Re: Could this device be built?

The way most of the phased-array systems I have seen operate is that there is a master clock which drives all of the individual transmitter modules, and each of the modules receives a control signal for a phase modulator, which shifts the output of that module by a preset amount. How this is accomplished in each module varies a lot depending on the speed of control required, the bandwidth required, and the size. When you have a few hundred channels, shrinking each channel in size becomes critical.

--scott

--
"C\'est un Nagra.  C\'est suisse, et tres, tres precis."

Do you have a PhD in what can't be done?

John

Heating of the radome might also be a problem.

No, but I have enough years of experience with RF in general and radar in particular to know building a phased array requires precise phase (or frequency) control and you can't do that with an ultra wideband device, which has a bandwidth of 500 Mhz.

Such a device may make a great wireless LAN at ranges of tens of yards, but is not the device of choice for building a phased array anything.

--
Jim Pennino

Remove .spam.sux to reply.

What is the bandwidth of modern radars? I'd expect it to be wide and using spread spectrum tricks to make jaming harder.

--
These are my opinions, not necessarily my employer\'s.  I hate spam.

At a minumum!

Spiffy modern radars hop and chirp, both of which broaden the working bandwidth.

With modern signal processing, wider radar bandwidth improves resolution. You can do all sorts of fun stuff with 1000 antennas and a few teraflops of compute power.

All the major powers - US, Russia, France, Germany, China, Israel, UK

- are working on HPM weapons and array radars. The Brits call their projects "Suave" and "Virus." MBDA and BAE are major players.

Google "mbda hpm" and "bae hpm", and believe it or don't.

John

Depends on what bandwidth you are talking about.

For the instananeous transmitted frequency, narrow.

Some military stuff has used frequency hopping since WWII to make it harder to jam.

Frequency hopping is a spread spectrum technique and the bandwidth over time is wide.

You could make a spread spectrum, phased array radar, but the frequency, phase, and amplitude of all the emitters has to be precisely controlled to form the beam, which implies that for a given pulse, all the emitters are transmitting very close to the same frequency.

The next pulse may be hundreds of megahertz away, but that's what processors are for.

--
Jim Pennino

Remove .spam.sux to reply.

I would hope so since the techniques have been around for at least a quarter century.

Narrower pulse widths and good receivers improves resolution.

A search for "mbda hpm" returns:

Your search - "mbda hpm" - did not match any documents.

And "bae hpm" returns: SIMPLE = T / file conforms to fits standard BITPIX = 16 / number

You are mixing apples, oranges and cherries.

Frequency agile radar, rudimentary spread spectrum, was originally developed in WWII.

Phased array radars have been around for decades.

And everyone WANTS a death ray, but no one has made a practical one yet.

--
Jim Pennino

Remove .spam.sux to reply.

The bandwidth of a radar pulse is determined by the required resolution of the distance. Thus there is generally no point to increase the pulse bandwidth beyond ~100Mhz unless for the very special tasks like a target feature recognition. However the carrier frequency and the spreading code can vary from pulse to pulse.

Vladimir Vassilevsky

DSP and Mixed Signal Design Consultant

Don't type the quote marks. Geez.

John

There are plenty of commercial death rays in the 54-72 MC and 76-88 MC bands. They don't cause death directly, but transmissions on these frequencies can cause severe brain damage even at low levels when demodulated and viewed. A number of studies have shown long-term exposure to cause all sorts of problems in children.

--scott

--
"C\'est un Nagra.  C\'est suisse, et tres, tres precis."

AIUI they are termed "brain-death rays"

Cheers Terry

Then don't say Google "mbda hpm" and "bae hpm".

And if I don't use quotes, will I get information on phased array radars, phased array death rays, or phased array, spread spectrum, death ray radars?

--
Jim Pennino

Remove .spam.sux to reply.

Vladimir Vassilevsky

DSP and Mixed Signal Design Consultant

Not a problem if you buy my aluminum foil long johns.

--
Jim Pennino

Remove .spam.sux to reply.

How do you even know? I am building an 836 channel RF stimulator for the JSF so they can test all of their phase alignments, etc.

The airframe senses and tracks hundreds of targets in all four quadrants.

You're an idiot. You have allowed your susceptibility to bullshit like that which Rosie O'Retard spews to color your ability to grasp the facts in a modern world.

It seems to me, that with modern electronics and information technology, that a high resolution, handheld, RADAR system is possible.

You could quasi-randomly modulate (Variable transmit and listen periods), a solid state microwave oscillator (Gunn Diode) with a digital code with good correlation properties (Gold Code),

cross-correlate the echoes received when in the listen mode with the Gold Code, then cross-correlate the correlations from the echoes with stored geo-patterns downloaded from a Google-Earth like data base covering the area of operation,

compare adjacent (In time) echo returns to spot moving targets, then present the pattern on a small, solid state, color display that shows the Google-Earth like picture of the area, with super-imposed moving targets.

One would not need a directional antenna, nor high power for such a device, but it would be necessary to sweep the device around to build up a good correlation of the area as one's body and other things would block the signals and, even though the Google-Earth like picture, and the location of the RADAR would still be valid, but blocked moving targets would not be detected.

Note that if a map of the area of operation is downloaded into the system, and a set of times from the radar to fixed targets is compared to the map, one could quickly correlate the map with the echoes and determine where one is.

With such a device, one could move around, and see where they were on a moving Google-Earth-like picture, and see the moving targets about them, perhaps even colored and shaped by the RADAR signatures of the targets. (People, cars, tanks, trains, an incoming missle, etc.)

Note that for many situations that such a device could replace GPS. Just like GPS, after the device determines where one is, it would be able to compute changes in position quickly.

Hey maybe, I should patent this device?

As I mentioned in old posts, I used "Data Mining" back in the 1980's in my businesses and applied for a patent on "Data Mining" just when they began to allow software to be patented, but I decided not to complicate my life, and didn't complete the patent.

In other words, if you want to commercialize this idea for non-military applications, go for it.

As any entrepreneur knows, ideas are a dime a dozen, and what requires blood, sweat and tears is getting an idea to the marketplace.

The bottom line is, no one should be able to hold progress hostage with a patent, that is obvious to many, as the state of the art exposes new approaches.

--
Tom Potter

*** Time Magazine Person of the Year 2006 ***
*** May 2007 Anti-Bigot Award ***
http://home.earthlink.net/~tdp
http://tdp1001.googlepages.com/home
http://no-turtles.com
http://www.frappr.com/tompotter
http://spaces.msn.com/tdp1001
http://www.flickr.com/photos/tom-potter
http://tom-potter.blogspot.com

I'd like to have a short-range imaging radar, sort of like my Flir handheld thermal imager, as a sort of super stud finder.

Imagine a pc board covered with etched patch antennas, one or more step-recovery-diode impulse generators, and a lot of sampling receivers. Run it at several MHz, do a lot of averaging and signal processing, and reconstruct the image. Maybe use Wii type accelerometers so as the array is moved around, additional signal paths can be crunched in to enhance resolution without blurring. The microwave side of the hardware would be dirt cheap, and the signal processing would have a high engineering cost but would also be cheap in production.

Take a look at McEwan's patents for an idea of how the hardware would work. He was mostly looking at stuff like auto collision detection,

1-dimensional ranging, but imaging is quite feasible if you dump enough DSP onto the problem.

Firemen could use this for smoke penetration, or cops could spot bad guys in the next room, and I could spot cats under beds without having to crawl around on the floor.

John

You can already buy these devices, but of course the price tag is still rather high (possibly something like $50k/unit!? -- I visited the booth of a company selling them a couple years back now, and they were targeting fire departments and other government funded agencies that had that kind of money to throw around).

Super stud finders would sell like hotcakes once you got them to the $500 level. Even at $2500 you'd probably get plenty of takers. But at $50k... not so much.