Skip to content
Tutorials Center
My learning

Topics

Lesson 24 of 36 · Laplace Transform

Region of Convergence and Laplace Properties

Understand the region of convergence of the Laplace transform, causal and anti-causal signals, linearity, shifts, derivative rules and the initial and final value theorems.

A Laplace transform is not just a formula in s. It also comes with a region of convergence (ROC): the set of values of s for which the integral actually converges. Two different signals can have the same algebraic expression for X(s) and differ only in the ROC, so the ROC is part of the answer. This lesson shows how to read it, then lists the properties you will use to solve problems quickly.

The ROC

For x(t) = e^(-at) u(t) (a right-sided signal), the integral converges when Re(s) > -a. For the left-sided signal -e^(-at) u(-t) the algebraic result is the same, 1/(s + a), but the ROC is Re(s) < -a. The ROC is always a vertical strip (or half-plane) in the s-plane, because convergence depends only on σ = Re(s).

Useful rules:

  • The ROC contains no poles.
  • A right-sided signal has an ROC to the right of its rightmost pole.
  • A left-sided signal has an ROC to the left of its leftmost pole.
  • A two-sided signal has an ROC that is a strip between poles (or empty).
  • A finite-duration signal has an ROC that is the whole plane.
  • If the ROC contains the imaginary axis, the Fourier transform exists and equals X(jω).

For a causal LTI system the ROC is a right half-plane, and the system is stable exactly when that half-plane includes the imaginary axis, meaning all poles lie in the left half-plane. See poles and zeros.

SymPy returns the abscissa of convergence (the boundary of the ROC) when you leave the conditions on.

Python Finding the convergence boundary

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

Program input · 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 growing exponential e^(2t) has the boundary Re(s) > 2: its ROC does not include the imaginary axis, so it has no Fourier transform in the ordinary sense. This matches the intuition that a signal growing without bound cannot be written as a sum of steady sinusoids.

Properties

Let x(t) ↔ X(s) with x(t) = 0 for t < 0.

Property Time domain s-domain
Linearity a x + b y a X + b Y
Time shift (t0 ≥ 0) x(t - t0) u(t - t0) e^(-s t0) X(s)
Frequency shift e^(-at) x(t) X(s + a)
Scaling (a > 0) x(at) (1/a) X(s/a)
Differentiation dx/dt s X(s) - x(0⁻)
Second derivative d²x/dt² s² X(s) - s x(0⁻) - x'(0⁻)
Integration integral from 0 to t X(s) / s
Convolution x * y X(s) Y(s)
Multiplication by t t x(t) -dX/ds

The differentiation rule is the reason Laplace is popular: initial conditions are built in. Solving y' + 3y = 0 with y(0) = 2 becomes sY - 2 + 3Y = 0, so Y = 2/(s + 3) and y(t) = 2 e^(-3t).

Initial and final value theorems

You can read the starting and ending values of a signal from X(s) without inverting it.

  • Initial value: x(0⁺) = lim as s → ∞ of s X(s).
  • Final value: lim as t → ∞ of x(t) = lim as s → 0 of s X(s), valid only if all poles of s X(s) lie in the open left half-plane (so the limit exists).
Python Derivative rule and value theorems

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

Program input · 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.

A time-shift example

A pulse that switches on at t = 1 is u(t - 1). Its transform is e^(-s)/s. In general a delay of t0 multiplies the transform by e^(-s t0). Because that factor is not rational, delays are not captured by finite poles and zeros; control engineers approximate them (a Padé approximation) when they need a rational model.

Python A delayed pulse

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

Program input · 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.

Common mistakes

  • Quoting X(s) without the ROC when the signal could be left-sided or two-sided.
  • Using the final value theorem on a system with poles on the imaginary axis or in the right half-plane; the theorem silently gives a wrong number.
  • Forgetting the initial-condition terms in the derivative rules.

Check your understanding

Quick check · 4 questions

of 4 answered You scored / 4 · perfect!

  1. 1. x(t) = e^(-2t) u(t) has X(s) = 1/(s + 2). What is its ROC?

    A right-sided signal converges to the right of its rightmost pole, which is at s = -2.
  2. 2. What is the Laplace transform of dx/dt for a signal with x(0⁻) = 3?

    Differentiation gives s X(s) - x(0⁻), so initial conditions enter directly.
  3. 3. The transform of x(t - 2) u(t - 2) is...

    A delay of t0 multiplies the transform by e^(-s t0).
  4. 4. When does the final value theorem fail?

    If s X(s) has poles in the right half-plane or on the imaginary axis, x(t) does not settle to a limit and the theorem gives a misleading value.

Summary

  • The ROC is a vertical strip or half-plane; it is part of the definition of the transform.
  • Right-sided signals converge right of their poles, left-sided signals left of them.
  • Key rules: shift in time gives e^(-s t0), shift in s gives e^(-at), differentiation gives s X(s) - x(0⁻), convolution gives a product.
  • Initial and final value theorems read limits straight from X(s), with a stability caveat.

Next: Inverse Laplace and Partial Fractions