Skip to main content

Energy density spectrum of CT signals

Short answer

Energy density spectrum is S_{xx}(ω)=|X(ω)|², derived from Parseval: E=∫|x(t)|²dt = (1/2π)∫|X(ω)|²dω. The energy in a band dω is (1/2π)|X(ω)|² dω. It is real, nonnegative, even for real signals, and equals the FT of autocorrelation.

Derive and explain the expression of energy density spectrum for a continuous time aperiodic signal.

The energy density spectrum (energy spectral density, ESD) of an aperiodic signal shows how the signal's total energy is spread over frequency. It is defined as ∣X(ω)∣2|X(\omega)|^2.

Derivation

The energy of a CT aperiodic signal is

E=∫−∞∞∣x(t)∣2dt=∫−∞∞x(t)x∗(t) dtE = \int_{-\infty}^{\infty}|x(t)|^2dt = \int_{-\infty}^{\infty}x(t)x^*(t)\,dt

Express x∗(t)x^*(t) through the inverse Fourier transform:

x∗(t)=[12π∫−∞∞X(ω)ejωtdω]∗=12π∫−∞∞X∗(ω)e−jωtdωx^*(t) = \left[\frac{1}{2\pi}\int_{-\infty}^{\infty}X(\omega)e^{j\omega t}d\omega\right]^* = \frac{1}{2\pi}\int_{-\infty}^{\infty}X^*(\omega)e^{-j\omega t}d\omega

Substitute and change the order of integration:

E=12π∫−∞∞X∗(ω)[∫−∞∞x(t)e−jωtdt]dω=12π∫−∞∞X∗(ω)X(ω) dω=12π∫−∞∞∣X(ω)∣2dω\begin{aligned} E &= \frac{1}{2\pi}\int_{-\infty}^{\infty}X^*(\omega)\left[\int_{-\infty}^{\infty}x(t)e^{-j\omega t}dt\right]d\omega \\ &= \frac{1}{2\pi}\int_{-\infty}^{\infty}X^*(\omega)X(\omega)\,d\omega = \frac{1}{2\pi}\int_{-\infty}^{\infty}|X(\omega)|^2d\omega \end{aligned}

This is Parseval's relation. Write it as

E=12π∫−∞∞Sxx(ω) dω,Sxx(ω)=∣X(ω)∣2E = \frac{1}{2\pi}\int_{-\infty}^{\infty}S_{xx}(\omega)\,d\omega, \qquad S_{xx}(\omega) = |X(\omega)|^2

In hertz: E=∫−∞∞∣X(f)∣2dfE = \int_{-\infty}^{\infty}|X(f)|^2df, with ESD ∣X(f)∣2|X(f)|^2 in J/Hz.

Explanation

  • 12π∣X(ω)∣2dω\frac{1}{2\pi}|X(\omega)|^2 d\omega is the energy carried by the frequency components in a small band dωd\omega around ω\omega. The energy between ω1\omega_1 and ω2\omega_2 is therefore 1π∫ω1ω2∣X(ω)∣2dω\frac{1}{\pi}\int_{\omega_1}^{\omega_2}|X(\omega)|^2d\omega for a real signal, counting both the positive and the negative band.
  • The ESD is real, non-negative, and holds no phase information. Signals with different phase but the same ∣X(ω)∣|X(\omega)| have the same ESD.
  • For real x(t)x(t), ∣X(−ω)∣=∣X(ω)∣|X(-\omega)| = |X(\omega)|, so the ESD is even.
  • The ESD is the Fourier transform of the autocorrelation Rxx(τ)=∫x(t)x∗(t−τ)dtR_{xx}(\tau) = \int x(t)x^*(t-\tau)dt (Wiener–Khinchin relation for energy signals).
  • LTI systems: if Y(ω)=H(ω)X(ω)Y(\omega) = H(\omega)X(\omega), then Syy(ω)=∣H(ω)∣2Sxx(ω)S_{yy}(\omega) = |H(\omega)|^2 S_{xx}(\omega).
  • It applies only to energy signals (finite EE). Power signals use the power spectral density instead.

Example

For x(t)=e−atu(t)x(t) = e^{-at}u(t): Sxx(ω)=1a2+ω2S_{xx}(\omega) = \frac{1}{a^2+\omega^2}, and

E=12π⋅πa=12aE = \frac{1}{2\pi}\cdot\frac{\pi}{a} = \frac{1}{2a}

Half of this energy lies in the band ∣ω∣<a|\omega| < a:

12π∫−aadωa2+ω2=12π⋅2atan⁡−1(1)=12π⋅2a⋅π4=14a\frac{1}{2\pi}\int_{-a}^{a}\frac{d\omega}{a^2+\omega^2} = \frac{1}{2\pi}\cdot\frac{2}{a}\tan^{-1}(1) = \frac{1}{2\pi}\cdot\frac{2}{a}\cdot\frac{\pi}{4} = \frac{1}{4a}

✓ Answer checked by Claude Opus 5.5 (high effort)

All Signals and Systems chapter 3 questions →