Hands-On Labs

Welcome, signal hunter!

Hi, I'm Echo. I hunt things by listening — that's literally what echolocation is, real-time signal processing with flippers. Over these 35 labs you're going to teach a $5 chip to do the same trick. Time to transform!

Thirty-five labs that take you from "what's a Thonny?" to "I hand-wrote an ARM assembly FFT and benchmarked it against eight competing versions."

You need no prior experience with FFTs, signal processing, or assembly language. Lab 1 assumes you own a computer. That's it.

Why this is worth your time

Here's the thing nobody tells you: a fast FFT is a superpower. Sound, vibration, radio — anything that wiggles — looks like meaningless noise until you transform it into the frequency domain. Then patterns leap out. A whistle becomes a number. An engine's rattle becomes a diagnosis.

Doing that transformation fast enough to keep up with the real world, on a chip that costs less than a sandwich, used to require expensive dedicated hardware. Most CS and EE undergraduates never get to touch this. You're about to.

What you'll need

Item	Approx. cost	Used from
Raspberry Pi Pico 2 (RP2350)	$5	Lab 1
SSD1306 OLED display, 128×64, SPI	$5	Lab 4
Two push buttons	$1	Lab 5
INMP441 I²S MEMS microphone	$3	Lab 7
Breadboard + jumper wires	$5	Lab 4

Every pin number lives in one file — config.py — which you'll set up in Lab 4 and import everywhere after that.

The labs

Module 0 — Getting Started

Your computer talks to a microcontroller for the first time.

#	Lab	You'll build
1	Hello World with Thonny	Your first program on real hardware
2	Blink: Your First Hardware Program	A blinking LED you control
3	Know Your Board	A report on your chip, read from its own registers

Module 1 — Peripherals

Adding eyes, hands, and a filing cabinet.

#	Lab	You'll build
4	The OLED Display	Text and pixels on a real screen
5	Buttons and Interaction	A menu you can click through
6	Deploying Code and Libraries	A program that runs on power-up, untethered

Module 2 — Sound as Numbers

The microphone arrives. Everything gets more fun.

#	Lab	You'll build
7	Your First Sound Capture	Raw audio samples, straight off the mic
8	Sound Levels: RMS and a VU Meter	A live loudness meter that reacts to your voice
9	Sampling Rate and Aliasing	A tone that lies about its pitch — and why
10	Bit Depth, Headroom and Clipping	Deliberately blown-out audio, and the fix

Module 3 — Discovering Frequency

The heart of the course. You will invent the DFT rather than be handed it.

#	Lab	You'll build
11	Sine Waves: Amplitude, Frequency, Phase	Waves from scratch, in code
12	Adding Waves: Superposition and Beats	Two notes that fight, and wobble
13	Correlation: Does My Signal Contain This Note?	A detector for one specific frequency
14	Sweeping All Frequencies: You Just Built a DFT	Your own DFT. Yes, really
15	Validating Your DFT on a Known Signal	Proof that it actually works
16	Your DFT Is Too Slow	The measurement that motivates everything next

Module 4 — The FFT

Same answer. Vastly less work.

#	Lab	You'll build
17	Divide and Conquer: From DFT to FFT	The trick that kills the wasted work
18	Bit Reversal and Twiddle Factors	The bookkeeping that makes it fit in place
19	The Butterfly	The four-line operation at the centre of it all
20	A Complete Python FFT	A working FFT, validated against your DFT

Module 5 — Real Spectra

Point it at the world.

#	Lab	You'll build
21	Spectrum of a Real Sound	A live spectrum — whistle and watch the peak move
22	Windowing and Spectral Leakage	Smeared peaks, then sharp ones
23	Peak Detection: Build a Tuner	A working instrument tuner
24	Real-Time Spectrum Analyzer	The full pipeline, profiled stage by stage

Module 6 — Measuring Performance

You cannot optimize what you cannot measure honestly.

#	Lab	You'll build
25	How Long Did That Take?	Nanosecond-accurate timing from a CPU register
26	Benchmarking Methodology	A harness that doesn't lie to you
27	The Abstraction Ladder	Python vs. native vs. viper vs. C vs. assembly

Module 7 — Assembly Language

Where the speed actually lives.

#	Lab	You'll build
28	Does Your CPU Have an FPU?	A capability probe — and a cautionary tale
29	Your First Assembly Function	Machine instructions you wrote yourself
30	Talking to the FPU	Float math at hardware speed
31	The Butterfly in Assembly	A complete assembly FFT

Module 8 — Optimization and Capstone

#	Lab	You'll build
32	Specialization and Branchless Code	Faster code by doing less
33	Beyond the Assembler: Hand-Encoding	An instruction your assembler refuses to write
34	Competing Variants: Predict, Measure, Explain	A comparison matrix, and some surprises
35	Capstone	Your own variant, benchmarked and written up

How to get the most out of these

Whenever a lab says Predict, then measure — actually write your guess down first. In Plan 02 of this project, every single performance prediction turned out to be too optimistic. Being wrong on paper is how you learn what the machine really does.