Hi all. Thanks for the interest. I'm going to post an impromptu blog post here just explaining what this is. The rather unfortunate ground stop today made this unexpectedly useful with the pilots here on Orcas just to get a sense of the chatter in our airspace re the situation.
This is an ETTUS B210 attached to a dual 4090 box. It's a polyphase channelizer that takes the entire airband, channelizes it using cuda code on GPU, demodulates it, detects transmissions, throws away data tx and then feeds the tx through a voice model I created for transcription. A separate LLM reads the transcriptions to create a situational awareness picture of the airspace.
I'm interested in open sourcing this and making it available for folks who can run an ETTUS B210 or high quality SDR with enoguh bandwidth and a local GPU machine, so that we can cover more airpsace.
This was born out of an incident we had not too long ago where a pilot around KORS (based at Friday Harbor) lost the engine in his 172, and put it in the water in front of a ferry and was picked up by the ferry. We're a pretty close bunch on Orcas Island and we only heard about it days later via social media. That felt completely unacceptable, and I set out to create a radio that could monitor everything in real-time, all the time, and alert on emergencies. That has extended to transcribing and creating situational awareness for a chunk of airspace.
I'm an IFR pilot. I fly a T206H regularly in IMC around here and I'm a ground instructor. You can email me at mark@defiant.com or find me on x at mmaunder if you'd like to connect. I'm particularly interested in hearing from folks who can run a station on their own hardware and are under bravo or a charlie or even just somewhere interesting re airspace.
Edit: OMG I just realized who you are. I used to read your blog obsessively before I became a pilot, and while I was working heavily on mysql. Thanks for that!!
This reminds me of a Claude hobby project from a while back that listens on public KiwiSDR receivers (on-demand, politely, never taking the last slot) and decodes ships’ DSC radio signals: https://research.datadesk.eco/skywave/
These are mostly quite boring, but I did manage to make some progress in triangulating position between multiple receivers using the delay, which would be useful as a method for tracking AIS-dark vessels.
These are raw transcripts unedited by me that I extracted possibly related to the groundstop, generated at time of posting:
2143Z, 126.400 Victoria Ground (weak signal): "Yeah, we just uh communication issues, so we got the traffic stopped on [unintelligible]"
2151Z, 134.800 Vancouver Centre, pilot of a flight ending "364": "We're saying they just have an alternate plan. They've been getting lower and lower 48 right now. We're on
the phone with you, they are saying cannot show any reason why we would be out of communication to Seattle. So just otherwise we do have some alternates that we're talking
about, but we just want to see, you see anything on your end that we cannot proceed to Seattle?"
2152Z, 134.800 Vancouver Centre, same pilot: "Okay, sounds good. Yeah, we appreciate that. We'll be staying by and we'll just kind of get the alternate plans for ready to
go. Thank you."
2159Z, 121.075 Victoria Terminal, controller: "Seacraft 304, just a heads up here, there is a ground stop for an aircraft about into Vancouver at this time. I'm sure you
have a VFR slot, seeing where you're headed there now, so I'd like to be still going to take you, but it just has stopped that you might end up getting turned around."
2159Z, 121.075 Victoria Terminal, controller: "Sorry, that's all inbound aircraft, ground stop."
2200Z, 121.075 Victoria Terminal, controller: "Yeah, it's not affecting anything airborne at this time. We're just about departing anything for Vancouver and Calgary at this
time."
2201Z, 126.400 Victoria Ground, pilot (weak signal): "Hey Ground, it's Jazz 270, and I'm guessing there's a flow for Vancouver."
2203Z, 121.075 Victoria Terminal, pilot: "Uh just wanted to confirm uh there's a ground stop uh going into Vancouver on this frequency?"
2203Z, 121.075 Victoria Terminal, controller: "Encore 3325, that's correct. Nothing is departing to land at Vancouver at this time."
2204Z, 134.800 Vancouver Centre, pilot: "Yeah, right now they said normal ops continue, but we'll see what happens once we get closer."
2217Z, 126.125 Vancouver Departure, pilot: "Yeah, you guys not having any of the issues that they're having in the US right now?"
2217Z, 126.125 Vancouver Departure: "Uh, apparently the ones uh seeing ground stop for the 14 major airports I've just got a few minutes. Okay, thanks."
2227Z, 126.400 Victoria Ground, pilot: "Encore 58, information Papa, requesting flow through to Vancouver for 2255." Reply: "Roger, 2255 on the flight plan."
Also, the recordings you hear online will often be more garbled (weaker signal) than what you hear in the cockpit or what ATC hears, because (1) in the air you have clear line of sight to the ground transmitter and to other planes, and (2) ATC has receivers that are designed to be able to cover the area they're controlling.
Ah this is such a hot debate in the community. Yes it's straight up AM simplex and the argument for is: With AM when you step on another tx it screeches. Also pilots can hear pilots to improve situational awareness vs duplex which would have pilots only hearing ATC. Huge legacy fleet of aircraft. Cost to upgrade anything in a certified aircraft. There are so many arguments for digital though. it's a tough one.
You get used to it. There's like three things you're likely to hear based on context, and deciding which one it was isn't too difficult. Discerning additional detail and nuance (left/right on course) then becomes a much lower cognitive load. It is definitely a learned skill, but it comes automatically with time.
The counterpoint to that is that for the same SNR where you're at barely (or not at all) audible or garbled on analog, you have absolutely clear digital audio.
Yeah, and with most comms being stereotyped, most messages carry little data, meaning you can easily add redundancy and still take up the same or less airtime. You will still need plaintext messages for complex situations, but even that can be transmitted more clearly and efficiently with digital encoding.
I’m supposed to be flying tomorrow afternoon for my wedding next week. I fly to Toronto and then Mexico, so no stops in the US but needless to say I’m worried.
Not trying to promote myself, but there's another project I created to provide pilots with IFR situational awareness so we know if there's going to be a ground delay when we call for clearance. It's a bit technical, but you can use it to check your own route. It's at https://ifrmeter.com/
Yeah sorry about that. This is a software defined radio based on a hangar at Orcas island, receiving the entire airband, channelizing it into over 700 channels, and then demodulating the channels and playing whatever is sent provided it's voice. It also transcribes using a voice model and an LLM parses the output and attempts to create cards on the right hand side. It's also trying to detect simulcasts and remove the duplicate transmissions. Hope that helps explain it. It's also archiving everything so you can listen to old transmissions. Right now I'm working to keep the server up. Thanks HN. :-) For context, I'm a pilot based at KORS and a nerd.
I don't think GP is confused about the technical details, they are confused about what they are looking at and what the point is. I think the answer they were looking for is (correct me if I'm wrong) "this is a live feed of transcribed ATC comms in the Vancouver area," and possibly some evidence that there is a ground stop since things seems to be running as normal based on the most recent logs.
Haven't listened by KORS is a small untowered airport in the San Juan islands, in the northwest corner of WA state. It's very near Victoria, Canada, whose ATC controls the nearby airspace, extending into US territory slightly.
The post title is implying that Victoria airport is also subject to a ground stop.
Hi all. Thanks for the interest. I'm going to post an impromptu blog post here just explaining what this is. The rather unfortunate ground stop today made this unexpectedly useful with the pilots here on Orcas just to get a sense of the chatter in our airspace re the situation.
This is an ETTUS B210 attached to a dual 4090 box. It's a polyphase channelizer that takes the entire airband, channelizes it using cuda code on GPU, demodulates it, detects transmissions, throws away data tx and then feeds the tx through a voice model I created for transcription. A separate LLM reads the transcriptions to create a situational awareness picture of the airspace.
I'm interested in open sourcing this and making it available for folks who can run an ETTUS B210 or high quality SDR with enoguh bandwidth and a local GPU machine, so that we can cover more airpsace.
This was born out of an incident we had not too long ago where a pilot around KORS (based at Friday Harbor) lost the engine in his 172, and put it in the water in front of a ferry and was picked up by the ferry. We're a pretty close bunch on Orcas Island and we only heard about it days later via social media. That felt completely unacceptable, and I set out to create a radio that could monitor everything in real-time, all the time, and alert on emergencies. That has extended to transcribing and creating situational awareness for a chunk of airspace.
I'm an IFR pilot. I fly a T206H regularly in IMC around here and I'm a ground instructor. You can email me at mark@defiant.com or find me on x at mmaunder if you'd like to connect. I'm particularly interested in hearing from folks who can run a station on their own hardware and are under bravo or a charlie or even just somewhere interesting re airspace.
That's so cool... I'm a pilot (have 3 planes) and would be interested in hosting one--or just running for my own edification, really.
I'll message you.
Awesome. Thanks.
Edit: OMG I just realized who you are. I used to read your blog obsessively before I became a pilot, and while I was working heavily on mysql. Thanks for that!!
This reminds me of a Claude hobby project from a while back that listens on public KiwiSDR receivers (on-demand, politely, never taking the last slot) and decodes ships’ DSC radio signals: https://research.datadesk.eco/skywave/
These are mostly quite boring, but I did manage to make some progress in triangulating position between multiple receivers using the delay, which would be useful as a method for tracking AIS-dark vessels.
If you scroll back to 2:03pm (the time this link was posted) there was this interaction:
14:03:04 Uh just wanted to confirm uh there's a ground stop uh going into Vancouver on this frequency?
14:03:10 Encore 3325, that's correct. Nothing is departing to land at Vancouver at this time.
These are raw transcripts unedited by me that I extracted possibly related to the groundstop, generated at time of posting:
OG scanner transcriber https://scanner.rachelbythebay.com/
Surprised at how garbled many of these transmissions come across.
I can barely understand 50% of what is being said and am impressed by these professionals who can.
Are the transmissions still analog for security / interoperability reasons?
Also, the recordings you hear online will often be more garbled (weaker signal) than what you hear in the cockpit or what ATC hears, because (1) in the air you have clear line of sight to the ground transmitter and to other planes, and (2) ATC has receivers that are designed to be able to cover the area they're controlling.
Ah this is such a hot debate in the community. Yes it's straight up AM simplex and the argument for is: With AM when you step on another tx it screeches. Also pilots can hear pilots to improve situational awareness vs duplex which would have pilots only hearing ATC. Huge legacy fleet of aircraft. Cost to upgrade anything in a certified aircraft. There are so many arguments for digital though. it's a tough one.
You get used to it. There's like three things you're likely to hear based on context, and deciding which one it was isn't too difficult. Discerning additional detail and nuance (left/right on course) then becomes a much lower cognitive load. It is definitely a learned skill, but it comes automatically with time.
You can still understand a noisy analogue transmission. With digital you get it or you don't.
The counterpoint to that is that for the same SNR where you're at barely (or not at all) audible or garbled on analog, you have absolutely clear digital audio.
Yeah, and with most comms being stereotyped, most messages carry little data, meaning you can easily add redundancy and still take up the same or less airtime. You will still need plaintext messages for complex situations, but even that can be transmitted more clearly and efficiently with digital encoding.
Only ground stop is the FAA one: https://spaces.navcanada.ca/workspace/ois/
I’m supposed to be flying tomorrow afternoon for my wedding next week. I fly to Toronto and then Mexico, so no stops in the US but needless to say I’m worried.
Not trying to promote myself, but there's another project I created to provide pilots with IFR situational awareness so we know if there's going to be a ground delay when we call for clearance. It's a bit technical, but you can use it to check your own route. It's at https://ifrmeter.com/
What model are they using for the speech to text transcription?
I'm confused by this one, when I click the link I am seeing a lot of live traffic?
EDIT: The title is confusing to read, there's 2 statements.
Yeah sorry about that. This is a software defined radio based on a hangar at Orcas island, receiving the entire airband, channelizing it into over 700 channels, and then demodulating the channels and playing whatever is sent provided it's voice. It also transcribes using a voice model and an LLM parses the output and attempts to create cards on the right hand side. It's also trying to detect simulcasts and remove the duplicate transmissions. Hope that helps explain it. It's also archiving everything so you can listen to old transmissions. Right now I'm working to keep the server up. Thanks HN. :-) For context, I'm a pilot based at KORS and a nerd.
I don't think GP is confused about the technical details, they are confused about what they are looking at and what the point is. I think the answer they were looking for is (correct me if I'm wrong) "this is a live feed of transcribed ATC comms in the Vancouver area," and possibly some evidence that there is a ground stop since things seems to be running as normal based on the most recent logs.
Yes, that's it!
What SDR did you use that has the bandwidth to capture the entire airband? You need to scan about 20MHz for the voice channels.
What STT system / model are you using?
I had a plan to try something like this from where I live, but I wasn't sure how to solve those two problems.
If you're trying to build something that integrates with the Linux desktop, check out dabri https://github.com/AshBuk/dabri
Not mine, I just found it 2 hours ago.. No mention of it on HN so not sure what others think about it.
If someone who understands what this is feels like explaining it, that would be very cool :)
Haven't listened by KORS is a small untowered airport in the San Juan islands, in the northwest corner of WA state. It's very near Victoria, Canada, whose ATC controls the nearby airspace, extending into US territory slightly.
The post title is implying that Victoria airport is also subject to a ground stop.
Please see my other reply. Happy to answer any questions in between trying to keep the server running. Amazing how well node performs under load.
Ok, thanks for that!
This is very nice, I always wanted something like this.