This is incredible stuff! One of the less-hyped slow revolution is what's happening in non-wearable sensing. Some of it is driven by the consumer-ization of advanced radar systems, a technology that was the domain of the military and space/satellite comm.
Some companions to this work:
1. Wifi sensing from academia, particularly in the ubiquitous computing community- Shwetak Patel's group in Washington [1], Dina Katabi's group in MIT [2], Patwari's group in Utah [3], etc.
[4] I use 'advanced' in the sense that it is way, way, more sophisticated in its use of math and signal processing/machine learning/linear algebra algorithms than your standard automotive speed monitoring radar.
Some of this is the sort of stuff you really don't want to interfere with--both for safety reasons, and because of the likelihood of enforcement.
Next time I suggest that the author look into whether his jurisdiction will grant him an experimental license so he can do this legally. In many jurisdictions that is cheap and easy (unless, of course, you want to operate in a band where the interference risk is just too high).
Edit: I see the author is in Finland. Unfortunately for the author, I'm guessing this changes nothing since these frequencies are regulated both nationally and internationally. But it does mean that there is probably some local variation in the exact services that might be subject to interference. And who knows: maybe in Finland the 5.5 GHz band is a "do whatever you want at any power level you want" band. But I doubt it...
Or he has his amateur radio license and is licensed to operate at higher power levels.
Even unlicensed, he has up to 25mw of power and 8.5db antennas (which with 25mw or less can give you upwards of a mile of clean signal), which may be sufficient. Hard to say.
Does an amateur license allow you to operate a RADAR system? I doubt it. I had thought that amateur operators were somewhat limited in the sorts of hardware and operations they are permitted. (E.g., type acceptance of equipment when operating outside designated amateur bands.)
And in the U.S. at least the bands around 5.5 GHz (the so-called U-NII-2 and U-NII-2 extended bands) are subject to lots of interesting rules for unlicensed operation, including the ability to detect and avoid incumbent RADAR signals. https://transition.fcc.gov/bureaus/oet/ea/presentations/file...
Of course, I know literally nothing about the Finnish rules that actually apply here...except that, thanks to the ITU, they probably bear some resemblance to the U.S. rules.
Radar systems are fairly common on boats larger than ~30 ft, and you can operate those without any license at all. The latest generation devices are all broadband, working on the principle that if you spread your broadcast over a wide enough range, the power you're broadcasting on any given band is very low, i.e. legal and not causing interference for others. Maybe this guy is working off the same principle?
You don't always need an operator license but, when you don't, you typically do need properly certified equipment. (This is also why you don't need a license to use Wi-Fi.) The rules vary by service and by frequency band.
I'm no expert on boat RADARs, but I'd bet that, 1) they operate in bands allocated for that sort of use and 2) that they have been either type accepted or certified by the FCC to comply with whatever rules govern their band of operation.
> Does an amateur license allow you to operate a RADAR system?
Yes, conditionally. RADAR does not fall under the "specifically prohibited" rule for communications, but it's also not explicitly listed as "specifically authorized" either. Likely up to the discretion of whatever FCC official you asked.
> the ability to detect and avoid [interference with other] signals.
This is a requirement for all amateur radio transmissions. However, so long as you take reasonable precautions (and react to any reports of issues), you can do it.
In my (unprofessional) reading of the phrase "You determine for yourself whether your communications should be transmitted on amateur service frequencies", a reasonably powered experimental radar system which doesn't interfere with other services would be OK.
I wondered about that, too. There's a worldwide ISM band at 5.725 to 5.875 GHz in which you can run radars. But he says he's sweeping from 5.2GHz to 6 GHz. That's a lot of spectrum to be using. There's lots of stuff in there, including satellite links and aeronautical radionavigation.
Here's a similar project which is in the 5.8 to 6 GHz range. With a little work, they could probably make it stay in the ISM band there. One big frustration of gigahertz RF work is that the test gear you need to see what's happening costs far more than the project. I tried to build a frequency-modulated LIDAR once, and ran into that.
This is pretty awesome. I used to write processing software for airborne SAR and often wondered what it would take to just strap some stuff to my car and do it. Unfortunately blasting EM radiation in almost any useful frequency is a pretty big no-no without going out to the middle of nowhere.
I’m not understanding how moving along a single axis can generate a 2D image. Is there a second axis that moves at a right angle to the flight path (or leadscrew)? I must be missing something..
A radar returns a vector of measurements along the line of sight of the beam. These are in-range measurements from a particular position and time. SAR stitches together a series of these in-range measurements taken at different positions and times to create a single 2D image of the static elements of a scene.
Anna, I'm replying to myself since your posts are coming up [dead] and I can't vouch for your reply to me like I did for your top-level comment.
Yes, that is basically how it works. The system takes measurements at a series of slant ranges (called range bins) for each beam position. The returns from those range bins can be mapped to a physical location as long as the location of the receiver is known. Record enough beam positions in a line and you can map a 2D image.
Right. If the radar is held stationary, it can produce a one-dimensional "image" of what is in its line-of-sight. Move it sideways and you generate the 2nd axis.
It's a little more complex than that, because the field of view is not a perfectly narrow beam, but same idea.
Nice. The down-range focus on the fenceposts is very good. Cross-range focus isn't as good, still impressive for a such a low-cost system. It'd be very interesting to see some images of a more target-rich environment.
Radar, SAR, and especially passive multistatic SAR are technologies that I'm sure we're going to see more of in future. Passive SAR (or transmitter-of-opportunity SAR) can be computationally expensive, but pieces seem to be there to build something that works really well.
I haven't used it, but from reading the datasheet it appears to be nothing but an RF motion sensor. The output is simply a voltage level reflecting the magnitude of the Dopper shift the sensor measured. There is no ability to perform range measurement or range gating, so all it will tell you is if something is moving within its (very wide) field of view.
I'd hire this guy. This, the homemade GPS guy, and the homebuilt apollo computer guy are all the sorts of people you want on your team to teach the young bucks a thing or five.
The young bucks? Implying this guy isn't young himself? Handheld radar wasn't an uncommon project in the EECE department at my school.[0] :) I'm not entirely sure the students building radars would want to be on teams where they had to teach their peers a thing or five!
The news talk show, founded by a Turkish-American, that you are familiar with and link to is actually meant to be a humorous reference to the idiomatic usage. The idiomatic usage used to be well known, but these days people are more familiar with the news show and the idiomatic usage is somewhat archaic now. It's not a negative reference, more a respectful acknowledgement of toughness and persistence - one doesn't want to mess with young turkish warriors unless they are ready to face a competent adversary.
Some companions to this work:
1. Wifi sensing from academia, particularly in the ubiquitous computing community- Shwetak Patel's group in Washington [1], Dina Katabi's group in MIT [2], Patwari's group in Utah [3], etc.
2. MIT OCW on how to build a radar system with advanced [4] capabilities http://ocw.mit.edu/resources/res-ll-003-build-a-small-radar-...
[1] https://ubicomplab.cs.washington.edu/
[2] http://groups.csail.mit.edu/netmit/wordpress/projects/projec...
[3] http://span.ece.utah.edu/neal-patwari
[4] I use 'advanced' in the sense that it is way, way, more sophisticated in its use of math and signal processing/machine learning/linear algebra algorithms than your standard automotive speed monitoring radar.