Skip to content
Tutorials Center
My learning

Topics

Lesson 8 of 36 ยท Signals and Classification

Energy and Power Signals

Learn how to compute signal energy and average power, classify signals as energy or power signals, and verify results numerically in Python.

How "big" is a signal? Peak value says how high it goes, but not how long it lasts. Physics gives a better measure. If x(t) is the voltage across a 1 ohm resistor, the instantaneous power is x(t)^2 and the energy delivered is its integral. Signal theory borrows this definition even when the signal is not a voltage, which gives one way to compare signals and decide which transform applies.

Energy

For a continuous-time signal the total energy is

E = integral from -inf to +inf of |x(t)|^2 dt

and in discrete time

E = sum over n of |x[n]|^2.

The magnitude squared makes it work for complex signals too. A signal is an energy signal if 0 < E < infinity.

Average power

For signals that never end, energy is infinite, so we divide by the length of the window. The average power is

P = lim (T -> inf) of (1 / 2T) integral from -T to T of |x(t)|^2 dt

and, in discrete time, P = lim (N -> inf) of (1 / (2N+1)) sum from -N to N of |x[n]|^2.

A signal is a power signal if 0 < P < infinity. For a periodic signal the limit simplifies to an average over one period:

P = (1 / T0) integral over one period of |x(t)|^2 dt.

How to classify

  • Finite energy implies zero average power: an energy signal has P = 0.
  • Finite nonzero power implies infinite energy: a power signal has E = infinity.
  • The two classes do not overlap, but some signals are neither: the ramp t u(t) and e^t grow without bound, so both E and P are infinite.
Signal Class Why
Rectangular pulse energy finite duration
e^(-t) u(t) energy the decay makes the integral converge
cos(w t) power periodic, endless
unit step u(t) power P = 1/2
ramp t u(t) neither grows without bound
Python Energy of a pulse and a decaying exponential

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The numeric integral is a Riemann sum: multiply the sum of squares by the sample spacing dt. For discrete signals no dt is needed.

Power of periodic signals

For x(t) = A cos(w t + phi) the average power is A^2 / 2, independent of frequency and phase. Over a full period, the average of cos^2 is exactly 1/2. For a constant A the power is A^2. The unit step has power 1/2, because over a long window [-T, T] it is 1 half the time.

Python Power of sinusoids and the step

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Sums of signals and cross terms

For x = x1 + x2 the power is P1 + P2 + 2 * average of x1 x2. When the two signals have different frequencies the cross term averages to zero, so powers simply add; this is a preview of orthogonality in the Fourier chapters.

Python Powers add for different frequencies

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Why classification matters

Energy signals (pulses, decaying signals) have a Fourier transform. Power signals (periodic, steady) have a Fourier series or a transform full of impulses. Parseval's theorem, later in this course, states that energy or power measured in time equals the same amount measured in the frequency domain.

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. A rectangular pulse of height 3 and duration 2 s has energy:

    E = integral of 3^2 over 2 s = 9 * 2 = 18.
  2. 2. What is the average power of 4 cos(10 t)?

    P = A^2 / 2 = 16 / 2 = 8.
  3. 3. Which signal is neither an energy signal nor a power signal?

    The ramp grows without bound, so both its energy and its average power are infinite.
  4. 4. An energy signal always has average power:

    Finite total energy spread over an infinite window averages to zero.

Summary

  • Energy is the integral (or sum) of |x|^2; average power is energy per unit time in the limit.
  • Energy signals have 0 < E < inf and P = 0; power signals have 0 < P < inf and E = inf.
  • A sinusoid of amplitude A has power A^2 / 2; the unit step has power 1/2.
  • Powers of sinusoids at different frequencies add because cross terms average to zero.
  • Some signals (ramps, growing exponentials) are neither.

Next: System Properties