Oh FFS
Im all for technoporn but really? Are they building a fast car or some sort of mini-tech empire?
Engineers from the Bloodhound supersonic car world record attempt have been testing their radio comms gear as plans to stream the 1,000mph (1,610kmph) event live come together. Live-streaming the car's record-breaking run is no trivial matter. Given that you can’t even maintain a phone call on a London train, trying to …
"and doing it in the Kalahari desert is nigh on impossible"
Agree, it'll be a challenge to make it work, but really? Getting a wireless data link in a flat open area with no buildings or obstacles to block or reflect the signal, and probably virtually no other transmitters or sources of interference for a large number of miles in any direction is harder than elsewhere?
Surely the problem is solved rather than testing P2P links traveling in opposite directions, just have a relay point some distance off the "runway" such that the relative speed of distance change is not that great if not zero at the point where the vehicle is expected to hit maximum velocity...?
(I'm not going to suggest that the car drives around to maintain an exact distance, because ground based vehicles traveling at that speed don't do turning very well...)
My simple calculations tell me that the car will (hopefully) be travelling at up to 447 m/s
The radio signals will be travelling at something in the vicinity of 3x10^8 m/s
Therefore the doppler shift will be..... miniscule. ~1.5ppm. Well within the PLL compensation of most radios.
I see this test as more of a publicity thing for the comms suppliers than anything else.
My simple calculations tell me that the car will (hopefully) be travelling at up to 447 m/s
The radio signals will be travelling at something in the vicinity of 3x10^8 m/s
Therefore the doppler shift will be..... miniscule. ~1.5ppm. Well within the PLL compensation of most radios.
I see this test as more of a publicity thing for the comms suppliers than anything else.
I was wondering the same thing. I in no way claim to know what I'm talking about here, but it seems to be if we can communicate with commercial aircraft pilots in real-time whilst they whizz around earth at 700 mph then this seems quite straightforward. Hell, we communicate with astronauts in spaaaaace at about 17000 mph quite happily.
It's not the Doppler effect, it's the timing within the protocol that gets screwed up.
GSM is limited to speeds well below this - the railway variant GSM-R has specially widened guard bands to allow operation on TGV and other HS trains. If they've got 4G working at these speeds, and at sufficient bandwidth to stream HD video, that's impressive.
With GSM it's not just about the hand-off for speed. GSMA is a TDMA technology, it has eight timeslots, The cell site flips between the seven people all using the slots (there is one not used for calls), but this needs the site to know how far away each phone is and allow for the signal to reach the user, albeit at the speed of light.
If the car (or whatever) is moving too fast the signal can't hit the timeslot. But as Bloodhound is using 4G which doesn't have slots it's not an issue.
Simon
So by my read of the above (thanks - I learned a lot today), it *is* sort of to do with Doppler, but not on frequency, on timeband multiplexing.
Again, if the base station needs the cell to be at a known distance (rather than very specific frequency) my original suggestion of just placing it far enough away from the runway would work, as the change in distance away will be dampened by Mr. Sine and Cosine as the car drives past.
You could even place many cells parallel to the route and pick up the one that the car/plane is passing.
Or you could solve the problem by not using timeband multiplexing.
Perhaps they should have a driver change though, may I suggest Jeremy Clarkson, though the weight penalty might be too much, he's much more disposable.
Seriously though, the technical challenges they need to solve, not least how to keep it stable and on the ground will be the long term benefit. I would have though the speed issue for comms had been solved by now, military aircraft have been flying a lot faster than this for some time. Mind you if you are developing a low cost mobile 4G solution for the third world, flatness aside.
Live HD video from Bloodhound is part of the pay-off for the sponsors.
I watched the last Falcon launch , and we have come a long way from the days of Apollo, when it needed a cine camera on the Saturn V, and there was no guarantee of recovering the film. We have a few, much-used, shots of the booster seperation. With Falcon, we were seeing the inside of a fuel tank, in zero gravity, live.
We expect more. The Bloodhound team are making sure they can deliver.
Meanwhile, they keep trying with Falcon, and the first successful landing of the booster on Just Read The Instructions is going to be stunning video. We don't need that video to be live but, like Apollo XI, it's something that people will remember.
But why are some people fussing about hoverboards in 2015?
No such thing. Literally does not exist.
Back in the analog days, the action-to-image latency of live TV could be measured in milliseconds, because nobody unintentionally buffered anything. These days, "Live" HD video is often ridiculously buffered. During one rocket launch, we had 45 seconds *delta* latency between two "Live" streams. Crazy.
Try watching New Years countdowns accompanied by an accurately set clock. Simply nonsense. A minute off is common. A minute!!
Even simple audio. By the time the BBC WS beep comes out of the speaker in Canada, it's 15 to 20 seconds late. Useless.
Too many engineers fail to consider latency. All they have to do is just think about it and implement some easy tech to get it back into the sub-one-second range. Not always possible, but most latency is just laziness or lack of thinking.
We had a digital video project. At the outset latency was considered. So end-to-end digital video was a small fraction of a second. Just need to specify it.
> Too many engineers fail to consider latency.
It's all bandwidth bandwidth bandwidth and so often its latency that counts.
No putting the proxy server for the Sydney office in London is not a good idea, really it isn't. No I know you think you've got a fast Internet connection, but...