From Signals to Meaning: Signal Processing
Turning messy data into usable data
Why preprocessing matters
Raw brain data often contains noise from movement, blinking, muscle activity, and the environment. Preprocessing helps remove unwanted parts and makes the signal more consistent.
Typical preprocessing pipeline
-
1
Step 1: Collect the raw signal
-
2
Step 2: Remove obvious noise or artifacts
-
3
Step 3: Normalize the data so values are comparable
-
4
Step 4: Split the signal into windows or segments
-
5
Step 5: Extract features for modeling
Feature extraction
A feature is a measurable property that helps a model learn. For brain signals, features may include signal power, frequency content, timing patterns, or activity across different sensors.
Key idea
AI models usually do better when the input is organized. Good preprocessing can matter as much as the choice of model.
Why is preprocessing needed?
Preprocessing cleans and structures data so models can learn more effectively.
Correct answer: To remove noise and prepare data for modeling
What is a feature?
Features are inputs that help AI detect patterns in data.
Correct answer: A measurable property used by a model