Directional RF spectrum analyzer

4Alex

Hello fellow Propeller chaps!

After much research, I can't find a practical solution to a problem that could potentially interest some of you, especially those using radio-beacons for robotics and dead-reckoning. I am turning to you all for your knowledge and some guidance.

In a nutshell, I would like to design a *small* unit, Propeller-based (of course!), that could monitor the wireless spectrum (from 100's MHz to low GHz), identify the top 10 or so highest RF signals (RSSI) in the ISM bands, and locate their origin. Basically some sort of a miniature directional spectrum analyzer.

For determining the origin of the RF signal, I was thinking of a few antennae, oriented in different directions, sequentially monitored, and mounted on a slow rotating base. A digital compass would provide the orientation of the primary 'index' antenna. This could constantly monitor the surrounding spectrum and log the results for each antenna.

What I can't locate is (a) an IC that would 'read' the wireless spectrum in the range of interest, 3v3, and I2C (if possible), and (b) a chip antenna, similar device, or etched pattern that would be directional enough to be useful.

I would gratefully appreciate any assistance on this matter. Many thanks in advance and my apologies for this off-topic posting.

Cheers,

Alex

Ale

I do not know such an ic. Maybe the answer is to have more than one receiver everyone with its own local oscillator and everything. I just remembered that a TV tuner may help you with exactly this task, I do not know exactly how you can use it, but having a look at the ic inside can give you an idea. The I2C protocol was exactly thought for that if I'm not mistaken!

Edit: I just disassembled a tuner from a satellite receiver that I had (almost) completely already dismantled and found a DA8722 inside.

pdf1.alldatasheet.com/datasheet-pdf/view/19549/PHILIPS/TDA8722M.html

Have fun

Ale

Post Edited (Ale) : 9/17/2008 4:20:11 PM GMT

Beau Schwabe

4Alex,
·
You not are going to get 100's MHz to low GHz detection with the Propeller unless you use a down converter of some kind to shift the frequency down to something more manageable.
·
For determining the origin of the RF signal, you don't need to rotate your antennas... You can use 3 fixed antennas that are a specific 'known" distance apart.· Using a Phase detector between the antennas through triangulation you can determine the signal direction.· You could get away with
two antennas if you know for certain that the signal is .... figuratively "in-front" of you rather than behind you at any time.
·
Distance is another issue, but direction should be easily obtainable.
·


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

Leon

Spectrum analysers typically up-convert to a high IF and then down-convert to a low second IF, so that the range can be covered by a minimal number of oscillators. Mini-Circuits Laboratories makes suitable mixers and oscillators. You need a logarithmic amplifier for the second IF.

Leon

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Amateur radio callsign: G1HSM
Suzuki SV1000S motorcycle

Greg Norton

Beau,

Can you expand on this method or provide links for further reading?· I would like to learn more about this technique.

Thanks.

Greg

VIRAND

Google "poor man's spectrum analyzer" for lots of very relevant DIY stuff.

4Alex

Thanks for all the replies.

@Beau: I'm sorry, I was not clear enough. I don't intend to use the PPL as the detecting element but as the main processing unit interfacing a detecting chip(s), preferably through I2C. For practical reasons, I can't have an array of 3 antennae located at various locations (triangulation), hence the rotating scheme. My idea was a bit like the way marine radar operates.

@Leon & Virand: Thanks for your suggestions. I've looked into several links provided and I came across a bug sweeper. Perhaps my application could use the methods applied to 'bug' sweeping. In a sense, its kind of a spectrum analyzer. I've seen models that can fit in the palm of the hand. I've never used a sweeper before. Could someone tell me if a bug sweeper is *really* directional (e.g., locating the bug instead of just detecting one) ? If so, then I could possibly used such a device as a starting point. Its battery powered, so it has to be in the range of a PPL application. Also, what's the basis of a bug sweeper: does it really sweep through each and every single frequencies and measure the RSSI or is it something else?

Thanks all for your assistance.

Cheers,

Alex

4Alex

I forgot to include the URL for the hand-sized sweeper:

http://www.pimall.com/nais/countertekbugtracer.html

This puppy is supposed to sweep from 1MHz to 6GHz, so my app is well within that range. However, it looks a bit too good to be true...

Cheers,

Alex

Cluso99

You might like to try the ARRL (American Radio Relay League??). It is the amateur radio organisation in the USA. They publish the book ARRL which is kits and circuits for all these kinds of things. The RSGB (Radio Society of Great Britain used to do the same). They shoud have aerial designs and detectors, etc, so it could be a good start.

Mike Green

A "bug sweeper" is just a simple sensitive broadband detector. Its function is to detect radiated electromagnetic radiation at any and all frequencies possible. I suspect it doesn't do any "sweeping" really at all. There's probably a printed circuit antenna, diode detector, some kind of op-amp to amplify the output of the detector, and a circuit to flash the LEDs sequentially at specific thresholds.

Beau Schwabe

Greg Norton,

"Can you expand on this method or provide links for further reading? I would like to learn more about this technique."

This style of detection is used in what is often called a "Fox hunt" There are different variations, some use two antennas, others use a center tapped circular loop that functions as both the antenna and the coil.

The technique relies on the propagation of the speed of light approximately 1ns per foot. If both of your antennas are equidistant from the transmitter, then the signal reaching the antenna happens at the same time. If you were to draw a line between both antennas, the line would be perpendicular to the direction of the transmitter.

If you change the orientation of the antennas in relation to the transmitter, then the signal reaching each antenna will be slightly different in Phase.

For example 98.357MHz has a wavelength of 10 feet (http://www.1728.com/freqwave.htm) ... If the antennas are placed 2.5 feet apart (1/4 of the wavelength), then the Phase relationship would be proportional to the direction of the transmitter.... If I'm facing 90 Deg to the right of the transmitter, I would expect a 90 Deg Phase shift with the Left antenna leading, and the Right antenna lagging. Likewise if I am facing 90 Deg to the left of the transmitter, I would expect a 90 Deg Phase shift with the Left antenna lagging, and the right antenna leading.

For the example, the antenna distances were nice and tuned to the wavelength that we were looking for, but as long as you know the distance between the antennas, and the frequency that you are tuned to, then this really doesn't matter, other than requiring more software math to make it more Human readable.


Google: "Amateur Radio Direction Finding" or ARDF

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

Post Edited (Beau Schwabe (Parallax)) : 9/19/2008 12:46:52 AM GMT

4Alex

Good morning all,

I've found an IC made by Analog Devices: AD8319. It's defined as a 1MHz to 10GHz, 45dB, Log Detector/Controller. And it is a 3v3 device!

Available at Digikey.

More to come on this.

Cheers,

Alex

Bob Lawrence (VE1RLL)

Example Fox hunt system:

users.bigpond.net.au/vk3yng/foxhunt/2m_sniffer/manual.htm

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Aka: CosmicBob