BBC Three to return in January 2022 as broadcast channel

D

Deleted member 473

I thought single frequency on DAB was touted as a good thing, seamless listening experience when driving around.🙉
 

gomezz

Well-Known Member
I no longer listen to FM in the car ever since I went through the car wash forgetting to push the FM aerial down. 😊
 

Black Hole

May contain traces of nut
I thought single frequency on DAB was touted as a good thing
Yes, and somebody on here tried to claim digital broadcasting and DSP made that possible, but the physics says that ain't so. The areas roughly equidistant from multiple transmitters must suffer fringing.

But maybe "single frequency" is only an approximation and it's actually "different but closely separated frequencies within one channel".
 
OP
O

Owen Smith

Active Member
Yes, and somebody on here tried to claim digital broadcasting and DSP made that possible, but the physics says that ain't so. The areas roughly equidistant from multiple transmitters must suffer fringing.

But maybe "single frequency" is only an approximation and it's actually "different but closely separated frequencies within one channel".

No it really is single frequency. But the guard interval (period during which nothing is transmitted) allows the receiver to see that there are multiple transmitters, and with DSP it actually is possible to receive one and reject the others. But what you have to be careful to avoid is distant transmitters beyond the guard interval time being powerful enough to interfere with the local transmitter. Or transmitters too close to each other. It's a delicate balancing act, and the guard interval does cause a loss of bandwidth. That is why Freeview COM7 as an SFN has less bandwidth than the BBC HD mux, COM7 has a much larger guard interval. You can see the guard interval in the HDR Fox T2 tuning menu.

MIMO wifi can use multiple antennas all transmitting and receiving on the same frequency which doubles or triples the bandwidth available, again because the DSP can pick out which one they want based on the spatial separation of the transmitting antennas.

To anyone with some knowledge of analogue reception and interference and fringing, it does indeed seem impossible that this can work. But it does.
 
D

Deleted member 473

I think Owen is saying that it does work, but not in the real world, unless you don't have enough transmitters, or perhaps too many.
 
OP
O

Owen Smith

Active Member
I think Owen is saying that it does work, but not in the real world, unless you don't have enough transmitters, or perhaps too many.
Correct. I have no idea if M4 DAB reception is in the too few or too many transmitters scenario. And the problem may be if it has too many (eg. due to terrain making many visible), shutting some down to make the M4 work may cause reception to be lost elsewhere. Equally if the M4 needs more transmitters, adding more may result in interference breaking reception elsewhere. It's a matter of finding the least worst compromise, and it's a complex task. If the countryside were completely flat it would probably be easier, since distance and TX power alone would determine what transmitters could be received where and we could have a nice evenly spaced grid of transmitters where power and the curvature of the earth determined which one(s) could be received where.

This is also why some COM7 and COM8 transmitters had to be shut down when those Freeview muxes went to SFN, they were too close to other main transmitters.
 
Last edited:
OP
O

Owen Smith

Active Member
The alternative to SFNs for DAB would be a lot less muxes, probably only one national mux and less space for local muxes. That means the same number of transmitters ie. same cost to transmit, but with less income to pay for it whether that be from the BBC licence fee or less stations carrying adverts. We live in a world where things have to be paid for.

If I had my way we'd have never wasted any money on DAB, and instead spent it on improving and expanding the FM transmitter network. But that would have left us with only the BBC and Classic FM as national stations, plus a bunch of local stations and a much better range of stations in London (to be fair the "more choice in London" problem occurs on DAB as well). But that's partly with the benefit of hindsight, DAB has not been a success and bit rates have been crushed to near unlistenable levels. And we didn't know at the time that internet radio was coming soon enough to take the shine off DAB.
 

Black Hole

May contain traces of nut
I have no idea if M4 DAB reception is in the too few or too many transmitters scenario. And the problem may be if it has too many (eg. due to terrain making many visible), shutting some down to make the M4 work may cause reception to be lost elsewhere. Equally if the M4 needs more transmitters, adding more may result in interference breaking reception elsewhere.
That's as maybe, and I can accept that in a hill shadow there might be a problem particularly considering the frequency band in question (another reason it's a bad idea), but the M4 north of Bristol in a straight section on level ground with nothing obvious to block it from (I suppose) Mendip.

No amount of signal processing can make up for total destructive interference. With one receiving antenna there is no opportunity to discriminate on phase. If the frequencies really are identical, they look like multipath distortion and I accept clever processing can clean up the payload modulation... but that really does rely on destructive interference not reducing the signal below the noise threshold.

It seems to me there may be beating going on between Mendip and Wenvoe.

MIMO is different - it's a two-way system and able to be adaptive, using beam forming. Broadcast can't be adaptive.
 
D

Deleted member 473

My house, or equipment, can't manage mesh WiFi. I am back with an Asus router now, with separate 2.4GHz and 5GHz SSIDDs, no extended 5GHz channels, and with a separate Draytek VDSL modem. With mesh, I was constantly losing connections.

Even sticking the mesh nodes on separate frequencies didn't solve the problems.
 
OP
O

Owen Smith

Active Member
Black Hole. The point of designing an SFN is to avoid, by design, total destructive interference.

We don't actually know what the problem with DAB on the M4 is, it could be any number of things. But you are determined to portray it as being the thing you most dislike about SFNs.

The fact remains it is what it is, and in future I won't waste my time trying to explain things to people that clearly don't want to listen.
 

Black Hole

May contain traces of nut
The point of designing an SFN is to avoid, by design, total destructive interference.
That is clearly impossible, as any schoolchild who's seen a ripple tank can tell you.

Certainly there's no point wasting your time trying to explain something you don't appear to have the answers for. Everything you've said are excuses for SFNs rather than a demonstration that they work in the real world.

Cleaning up a signal with multipath distortion is a process of deconvolution, and the modulation characteristics can be designed to make the deconvolution process easier. It occurs to me successful deconvolution relies on the convolution being stable in the short term, whereas in a moving vehicle in a fringe area the convolution will actually be changing rapidly. This could explain why I experience blackout over a longer distance than a fringe ought to be (about 1m, because signal extraction will be disrupted by the rapid fluctuation of the signal rather than just during the destructive fringe), and the receiver might have significantly better chance if it were standing still.

I can test this. Next time I have the opportunity and it is convenient, I'll park up (not on the motorway!) when I lose DAB. If it remains lost when stationary, creeping half a metre should take me out of that particular fringe (assuming I'm not running along the fringe rather than across it). At last, a testable hypothesis, which has some credibility because DAB reception is known to vary from spot to spot within the house.

I won't waste my time trying to explain things to people that clearly don't want to listen.
I don't want to listen to dogma, no. I expect arguments to be backed up with data. If you feel like making a useful contribution, cite a reference where it shows how DAB was designed to cope with a moving receiver.
 
Last edited:
Top