Fourier Series

Signals

Eigen stuff

Say we have an LTI system.

Given an input \(x(t) = \phi(t)\), If the output \(y(t) = k\ \phi(t)\), where \(k\) is a constant, then \(\phi(t)\) is an eigen-function.

The constant \(k\) is the eigen-value. It can be understood as some amplitude factor.

In the context of LTI signals, we know \(y(t) = x(t) * h(t)\).

A known eigenfunction for these LTI systems is \(\phi(t) = e^{j \omega t}\), such that:

\[\phi(t) * h(t) = H(\omega) \phi(t)\]

Where

  • \(\phi(t) = e^{j \omega t}\) is an eigen-function
  • \(h(t)\) is the impulse response
  • \(H(\omega) = \int_{-\infty}^{\infty} h(\tau) e^{-j \omega \tau} d\tau\) is the eigen-value, which is a function of the frequency \(\omega\)

We can use this to define the Fourier Series:

Fourier Series definition

\[x(t) = \sum_{k = -\infty}^{\infty} a_k e^{jk \omega t}\]

Where

  • \(k\) is the counter (integer, real number)
  • \(a_k\) is the constant of the k-th element
  • \(\omega\) is the frequency (since this are sines and cosines in disguise)

How do we go from eigenfunctions to here? Of course I’m skipping a lot of stuff, but if you need a fuller version, go read this.

A general way of getting the constants is:

\[a_n = \frac{1}{T} \int_{-T/2}^{T/2} x(t) e^{-jn \omega t} dt\]

Where

  • \(a_n\) is the n-th constant of the Fourier Series expansion of \(x(t)\)
  • \(T\) is fundamental period
  • \(x(t)\) is a signal
  • \(n\) is the index of the Fourier Series expansion
  • \(\omega\) is frequency

Reminder: \(T = \frac{2\pi}{\omega}\)

Discrete time

\[x[n] = \sum_{k = 0}^{N-1} a_k e^{jk\Omega n}\]

Where

  • \(N\) is the period
  • \(\Omega\) is frequency
\[a_k = \frac{1}{N} \sum_{n = 0}^{N-1} x[n] e^{-jk\Omega n}\]

Properties

Periodicity

\[x[n] = x[n + aN]\]

Where

  • \(N\) is the period
  • \(a\) is some real int

Linearity - superposition

Time shifting

\[x[n-m] \rightarrow e^{-j (2\pi / N) km} c_k\]

Where

  • \(m\) is the shift
  • \(c_k\) is the fourier series sum of \(x[n]\)

Problems

Type 1: Find \(a_k\) without the integral

Given:

  • \(x(t)\), containing sin, cos, and constants

Find:

  • The constants of its Fourier Series expansion \(a_k\)

Process

1. Rewrite all sines and cosines in their exponential form:

\[\cos{\omega t} = \frac{1}{2}(e^{j\omega t} + e^{-j\omega t})\] \[\sin{\omega t} = \frac{1}{2j}(e^{j\omega t} - e^{-j\omega t})\]

2. Group the exponentials \(e^{jk\omega t}\) and simplify their coefficients

3. For each \(e^{jk\omega t}\), write the accompanying coefficients to \(a_k\)

Tips:

  • Keep the expansion form in mind
  • \(k\) MUST be of type int
  • Consider the case \(k = 0\) separately (since \(e^0 = 1\), then \(a_0\) will be any left over constants)
  • The sin exponential form is divided by a j, don’t forget it
  • Remember \(e^{a+b} = e^a e^b\)

Type 2: Find \(a_k\) with the integral

Given:

  • \(x(t)\), either a function or graph

Find:

  • The constants of its Fourier Series expansion \(a_k\)

Process

1. If given a graph, rewrite as an equation and stating the limits.

  • The graph may contain the period T as a variable or a number. If it is a number, make sure to replace it both in T and \(\omega\).

2. Consider the integral:

\[a_k = \frac{1}{T} \int_T x(t) e^{-jk \omega t} dt\]

Tips:

  • You should split the integral depending on the limits (if the functions is piecewise)
  • If the case \(k = 0\) is problematic (undefined), re-evaluate the integral only for the case \(k = 0\), so that \(a_0 = \frac{1}{T} \int_T x(t) dt\)