Signal analysis

Signal analysis · Course page

Clean a signal with DSP filters, look in the frequency domain with FFT, then squeeze a sliding window into the feature vector a model actually sees.

Module goal

Understand that a model never sees a raw signal — it sees the features a front-end prepares for it, and that front-end must match everywhere.

Lessons

Lesson	Topic	Time (min)	Slides
4.1	DSP filters: EMA, Median, Kalman and the radar range profile	65	slides.md
4.2	Hands-on: filters cleaning a signal live	75	slides.md
4.3	FFT and the frequency domain: bins, Nyquist, DC, leakage and the Hann window	65	slides.md
4.4	Hands-on: a live spectrum from the IMU	75	slides.md
4.5	Features and windows: what a model actually sees	65	slides.md
4.6	Hands-on: a feature vector from a sliding window	75	slides.md

Lessons come in pairs: a concept lesson followed by a hands-on lesson with a practice file, a solution, and a lab.

Module checkpoint

You pass this module once you can do all of the following (details are in the Lab section of each hands-on lesson):

• A filter visibly improves a noisy sensor signal — the Filtered line is clearly smoother than Raw, confirmed with a noise-down % figure (lesson 4.2).
• Run s09_fft_spectrum.py and read a live spectrum — the frequency bars and the peak (Hz) genuinely change with faster or slower shaking (lesson 4.4).
• Build a feature vector from a raw signal by hand: slice a window (window + hop), then squeeze it into mean, std and band values that change with motion (lesson 4.6).