int_0^{\pi} 1 dt) = 0$.
2. Find $a_n$: Since the function is odd, $a_n = 0$ for all $n$.
3. Find $b_n$: $$b_n = \frac{1}{\pi} \int_{-\pi}^{\pi} f(t) \sin(nt) dt = \frac{2}{\pi} \int_0^{\pi} \sin(nt) dt$$ $$b_n = \frac{2}{\pi} \left[ -\frac{\cos(nt)}{n} \right]_0^{\pi} = \frac{2}{n\pi} (1 - \cos(n\pi))$$

If $n$ is even, $\cos(n\pi) = 1$, so $b_n = 0$. If $n$ is odd, $\cos(n\pi) = -1$, so $b_n = \frac{4}{n\pi}$.

Thus, the series is $f(t) = \sum_{n=1,3,5...}^{\infty} \frac{4}{n\pi} \sin(nt)$.

Common Mistakes

• Ignoring Symmetry: Always check if a function is even or odd. Even functions have only cosine terms ($b_n=0$), and odd functions have only sine terms ($a_n=0$). This saves significant time.
• Incorrect Period: Ensure your integration limits match the period $T$. If the period is $2L$, the fundamental frequency is $\pi/L$, not $2\pi/T$.
• Coefficient Normalisation: Students often forget the factor of $2/T$ for $a_n$ and $b_n$, while using $1/T$ for $a_0$. Always double-check your constants.

FAQ

What is the difference between Fourier series and Fourier transform? Fourier series are for periodic signals, resulting in a discrete spectrum of harmonics. Fourier transforms are for non-periodic signals, resulting in a continuous frequency spectrum.

Why do we use harmonics? They are the natural building blocks of periodic systems, representing the simplest oscillations that satisfy the boundary conditions of linear systems.

What is the Gibbs phenomenon? It is the peculiar behaviour of Fourier series near discontinuities, where the partial sums exhibit persistent oscillations (overshoot) that do not disappear as more terms are added.

Conclusion

Fourier series are an essential tool for any mathematician. By mastering the decomposition of periodic functions into harmonics, you gain the ability to analyse complex systems with ease. To see these concepts in motion, visit MathInstructor AI to generate a free, narrated animated lesson on Fourier series tailored to your specific study needs.