Why signal processing extracts information in SciPy - See It in Action

Why Signal Processing Extracts Information

📖 Scenario: Imagine you have a noisy recording of a bird singing in a park. You want to find out the main melody of the bird's song, but the noise makes it hard to hear. Signal processing helps us clean the noise and find the important parts of the sound.

🎯 Goal: You will learn how to use signal processing to extract useful information from a noisy signal. We will create a simple signal, add noise, filter the noise out, and see the cleaned signal.

📋 What You'll Learn

Create a noisy signal using numpy

Set a threshold frequency for filtering

Use a filter from scipy.signal to remove noise

Print the filtered signal values

💡 Why This Matters

🌍 Real World

Signal processing is used in many areas like cleaning audio recordings, improving medical signals like ECG, and analyzing sensor data.

💼 Career

Understanding how to extract useful information from noisy data is a key skill for data scientists, engineers, and researchers working with real-world data.

1

Create a noisy signal

Create a numpy array called time with 500 points from 0 to 1 using np.linspace(0, 1, 500). Then create a signal called signal as np.sin(2 * np.pi * 5 * time). Finally, create a noisy signal called noisy_signal by adding random noise using np.random.normal(0, 0.5, 500) to signal.

SciPy

import numpy as np
# Create time array, signal, and noisy_signal
# Your code here

Need a hint?

Use np.linspace to create the time points. Use np.sin for the signal. Add noise with np.random.normal.

2

Set the cutoff frequency for filtering

Create a variable called cutoff_freq and set it to 10. This will be the frequency threshold for filtering noise.

SciPy

import numpy as np

# Set cutoff frequency for filter
cutoff_freq = 10
# Your code here

Need a hint?

Just create a variable named cutoff_freq and assign it the value 10.

3

Filter the noisy signal

Import butter and filtfilt from scipy.signal. Create a Butterworth low-pass filter using butter(4, cutoff_freq, fs=500, btype='low', analog=False) and store the result in b, a. Then apply the filter to noisy_signal using filtfilt(b, a, noisy_signal) and store the result in filtered_signal.

SciPy

import numpy as np

cutoff_freq = 10

from scipy.signal import butter, filtfilt
# Create Butterworth filter and apply it to noisy_signal
# Your code here

Need a hint?

Use butter to create filter coefficients and filtfilt to apply the filter.

4

Print the filtered signal

Print the filtered_signal array to see the cleaned signal values.

SciPy

import numpy as np

cutoff_freq = 10

from scipy.signal import butter, filtfilt

b, a = butter(4, cutoff_freq, fs=500, btype='low', analog=False)
filtered_signal = filtfilt(b, a, noisy_signal)

# Print the filtered_signal array
# Your code here

Need a hint?

Use print(filtered_signal) to show the cleaned signal values.