LTI System in Signal Processing: Definition and Examples

An LTI system in signal processing is a system that is both Linear and Time-Invariant. This means its output is directly proportional to its input and does not change over time, making it predictable and easier to analyze.

How It Works

An LTI system works by following two simple rules: linearity and time-invariance. Linearity means if you put in two signals separately and add their outputs, the result is the same as putting in the sum of those signals at once. Think of it like mixing colors: mixing red and blue separately then adding the results is the same as mixing red and blue together first.

Time-invariance means the system's behavior does not change over time. If you send a signal now or later, the system responds the same way, just shifted in time. Imagine a coffee machine that always makes the same coffee no matter when you press the button.

These properties make LTI systems very predictable and easy to study using mathematical tools like convolution and Fourier transforms.

Example

This example shows how an LTI system responds to an input signal using convolution with its impulse response.

python

import numpy as np
import matplotlib.pyplot as plt

# Define an impulse response of the system (e.g., a simple smoothing filter)
h = np.array([0.2, 0.5, 0.3])

# Define an input signal
x = np.array([1, 2, 3, 4, 5])

# Compute the output using convolution (LTI system property)
y = np.convolve(x, h, mode='full')

# Plot input, impulse response, and output
plt.figure(figsize=(8,4))
plt.stem(x, linefmt='b-', markerfmt='bo', basefmt='k-', label='Input Signal x[n]', use_line_collection=True)
plt.stem(h, linefmt='g-', markerfmt='go', basefmt='k-', label='Impulse Response h[n]', use_line_collection=True)
plt.stem(y, linefmt='r-', markerfmt='ro', basefmt='k-', label='Output y[n] = x[n]*h[n]', use_line_collection=True)
plt.legend()
plt.title('LTI System Example: Convolution Output')
plt.xlabel('n (time index)')
plt.ylabel('Amplitude')
plt.tight_layout()
plt.show()

Output

A plot showing three stem plots: blue for input signal, green for impulse response, and red for output signal which is the convolution result.

When to Use

LTI systems are used whenever you want to analyze or design systems that behave predictably over time and respond proportionally to inputs. This includes audio processing, image filtering, communication systems, and control systems.

For example, in audio equalizers, the filters are designed as LTI systems to shape sound without unexpected changes. In image processing, smoothing or sharpening filters are LTI systems that modify images consistently.

Key Points

• An LTI system is both linear and time-invariant.
• Linearity means outputs add up when inputs add up.
• Time-invariance means system behavior does not change over time.
• Convolution is the main tool to find output from input and impulse response.
• LTI systems simplify analysis and design in many signal processing tasks.

Key Takeaways

An LTI system's output is a linear combination of its inputs and does not change over time.

Convolution with the system's impulse response gives the output for any input signal.

LTI systems are fundamental in audio, image, and communication signal processing.

Their predictable behavior makes them easier to analyze and design.

Understanding LTI systems helps in building filters and control systems.