Z-Transform GATE ECE Notes + ROC Formulas + PYQs | Signals

Introduction

In real digital electronics, signals are often not available at every instant of time. A microphone signal is sampled, a sensor is read periodically, and a DSP processor works on indexed values such as $$x[n]$$.

Z-Transform gives a clean mathematical language for such discrete-time signals. It plays for sequences what Laplace Transform plays for continuous-time systems: it reveals stability, causality, frequency behavior, and system response.

Why It Matters

Engineering problem solved

Digital filters are described by difference equations, which become simpler algebra in the z-domain.
Discrete convolution becomes multiplication, so LTI system output becomes easier to compute.
Pole locations immediately show whether a digital system response decays, grows, or oscillates.
The unit circle connects Z-Transform with DTFT and digital frequency response.

Exam and interview value

GATE ECE questions often combine ROC, causality, stability, and inverse Z-transform.
University exams test one-sided and two-sided Z-transform definitions.
Technical interviews use Z-Transform to check DSP fundamentals and filter intuition.

Prerequisites

Discrete-time signals such as $$x[n]$$ and $$u[n]$$.
Geometric series and convergence.
Complex numbers and polar form.
Convolution of sequences.
LTI system response and impulse response.
Basic idea of Laplace Transform and Fourier Transform.

Basic Intuition

Think of Z-Transform as a special lens for sequences. A sequence may look like a list of numbers in time, but in the z-plane it becomes a pattern of poles, zeros, and convergence regions.

The variable $$z$$ is complex and is commonly written as $$z=re^{j\omega}$$. The angle $$\omega$$ represents discrete-time frequency, while the radius $$r$$ controls growth or decay. This is why the unit circle $$|z|=1$$ is so important for frequency response.

Z-Transform is not just a formula. It is a map that turns sequence behavior into geometry.

Z-Plane Pole-Zero Diagram Here

Animated Unit Circle and ROC Visualization

Core Theory

Two-sided Z-Transform

$$X(z)=\sum_{n=-\infty}^{\infty}x[n]z^{-n}$$

This definition looks at the entire sequence from negative infinity to positive infinity. It is used heavily in Signals and Systems because it clearly separates right-sided, left-sided, and two-sided sequences using ROC.

One-sided Z-Transform

$$X^{+}(z)=\sum_{n=0}^{\infty}x[n]z^{-n}$$

The one-sided form begins at $$n=0$$ and is useful for solving difference equations with initial conditions in DSP, digital control, and embedded systems.

Region of Convergence

ROC is the set of $$z$$ values for which the summation converges. Without ROC, an expression like $$X(z)=1/(1-az^{-1})$$ is incomplete because it may represent different time sequences.

Working Principle

Z-Transform works by multiplying each sample $$x[n]$$ with $$z^{-n}$$ and adding all weighted samples. This weighting tests how the sequence behaves against complex exponential patterns.

Step 1: Weight samples

Each sample is scaled by $$z^{-n}$$, so earlier and later samples are treated according to their index.

Step 2: Sum the sequence

The weighted samples are accumulated. If the sum stays finite, that value of $$z$$ belongs to the ROC.

Step 3: Read system behavior

Poles, zeros, and ROC reveal causality, stability, transient behavior, and frequency response.

Animated Sequence Weighting and Z-Transform Summation

Formula Explanation

Definition

$$X(z)=\sum_{n=-\infty}^{\infty}x[n]z^{-n}$$

The signal is converted from a time-indexed sequence into a complex-variable function.

Inverse Z-Transform

$$x[n]=\frac{1}{2\pi j}\oint_C X(z)z^{n-1}dz$$

This recovers the original sequence from its z-domain representation.

Convolution property

$$x[n]*h[n]\Longleftrightarrow X(z)H(z)$$

A difficult time-domain summation becomes multiplication in the z-domain.

Frequency response

$$H(e^{j\omega})=H(z)\big|_{z=e^{j\omega}}$$

If the unit circle lies in the ROC, the DTFT exists and the frequency response can be evaluated on $$|z|=1$$.

Diagram Explanation Placeholder

A complete visual explanation should show the z-plane with the real axis, imaginary axis, unit circle, poles, zeros, and shaded ROC. For causal systems, the ROC extends outward from the outermost pole. For stable systems, the ROC must include the unit circle.

Unit Circle, Poles, Zeros, and ROC Diagram Here

Animated ROC Expansion Across the Z-Plane

Frequency Behavior

In digital signal processing, frequency behavior is read on the unit circle. If a pole is close to the unit circle near a particular angle, the system tends to emphasize that frequency. If a zero is near the unit circle, it tends to suppress that frequency.

Z-plane feature	Time-domain meaning	Frequency effect
Pole inside unit circle	Decaying response	May create a peak near its angle
Pole outside unit circle	Growing response	Usually unstable for causal systems
Zero on unit circle	Cancellation at a frequency	Creates a notch
Unit circle in ROC	Absolutely summable impulse response	Frequency response exists

Real-World Applications

Design and analysis of FIR and IIR digital filters.
Solving digital filter difference equations in DSP processors.
Stability checking in digital control systems.
Audio equalizers, noise reduction, and speech processing.
Image processing filters used in cameras and medical imaging.
Embedded sensor processing in IoT and communication receivers.
Modeling sampled-data systems after analog-to-digital conversion.

Solved Examples

Example 1: Z-Transform of $$a^n u[n]$$

For $$x[n]=a^n u[n]$$:

$$X(z)=\sum_{n=0}^{\infty}a^n z^{-n}=\sum_{n=0}^{\infty}(az^{-1})^n$$

$$X(z)=\frac{1}{1-az^{-1}},\quad ROC:\ |z|>|a|$$

The ROC is outside the pole because the sequence is right-sided and convergence needs $$|az^{-1}|<1$$.

Example 2: Stability from ROC

A system has $$H(z)=1/(1-0.5z^{-1})$$ with ROC $$|z|>0.5$$.

$$|z|=1\ \text{lies in ROC}\quad \Rightarrow\quad stable$$

Because the unit circle is included in the ROC, the impulse response is absolutely summable.

Example 3: Difference equation

For the system $$y[n]-0.4y[n-1]=x[n]$$ with zero initial conditions:

$$Y(z)-0.4z^{-1}Y(z)=X(z)$$

$$H(z)=\frac{Y(z)}{X(z)}=\frac{1}{1-0.4z^{-1}}$$

The pole at $$z=0.4$$ lies inside the unit circle, so the causal system is stable.

Common Mistakes

Writing the Z-Transform without mentioning ROC.
Assuming every rational expression represents a causal sequence.
Confusing the z-plane unit circle with the whole ROC.
Forgetting that stability requires the unit circle to be inside the ROC.
Using continuous-time Laplace pole rules directly without adjusting to discrete-time unit-circle rules.
Mixing one-sided and two-sided Z-transform while solving difference equations.

Interview Questions

Why is Z-Transform called the discrete-time counterpart of Laplace Transform?
What information does ROC add beyond the algebraic expression X(z)?
How do you decide stability of a discrete-time LTI system from H(z)?
What is the relation between the unit circle and frequency response?
How does a pole near the unit circle affect the discrete-time response?
Why are difference equations easier to solve in the z-domain?

Exam Notes

For right-sided sequences, ROC is outside the outermost pole.
For left-sided sequences, ROC is inside the innermost pole.
For two-sided sequences, ROC lies between poles.
A causal discrete-time LTI system is stable only if all poles are inside the unit circle.
A stable system has ROC including $$|z|=1$$.
Check whether the problem asks for one-sided or two-sided Z-transform before using initial conditions.

Revision Summary

Z-Transform converts discrete-time sequences into z-domain functions.
$$z=re^{j\omega}$$ combines growth or decay with discrete-time frequency.
ROC is essential for identifying the actual sequence.
Unit circle in ROC means the frequency response exists.
Poles and zeros explain stability, transient behavior, and filtering action.
Difference equations become algebraic equations in the z-domain.

Practice Questions

Conceptual

Explain why ROC is required along with $$X(z)$$.
What is the physical meaning of the unit circle in Z-Transform?
How do poles and zeros affect digital filter behavior?
Why does convolution become multiplication in the z-domain?

Numerical

Find the Z-Transform and ROC of $$(0.8)^n u[n]$$.
Determine whether a causal system with poles at $$0.2$$ and $$1.1$$ is stable.
Find $$H(z)$$ for $$y[n]-0.7y[n-1]=x[n]+x[n-1]$$.
Use partial fractions to find the inverse Z-Transform of $$X(z)=z/(z-0.5)$$ for a causal ROC.

MCQs

For a stable discrete-time LTI system, ROC must include: unit circle / origin only / infinity only / every pole.
For a causal rational system, ROC lies: outside outermost pole / inside innermost pole / exactly on poles / only on unit circle.
A zero on the unit circle generally causes: frequency notch / instability / infinite gain at all frequencies / no effect.