2.4GHz 'Crystal set'

Bernie has built up a 2.4GHz 'Crystal set'.

OK it's really a wide band AM detector for the 2.4 GHz band.

The human brain is damn clever - good at recognising sounds. Agreed it is not a D.S.P. and cannot resolve signals, but listening is very intuitive. (As one that spent a childhood listening to TV - literally! I can spot 50Hz with 10.125 kHz a mile off!)

The entire set is portable. With the corner reflector it's possible to direct and change polarisation very quickly.

Better than netstumbler! Weak signals can be heard that are too faint for a 802.11b card.

Pictures

corner reflectorCorner reflector
Bernie's ElectonicsBernie's electronics.
VideoVideo signal. (Severely bandwidth limited!)
AP beaconAccess Point beacon. H.F. data being filtered out.
BluetoothBluetooth. The hopping frequency is 1600Hz. Heard as a whine when "searching for devices".

The latest TDK driver is a bit more together. The current version (April 30th 2003)1.3.2.19.

Download size of 19.5 Mbytes. The driver takes 45 minutes to download! (Yes, I'm living in a modern city in England. I must do something about this!)

 

Circuit

Corner reflector-> MMIC-> 4  stage filter -> MMIC -> 4 stage filter -> MMIC ->  3 stage filter ->diode detector -> amp -> earpiece.

Notes

There's no instability, unless you really try hard. The noise floor is determined by the MMICs, and not the AF amp.

Selectivity looks good enough (This is usually the failing of TRFs) and it is sensitive.

The 802.11b beacon Tx pulses are very short, so what sounds a weakish signal on the speaker, looks better on a scope - I don't think this a real problem though. A meter fed by a peak hold detector could be added.

There is HF curtailment - first, because of the op-amps themselves but mainly because of shunt C added across the feedback resistor (I wanted to restrict the bandwidth to improve the LF s/n ratio.)

The MMICs want +5v, so its the obvious thing to use for the diode bias too. Forward biased diodes make much better AM detectors.

More pictures here.


Playing with the set

Andrew gets a report...

I've just got back. I have Bernie's crystal set with me!

How good is it?

Very instructive. I'll bring it over tomorrow.
I can hear 802.11b - pulses.
Bluetooth - a carrier with a 1kHz whine.

Is that the hopping?

I guess so. FM TV carriers... complete with mobile flutter when driving along in a car.
And microwave ovens!

Of course!

There's a video sender near my house. The 50Hz frame sync is very obvious.
I'm listening to Anubis, Radio 4, over the 802.11b network - and listening to the 802.11 carrier...
The crystal set drives headphones: there is some more work to do with it - but it works well as is.
We'll play tomorrow!

Cool!


A report sent to Bernie

I've been playing!

1, The crystal set gives an immediate indication of 802.11b problems. The UI does not tell you what's going on - but you can hear! I set up a file transfer over 802.11b. I listen ...I cover the pc's aerial... I hear the thing struggle / retry.

2, The AP beacon can be adjusted. The default is 100ms. I set it to 5ms. I now hear a 200Hz tone. I walk a circuit around my block in MK - my AP being obvious. I hear a video sender, Bluetooth, a microwave oven, and another 2 AP's. Where I can see my AP in between obstructing houses I hear the 200Hz tone.
It's really line of site - just like driving through the cuttings this afternoon on the A14.

3, The video signal is confirmed by inspecting on a scope. Vertical sync and back porch are PAL TV, 20ms period. I'd like to see your circuit... too much H.F. cut off to get video. (Also the neighbours video sender must be a bit dodgy as I can see the AM...)

4, I connected up my second corner reflector, via cables you supplied, to the crystal set. All works as expected.
I note regeneration. By aiming my corner reflector at the strip line I get instability - also increased sensitivity - the neighbouring video signal is stronger.