Skip to content

Convolution#

Corresponds to multiplication in the frequncy domain.

x(t)h(t)=x(τ)h(tτ)dτx(t) * h(t) = \int\limits_{-\infty}^{\infty} x(\tau) \cdot h(t-\tau) \diff \tau

with signals x(t)x(t), h(t)h(t).

Discrete: (fg)[n]=k=f[k]g[nk]\((f*g)[n] = \sum\limits_{k = -\infty}^{\infty} {f[k] g[n-k]}\)

Comparison of convolution, cross-correlation, and auto-correlation. From Wikimedia

Convolution of f(t)g(t)f(t) * g(t) showing f(τ)f(\tau) and g(tτ)g(t - \tau)

Properties#

  • Kommutativität: f(t)g(t)=g(t)f(t)f(t)*g(t) = g(t)*f(t)

  • Assoziativität: f(t)(g(t)h(t))=(f(t)g(t))h(t)f(t)*(g(t)*h(t)) = (f(t)*g(t))*h(t)

  • Distributivität: f(t)(g(t)+h(t))=f(t)g(t)+f(t)h(t)f(t)*(g(t) + h(t)) = f(t)*g(t) + f(t)*h(t)

  • Faltung mit Dirac: x(t)δ(tb)=x(tb)x(t)*\delta(t-b) = x(t-b) (Gleiches Signal verschoben)

  • Kausalität: h(tτ)=0h(t - \tau) = 0 für τ>t\tau > t
    h[nl]=0h[n - l] = 0 für l>nl > n

Refrences#