The Fourier transform of the step function is a classic example of how generalized functions (distributions) like the delta function allow us to include non-convergent but physically meaningful signals into the frequency domain framework.
At first glance, finding its Fourier transform seems impossible. The Fourier transform of a function ( f(t) ) is:
The Fourier transform of ( \textsgn(t) ) is ( 2/(i\omega) ) (without a delta, since its average is zero). Thus: fourier transform step function
[ \int_0^\infty e^-\alpha t e^-i\omega t dt = \int_0^\infty e^-(\alpha + i\omega) t dt = \frac1\alpha + i\omega ]
[ u(t) = \frac12 + \frac12 \textsgn(t) ] The Fourier transform of the step function is
(its value at ( t=0 ) is often set to ( 1/2 ) for Fourier work), it represents an idealized switch that turns “on” at time zero and stays on forever.
[ \lim_\alpha \to 0^+ \frac1\alpha + i\omega = \frac1i\omega ] Thus: [ \int_0^\infty e^-\alpha t e^-i\omega t dt
The unit step function, often denoted ( u(t) ), is one of the most fundamental, yet mathematically troublesome, signals in engineering and physics. Defined as: