Our friend [Hunter Scott] gave a talk at a past Supercon about phased array antennas. He mentioned he was looking for collaborators to create an antenna with the SiBeam SB9210 chip. This is a specialized chip for WirelessHD, a more or less failed video streaming protocol, and it’s essentially an entire 60 GHz phased array on a chip with both transmit and receive capabilities. For $15, it seems like quite the bargain, and [Hunter] still wants to put the device to work.
The downside is that Lattice bought SiBeam and killed this chip — not surprising considering WirelessHD never really took off. However, [Hunter] says the chip was in some old smart TVs and laptops. If you can find replacement boards for those devices on the surplus market, you can get the chip and the supporting circuitry for a song.
The situation is a little sticky. [Hunter] has the datasheets for the parts, but is still bound by a nondisclosure agreement. He’s still working on getting that encumbrance removed. But until then, he has some advice about similar chips that have public datasheets and a controller chip that — if you would sniff its bus in a working system — might very well shed light on how to set up the cheap antenna chip.
There are also a bunch of public links that should make things clearer. However, we hope that Lattice will allow the NDA to expire on the datasheets. Meanwhile, you can catch the Supercon talk that started it all in the video below.
You might be interested in visualizing phased arrays. These are really common in radar systems.
Very satisfying to hear that their NDA’d chips were a total failure. Bad enough to have a black box chip, but when even the documentation needed to use it is hidden it’s nice to see the company lose money on it.
Ha, the schadenfreude. Makes sense to me. Would be even better poetic justice if they opened it up and made the product come back to life again since now people can actually, y’know… do stuff with it.
Nice, but I don’t think that was the only reason the idea didn’t take off. After all others could have came out with non-NDAd chips if they wanted to.
I agree the intended use was pretty hopeless given the competitive technologies, but if they had made it possible for people to use it by providing the documentation, it’s anyone’s guess what kind of products may have come out of it. Certainly wouldn’t be the first time a chip found it’s main use in something the manufacturer never thought of.
My company produces a lot of what we call vertical market chips: complicated and intended for a dozen specific large corporate customers. We don’t give out the documentation to anyone else, specifically because we don’t want other people trying to produce stuff based on it. Here’s why: because we spend more on engineering support than we get in revenue on these parts, unless they’re selling in quantities of 500,000+ to a single customer.
I don’t like it, but for a lot of specialized parts, the business case for broad market is not only nonexistent, it’s a money pit.
Failure had nothing to do with NDA, it was due to market changes. There are more NDA-protected IC’s out there than you can count. Market is competitive, which is a good thing, and result of that is that anything cutting edge is NDA-only. If you were doing something market relevant and had financial base to justify it you could have access too if you were willing to sign your life away on that NDA like I do on a regular basis. There is nothing bad about a “black box” product, and in many cases, it’s the only way to make money. If there were no way to make money with it, the chip wouldn’t exist. This is called reality, deal with it.
Having been a AN-TPQ 37/47 and Sentinel RADAR tech in my former life and knowing just how precise of a location tracking/aiming you get with phased array beam stearing, I was a little creeped out to hear that that’s the tech 5g is based on.
No more than a femtocell with a built-in GPS. Not to mention always on location services in Android and practically every phone having GPS because of e911.
I’ve never seen that side of 5G discussed. It’s always some tin foil hat stories about the frequencies, while accurately locating anyone using the technology seems a much more realistic and acute concern. Granted, the current situation isn’t much better with high resolution maps of local networks (cell, WIFI and Bluetooth) being made by the likes of Google, so a rather accurate location of any device relaying the local networks can be established without ever using GPS.
I wonder whether older phones can also be located with newer 5G towers. I’m not sure how backwards compatible the technology is. Thanks for the interesting insight.
5g does NOT use phased arrays in the same way as radars do…it just semi-randomly mangles the output characteristic to best match the conditions. Proper 2-D (or even spherical if you’re masochist enough) phased arrays come with a proper price tag.
Remember that unlike radars, 5g has to actually deal with multipath returns and the signal going through stuff, doppler radars designed to find and track moving stuff deliberately ignore all of this…
5G has full on tracking and quite accurate 3D beam forming. It’s even smart enough to not only use beam forming on multiple targets at once, but also to use null steering to move the low/zero signal areas which also form during beam forming between the high signal lobes actively over other targets to reduce interference.
While I’m not acutely aware of how phased array radar works and whether this is comparable, it’s enough to pinpoint the location of any single target in three dimensional space from a single tower location.
Phased array beam forming is seen as instrumental in 5G, as the high frequency and associated absorption rates would require very high powered classical transmitters, which cause all kinds of safety, legislative and financial issues.
5G New Radio location awareness, under ideal conditions as a massive MIMO will multilaterate the outdoor position of devices in 3D space to within a decimetre. And sub-meter indoors or outdoors within 2D space. At worst (in a tall skyscraper) your location awareness would be within ~30m.
In many ways it is no different than having a GPS in your phone or even WiFi. Your phone already knows its position quite well, the main difference now is that your network provider does as well by default at a higher granularity then before (LTE ~15m to ~1.5km – cell size – not using multilateration). And this granularity is required for the higher data rates and more efficient use of spectrum.
RMS is said to have no phone in his hand precisely for this reason, since GSM era. It may be more accurate now or tomorrow but not new.
What’s the betting that 7/10ths of the data is about ways not to create ozone and poison your customers. Given that 60ghz is strongly absorbed by O2. Don’t cross the beams! :-D
uhhh, miniscule output power, an open window in the summer will cause a lot more ozone then this ever could…