
What a three-day RF survey of my own airspace turned up
You can’t secure a wireless network you’ve never actually looked at. So I set up a proper Wi-Fi audit rig — a capable USB adapter on a Raspberry Pi — and pointed it at my own airspace for three days. What came back was a genuinely humbling list of things wrong with a network I thought I’d locked down, plus a clear-eyed picture of what these tools can and can’t do. Everything here was against my own equipment; the neighbours’ networks I only ever listened to, passively, and never touched.
The rig, and its honest limits
The adapter is a Wi-Fi 6E, 2×2 MIMO device — it can put its radio into monitor mode to capture raw frames, and it can inject frames for testing. The single most useful thing I learned about it wasn’t a feature, it was a constraint: on one radio, you cannot transmit while you’re associated to a network. You can listen while connected, but the moment you need to actively test, you have to give up the connection and dedicate the radio to monitoring. Injection on a dedicated monitor interface works perfectly — thousands of frames, no drops — but trying to do it while also staying online silently fails. Knowing that operational rule up front saves hours of chasing a “broken” adapter that’s actually just doing what the hardware allows.
I also learned to respect how easily this class of adapter can wedge or brown out a Pi if you reset the driver carelessly — enough that I added a hardware watchdog so a lockup self-recovers instead of needing me to physically pull the power. Small detail, real lesson: give anything that can hang your box a way to reboot itself.
What the survey found on my own network
I left a capture running for three days and then, crucially, actually read it — which is the step most people skip. Three findings stood out, all on my own kit:
- An open access point broadcasting inside my house for 54 hours. One of my own IoT gadgets, unable to join its assigned network, had quietly fallen back to hosting its own open, passwordless setup portal — and left it running for over two days. Anyone in range could have joined it and repointed the device wherever they liked. It only closed when the gadget finally rebooted onto a network it could reach. That’s the kind of thing that never shows up in a router’s dashboard and only appears when you’re actually watching the air.
- An IoT device sitting on my trusted network instead of the quarantined one. By correlating its Wi-Fi and Bluetooth signatures — which the adapter can watch at the same time — I identified a cheap smart device with an open local-control port sitting on the network it absolutely should not have been on. It belonged in IoT jail.
- A silent security downgrade. My main network advertises the modern WPA3 standard, but it runs in a “transition mode” that still accepts older WPA2 clients for compatibility. My adapter proved this by connecting as a plain WPA2 client even though WPA3 was on offer. That backwards-compatibility is convenient and it’s also a weakness — it keeps the door open to the older attacks WPA3 was designed to close.
The lesson that reframed a decision
There was a nice twist. I’d previously turned on Protected Management Frames set to required on my IoT network, because PMF is the correct defence against the classic deauthentication attack. But the survey showed the trade-off: my cheap IoT devices don’t support PMF at all, so “required” was locking the very gadgets that network exists to contain out of it — which is exactly how that open portal ended up broadcasting. Security controls have costs, and the only way to see them is to measure the real behaviour of your real devices, not the theory.
Attacking — only my own, only to learn the defence
I did prove the offensive side end to end, entirely against my own access point and my own devices: a targeted deauthentication genuinely knocks a client off, and the re-join that follows is the moment a handshake crosses the air. Seeing it work taught me why it works — and, more importantly, why the defences matter. On the network segment with PMF properly in place, the same attack simply did nothing. That’s the whole point of doing this: you learn to shut a door by opening it, on a lock you own.
The takeaway I keep coming back to: your network is doing things you don’t know about until you listen. An afternoon of reading your own airspace is one of the most honest security exercises there is.
Tech: Wi-Fi 6E USB adapter (monitor + injection), Raspberry Pi, Kismet, airodump-ng, 802.11w (PMF), WPA3 transition mode.