Region of Convergence

The region of convergence (ROC) defines the set of complex values of s for which the Laplace transform integral converges to a finite value. Knowing the ROC is not optional — two different signals can have the same algebraic Laplace expression but different ROCs, meaning they are entirely different signals.

Core Concept

The bilateral Laplace transform is X(s) = ∫ x(t)e^(-st)dt where s = σ+jω is a complex variable. The integral only converges for certain values of σ — the real part of s. The ROC is a vertical strip (or half-plane) in the complex s-plane that specifies exactly which values of s make the integral finite.

For a right-sided signal like e^(-at)u(t), the exponential factor e^(-σt) damps the integrand only if σ > -a. The ROC is the half-plane Re(s) > -a, which lies to the right of the pole at s = -a. For a left-sided signal -e^(-at)u(-t), the same X(s) = 1/(s+a) applies, but convergence requires σ < -a, so the ROC is to the left of the pole.

A causal LTI system is BIBO stable if and only if its ROC includes the entire imaginary axis (s = jω). This is equivalent to all poles lying in the left half of the s-plane, meaning Re(pole) < 0. This is the algebraic stability criterion used in control system design.

Key Formula

The bilateral Laplace transform: X(s) = ∫_{-∞}^{∞} x(t)e^(-st)dt. The ROC is the set {s = σ+jω : integral converges}. For rational X(s), the ROC never contains poles. ROC types: right-sided signal → ROC is Re(s) > σ_max (rightmost pole real part); left-sided signal → ROC is Re(s) < σ_min; two-sided signal → ROC is a vertical strip σ₁ < Re(s) < σ₂; finite-duration signal → entire s-plane (possibly excluding s=0 or s=∞).

Problem: Find the Laplace transform and ROC for:
  (a) x(t) = e^(-2t)u(t)  (causal)
  (b) x(t) = -e^(-2t)u(-t) (anti-causal)

Case (a) — causal:
  X(s) = ∫₀^∞ e^(-2t)e^(-st)dt
       = ∫₀^∞ e^(-(s+2)t)dt
       = 1/(s+2)   for Re(s+2) > 0
  ROC: Re(s) > -2  (right half-plane excluding pole at s=-2)

Case (b) — anti-causal:
  X(s) = -∫_{-∞}^0 e^(-2t)e^(-st)dt
       = -∫_{-∞}^0 e^(-(s+2)t)dt
       = 1/(s+2)   for Re(s+2) < 0
  ROC: Re(s) < -2  (left half-plane)

Observation:
  Both have X(s) = 1/(s+2) but different ROCs.
  Case (a) is causal and stable if -2 < 0.
  Case (b) is anti-causal and unstable (ROC excludes jω axis).

Final Answer: ROC determines the signal uniquely alongside X(s).

Exam Tip: Always state the ROC when writing a Laplace transform — the expression X(s) alone is incomplete. BIBO stability of a causal system requires ROC to include the jω-axis, which means all poles must have Re(s) < 0. If a problem gives you X(s) and asks whether the signal is causal and stable, check: all poles in the left half-plane with ROC Re(s) > max(pole real parts) means causal and stable.

Properties Summary

• ROC definition: strip of σ = Re(s) values for which ∫|x(t)e^(-σt)|dt < ∞.
• Right-sided signal: ROC is Re(s) > σ_max, a right half-plane.
• Left-sided signal: ROC is Re(s) < σ_min, a left half-plane.
• Two-sided signal: ROC is a vertical strip, may be empty if σ_max ≥ σ_min.
• Finite-duration signal: ROC is the entire s-plane (except possibly s=0 or s=∞).
• Stability: ROC includes jω-axis ↔ system is BIBO stable.
• ROC never contains poles — poles are always on the boundary or excluded.

Quick Revision

• ROC is a vertical strip or half-plane in the s-plane.
• Same X(s) + different ROC = completely different signal.
• Causal signal → ROC is right half-plane beyond rightmost pole.
• Anti-causal signal → ROC is left half-plane.
• Stable system: all poles Re < 0 and ROC includes jω-axis.
• Finite-duration signals have the largest possible ROC.
• For sum of signals, ROC is intersection of individual ROCs.
• Exam trap: stating X(s) = 1/(s+a) without the ROC — this answer is incomplete and may be marked wrong.