Rendered at 08:09:01 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
mysterydip 1 days ago [-]
I was about to say, “coolnamesalltaken already has a pretty small one,” but you are them! I’m definitely excited by this project, I have something I wanted to add ADS-B integration to and this looks like an even better option!
As for a question, how do you balance low power with performance? Is there a target range or use case to decide the performance envelope?
CoolNamesAllTkn 1 days ago [-]
Ha, it's very fun to see that people know about the project!
ADSBee 1421's very low power consumption is determined by its architecture. Where ADSBee 1090 has a discrete RF frontend with multiple power hungry LNAs and a power detector chip, plus a dual core RP2040, an ESP32 S3 module, and a CC1312, ADSBee 1421 just has an LR2021 chip and a CC1314. A single LNA on the ADSBee 1090's RF frontend actually draws about as much power as the whole m1421!
The tradeoff is that we have fewer CPU cycles available for crunching packets compared to the ADSBee 1090, and the LR2021 has some quirks that result in it only being able to receive ADS-B packets in its maximum dynamic range configuration (so we support Mode S DF17 only, no squitter packets or altitude / surveillance replies like you get with DF4/5/11). That said, this has been plenty for customers who are looking for an aircraft detect and avoid solution plus UAT traffic / weather.
Based on our testing so far, ADSBee m1421 is well suited to general aviation applications (portable flight bag / cockpit avionics), drone flight controllers (aircraft detect and avoid), and anything that needs to be low cost or easy to run off a solar panel or little battery.
For customers looking to build something like an airport ground station, where they want all squitter and extended squitter packet formats and a higher max packet rate, we still recommend the ADSBee 1090 three-MCU architecture.
mysterydip 21 hours ago [-]
Thanks for the info! That helps with the decision.
btreecat 20 hours ago [-]
Finally UAT support!
How's the software side though?
Last time I looked dump1090 worked well, but there was no way to decode UAT packets.
Something on my to-do list, that's now become more plausible with Claude assisting the research.
CoolNamesAllTkn 15 hours ago [-]
On the SDR side, dump978 is what most people use, but SDRs are generally not well suited to embedded applications in terms of power / form factor / cost.
Our firmware is fully open source, and our UAT decoder was introduced last year and has been working well (I think we've shaken all the bugs out at this point). We implemented everything from scratch so that it would run efficiently on our embedded RF MCU (TI CC1314). We support both UAT ADSB and UAT uplink (weather / traffic data).
We're focusing on embedded applications with ADSBee m1421 applications right now but it's definitely tempting to roll m1421 into a smaller flight bag device. It's something we might get back around to once we have Winglet off the ground.
As for a question, how do you balance low power with performance? Is there a target range or use case to decide the performance envelope?
ADSBee 1421's very low power consumption is determined by its architecture. Where ADSBee 1090 has a discrete RF frontend with multiple power hungry LNAs and a power detector chip, plus a dual core RP2040, an ESP32 S3 module, and a CC1312, ADSBee 1421 just has an LR2021 chip and a CC1314. A single LNA on the ADSBee 1090's RF frontend actually draws about as much power as the whole m1421!
The tradeoff is that we have fewer CPU cycles available for crunching packets compared to the ADSBee 1090, and the LR2021 has some quirks that result in it only being able to receive ADS-B packets in its maximum dynamic range configuration (so we support Mode S DF17 only, no squitter packets or altitude / surveillance replies like you get with DF4/5/11). That said, this has been plenty for customers who are looking for an aircraft detect and avoid solution plus UAT traffic / weather.
Based on our testing so far, ADSBee m1421 is well suited to general aviation applications (portable flight bag / cockpit avionics), drone flight controllers (aircraft detect and avoid), and anything that needs to be low cost or easy to run off a solar panel or little battery.
For customers looking to build something like an airport ground station, where they want all squitter and extended squitter packet formats and a higher max packet rate, we still recommend the ADSBee 1090 three-MCU architecture.
How's the software side though?
Last time I looked dump1090 worked well, but there was no way to decode UAT packets.
Something on my to-do list, that's now become more plausible with Claude assisting the research.
Our firmware is fully open source, and our UAT decoder was introduced last year and has been working well (I think we've shaken all the bugs out at this point). We implemented everything from scratch so that it would run efficiently on our embedded RF MCU (TI CC1314). We support both UAT ADSB and UAT uplink (weather / traffic data).
Our firwmare repo is here: https://github.com/PantsForBirds/adsbee. It's always open for bug reports or PRs!
A normal non ADS-B iPad GPS flight module is almost 10% the price of the Sentry
https://www.garmin.com/p/645104/
We're focusing on embedded applications with ADSBee m1421 applications right now but it's definitely tempting to roll m1421 into a smaller flight bag device. It's something we might get back around to once we have Winglet off the ground.