HI! I have been searching through datasheets, user manuals, etc.. And for the love of god, I can’t find any AoA FOV specs. Does any of your UWB ICs support 2D 360°AoA?
Thanks in advance,
Bob
Hi Bob,
We have support for 2D 360 AoA in our QM35825 chipset. You can find a little bit of information on that here. For additional questions on this topic, please contact your local sales representative: Contact a Qorvo Sales Representative - Qorvo
1 Like
By 360 degrees do you mean that it can solve the ambiguity as to whether the signal is from in front or behind the antennas? Or do you mean it can detect from any angle but the user needs to apply some additional logic to calculate which hemisphere the signal is from?
1 Like
Sorry, it looks like I forgot the link from my previous post. Here it is: Samples/Python/UWB-Qorvo-Tools/scripts/fira/fira_360_aoa/README.md · main · QORVO_SDK / Public / DevKits / qm35-sdk · GitLab
We have the capability for both without the additional logic but it requires an antenna that is not part of the standard DK.
1 Like
Hello,
sorry for the late response.
I am making a student proof-of-concept and my budget is quite limited. I would just like to know when a tag is in a focal cone of 10 degrees in front of my anchor. With no interference from tags that are all across the room.
Sincerely,
Bob
@akash Could you recommend an IC for that or should I contact the sales rep? 
Our other development kit option for AoA is QM33120WDK1 - Qorvo but this chip (QM33120W) only supports 2D AoA in one plane.
Hello. I am also interested in getting 2D 360 degree AoA. Can this be done with a simple Tag (possibly a QM35825) and an anchor - another QM35825. I need the anchor to know, in 2D, the distance and angle of the tag.
I see there is the QM35825-05 dev kit. @AndyA you mentioned a different antenna required, were you referencing this devkit? If so, can I get hold of the antenna?
Does any of your UWB ICs support 2D 360°AoA?
The QM35825 does support up to 4 antennas so it could in theory be used to make a 4 antenna AoA system. You need at least 3 antennas to disambiguate angular location in a 2D plane, but more is better to refine the precision.
You don’t need a single chip to do this, however. Long ago, using DW1000, we built a 5 antenna phased array system for full coverage using AoA and TWR to locate a tag.
Pardon the crude video, that was 5+ years ago.
The basic design is 5 separate UWB chips receiving at the same time, tied to the same reference clock, and extracting the phase of arrival. We then run that through an algorithm to find the angle of arrival. The tricky part is that we spaced the antennas much further apart than 1/2 wavelength, so we have to deal with integer phase ambiguities, but the benefit is that we get very high AoA precision, less than 1 degree. It provided a 3D answer, azimuth, elevation, and distance.
The QM series have better phase measurement than the DW1000 had, so AoA projects work better with them and we could improve the concept using QM, probably to 1/2 degree using the same geometry. Also, if you can do two interactions on different channels, that can help with precision and phase ambiguities. There is a tradeoff of antenna count, antenna spacing, and array size that determines your precision. We found 5 antennas was significantly better than 4 (which had ambiguity problems) and 3 (which is minimum and therefore produced fragile results). Our largest system was 7 antennas which didn’t add much to the results, so 5 was the best fit.
The basic take away is that you don’t need a single chip to do precise 360 degree AoA, but you do need each receiver to be tied to the same reference clock, and you need some non trivial coding to make it all work.
We didn’t commercialized the technology, though we may circle back to doing so at some point. The basic value was being able to install one device and track an entire room with it, which reduced the cost and install effort.
Mike Ciholas, President, Ciholas, Inc
3700 Bell Road, Newburgh, IN 47630 USA
mikec@ciholas.com
1 Like