How I built a mmWave material classification radar

2025 · [radar, rf, dsp, embedded, beamforming, startup]

the prototype — IWRL6432 + ESP32 inside the casing

Software is now a commodity thanks to Claude Code. So the next step is obviously hardware. I spent the last 6 months on building a hardware startup, which was fucking hard.

I made a radar that could classify materials, and this is the story of how I did it as my "end of studies" project. By the way, the project never ended (as you will see down this article, because of a lack of funding)

I live in Europe, where asbestos is a huge and common pain across every country here. That stuff fills walls, and requires people to come at your place to tell you if you have asbestos contaminated materials in your building. If so, you might have been breathing poison since you were a kid. Asbestos gives you cancer, and can fuck you up pretty bad.

The traditional ways of detecting it, is basically to pick a wall sample, send it to a lab, and they tell you if you have asbestos in your walls. Of course, there a lot of intermediaries that come into place to catch a good amount of money based on regulations and your fear of getting poisoned. So a 1$ analysis becomes a 60$ one to the end payer, and when you have to make tens of them, price skyrockets.

The idea

So my project was to build a radar that detects asbestos for you. It's based of material sciences, and wave physics, which happens to be my areas of expertise (recent eng grad). There I had my device plan : make an asbestos sensing radar.

Then, I had to design electronics, so because I am not a bozo, and I hope you are not one either, I ordered dev boards to quickly prototype. In my case : a Texas instrument IWRL6432 BOOST, and an ESP32 dev kit. Assembled them, and started tinkering with DSP algorithms to detect materials.

Built this test bench to test out the electromagnetic response of materials to my special material radar. I could hot swap materials and tighten materials to test out my device.

training the classifier

I finalised my approach with capon beamforming to get a density spectre, which I fed out to a neural network. And this allowed me to classify material surfaces, which under the "same surface, same layer" material hypothesis and "material change is sudden and discontinuous", would then allow me to say "hey this entire layer is made out of x,y,z material."

The Digital signal processing chain

the DSP chain, end to end

The radar is FMCW : it doesn't emit a single tone, it sweeps its frequency upward over time (chirp). The whole job of the DSP chain is to turn the echoes of those chirps into a material signature. Here is how it works :

Chirp generation and characterization the TX antennas emit a linear frequency sweep. Everything downstream depends on the exact shape of that sweep (start frequency, slope, bandwidth), so the first thing I had to do was characterize the chirp actually coming out of the front-end.

Mixing and beat signal the received echo is mixed with the transmitted chirp. What comes out is a low-frequency "beat" tone whose frequency is proportional to the round-trip distance to the reflector.

Range FFT an FFT on that beat signal turns frequency into distance. Each bin is a depth into the material, so I get reflected energy as a function of range.

Capon beamforming (AOA) across the MIMO RX array, Capon (MVDR) beamforming resolves the angle of arrival for each range bin, giving a sharp angular density spectrum instead of one blurry return.

The output of all this is a per-range, per-angle density "spectre" the electromagnetic fingerprint of whatever surface sits in front of the radar. That spectre is exactly the tensor I fed to the neural network to classify the material.

scanning a surface with the radar — the density spectre, before the AI step

chirp characterization

I spent two months on this, getting the toolchain right, and on the infamous compile flash debug loop of embedded firmware.

AI part

Since everyone is using AI right now, might as well put some AI in your invention to put new capabilities that you could never before.<br>So what we did is put neural network that ingests the output of the DSP chain, and outputs a class of material.

This model actually "learns" the electromagnetic properties of materials (epsilon' and epsilon''), it's just a classic CNN.

Mechanical & RF

Then, came the mechanical casings, started making a look-a-like, when I finally came to the conclusion that designing around the devboards was a way better approach.

producing the mechanical casing

I also went into the rabbit hole of RF prototyping. Ask anyone, they will tell you that RF is straight up black magic. PDEs, Physics, and Chaotic systems, all packed in a single field. So I started modeling my antennas to capture what was happening in my circuit to reverse engineer it.

Here is how I optimized the antenna shapes so that they would have the same characteristics as the ones in...