Streaming live airborne video of wildfires

[Mark Zaller]

Hello,
I'm an Air Attack pilot wanting to stream visible and IR video real-time to the ground over a .5-1 Mbit link. Is there someone or a company who can help with a cost effective solution? I'm currently supporting several aircraft on the Oregon fires that would like to do this now.

Areas I'm exploring and would appreciate smart people to help with:
- Live web-cam streaming. This would transmit adequate visible video, but not the aircraft's IR (320x240 NTSC). Any suggestions for the best approach?

-Streaming live video selected in the cockpit and captured on the plane's computer. What is the best software for this kind of live streaming?

-Screen capture like Microsoft Expression. On a Tablet PC (already in the cockpit) we could position both Visible and IR video windows next to the the GPS coupled moving map to transmit the screen. This provides the immense benefit of showing moving map exactly where the video is being shot, and reducing narration from the Air Attack team.

NTSC/640x480 is great, but 320x240 is ok if needed to minimize bandwidth. Workflow in the air must be extremely easy. Pilots and the aerial firefighters (ATGS-Air Attack Group Supervisor) are very busy, and often shy away from computers. Usually On/Off is all they will do. Once On and transmitting the fix mounted cameras to the ground, they are willing to occasionally talk over the FM radios to narrate for ground people. On the ground the UI must also be easy...hopefully just a web address the fire commanders can bookmark.

I run a non-profit for Aerial firefighters (Aerial Fire Tech), and would appreciate any help I can get.

-MarkZ
mszaller@yahoo.com

[Bill Davis]

Mark,

I don't know of ANY working system that could do what you seek.

In fully developed areas such as big cities, it's possible to establish reasonably stable ground link communications between an aircraft (typically a news helicopter) and ground stations via a system of microwave links - but that requires significant permanent infastructure.

You're asking to export a similar system out into the wild? Dropping fully functioning ground stations on a variety of mountain peaks? Good luck with that. And with the subsequent need to tether them to power and/or run additional links to a master ground station.

The only folks I know about who can do a form of this is the US military with their battlefield of the future initiative that shuffles real-time video from drones and other moving assets back to base stations.. And their budget is a LOT over what ANY of us is likely to be able to afford to rent on the open market.

Or maybe you'd like to station some guy in the open window of the aircraft with a dish and have them "AIM" the microwave transmitter at a receiver station on a mountain peak? What happens if the plane has to bank?, or gets buffeted? Or simply keeps moving forward moving farther and farther out of range.

Whole thing, in fact, reminds me of the old Al Franken SNL gag when he strapped a "satellite station" to his head and we the viewers lost the signal every time he turned to answer someone's question or looked away.

Nice in theory - but unless there's been a HUGE leap in tech since I became familiar with broadcast tech - this is not going to be simple, cheap or easy to do.

Camcorder capture and then process on the ground? Easy. Live aircraft to ground feed = not so much.

If anyone knows different, I'd be very interested to hear about it.

Sorry.

[Tom Majeski]

I do this for UAV's. Without going into detail, you're talking $15K to $20K in tracking antenna and RF equip. If your video bandwith and telemetry and gps data add up to around 300 K data rate, figure about $6 to $7 PER MEG of Inmarsat satellite time. That adds up fast.

Another option is the Bread Crumb system. I've used it in the past for setting up a network in the middle of nowhere for UAV video and aircraft telemetry. Their high power boxes cover for 10 miles line of sight. Still high priced, but less than satellite time. See Rajant.com.

[Mark Zaller]

We have a working communication link. We email down photos and maps now. I'm weak on how to do the video streaming across an existing .5-1 Mbit internet link.

I am hoping for software to automatically push our video capture (IR & visible) down to an existing ground distribution server. Maybe MSN or something else? No one on the fires is currrently budgeted for the video server, and I don't have a place to do it myself.

Our internet link isn't perfectly consistent because of the airborne remoteness, so peer to peer or web server in the sky is not my first choice, but maybe an approach. Help on how to do this with some fault tolerance would be appreciated.

-MarkZ
mszaller@yahoo.com
408-623-4303

[Ed Roo]

Find some HAM radio operators in your area. They may know someone local who can rig up something inexpensive.

[Bill Davis]

I still don't see how you're going to pull this off.

You mention .5 to 1 MegaBIT transfer rates.

Even with PhotoJPEG compression at 25:1 - single stream video takes 1MegaBYTE a second - or 8 times the bandwidth your current system runs at.

Yes, you might try dropping frame rates down to 10fps, or knock your frame sizes down to 160x220 or even both, and eventually get a small, blocky picture that MIGHT stream in real-time using standard tools like you're discussing. But I can't see a practical way to stream live video from a moving target using this kind of bandwidth gear with any practical "lock on" stability.

Plus, if you're shooting earthward from a fast moving option like an aircraft - you're generating video that is the MOST difficult to compress - scenes where every pixel is refreshing rapidly with distinctive content.

As other's have mentioned, it's not IMPOSSIBLE. And if you can throw enough money at the task, there's nothing technically stopping you. But I think you're going to find that if you want to create a reliable broadcast to ground station system with real size aircraft over realistic real-world terrain and enough geographic coverage to make the system practical for simple remote field deployment - you're likely looking at something approaching at LEAST mid five figures, and more likely well into six figures before you're all done.

I'd LOVE to be proved wrong about this. And if you find a solution - PLEASE let us all know.

This would be a GREAT boon out here in the southwest as well where real-time video coverage with decent base station monitoring of remote areas would help a LOT of industries and make folks a lot of money.