Re: [IRCA] NAB wants it both ways

[Barry McLarnon <bdm@xxxxxxxxxx>]

On Saturday 13 October 2007 22:50, Craig Healy wrote:
> Why not on AM?  There are many times when the LSB or USB provides far
> better reception.  Why wouldn't that apply here?  It's not selective
> fading, but interference.  One sideband is almost always better except for
> very close locals.  Why is the time diversity important on hybrid?  I
> thought the delay in the analog channel was just to match the delay needed
> to process and create the digital sidebands.  I was unaware there was any
> other reason.  Do they use some sort of Forward Error Correction that needs
> time to work? Every other signal I've seen with serious FEC doesn't need 8
> seconds.

You're right - frequency diversity does help in the hybrid system when the 
interference levels on the adjacent channels are different.  But in the 
all-digital system, this doesn't apply anymore.  The primary digital 
sidebands are moved into the central 10 kHz of the channel, so the core audio 
stream is now subject only to co-channel interference.  The secondary digital 
sideband goes to one 1st adjacent channel, and the tertiary digital sideband 
goes to the other one, so you would need to decode them both in order to get 
the stereo enhancement stream.

Re the delay between the analog and digital streams: part of this is due to 
the digital processing delay, and part of it (about 4.5 sec) is added for 
time diversity.  The idea behind this is if there is a transient event (such 
as a static crash or dropout from driving under a power line) that kills the 
digital decoding, the system can blend back to analog and stitch in some 
audio that is unaffected by the event.  It doesn't always work that well in 
practice, apparently.  IIRC, there is also a diversity delay between the two 
primary digital sidebands, and this is retained in the all-digital system, so 
it also has some time diversity.

For more of the gory details, grab this paper from the iBiquity website:
http://www.ibiquity.com/i/pdfs/Waveforms_AM.pdf

> I think in some cases it would be worse.  For example, if WNBH-1340 went
> all digital, it's entire signal would be within the 1340 channel.  That's
> shared by a bunch of other stations around here.  If it were in hybrid
> mode, the digital information would be on 1330 and 1350, both far more
> clear than 1340.

True, the Class C graveyarders could be an exception.  Since they have 
essentially no co-channel protection, they have the most to gain by grabbing 
the neighboring channels.  They're still gonna have lousy digital coverage at 
night, though it might not be any worse than their analog coverage.  But if 
they ran all-digital, they could run more power (i.e., somewhere between the 
current analog power and the hybrid digital power).

> That same concentration of signal also makes for an increase in
> interference.  We all know the devastating effects of the 100% digital
> hiss. It goes much further than analog.  Look at the difference in the US
> local frequencies when they allowed them to stay at 1,000w nights.  Overall
> it made little difference.  All it did was raise the signal above some
> ambient noise.  It really didn't improve reception.

If "digital in-fill" is allowed, the allocation/protection rules must be 
adapted accordingly, just as they've done in Europe to accommodate DRM.
For example, a Class C station that went all-digital might be required to run 
only 250 W digital power.  According to the ITU studies, if the digital power 
is set at 6-7 dB below what would be permitted under the protection rules for 
analog signals, it will provide adequate protection to co-channel analog 
stations (and more than enough for co-channel digital stations, if any).  
This was determined by comparing SNR measurements in receivers when subjected 
to analog and digital interference.  Of course, DXers would see it 
differently, since the 100% duty cycle of the digital signal does not provide 
any quiet gaps for hearing weaker signals underneath.

> I don't think the FCC will do the right thing.  It will be up to the
> courts, or perhaps the Congress.  It may even be the State Department if
> some of your countrymen get your government to act on the interference. 
> There are a number of stations in Canada suffering because of this. 
> Doesn't the government act on such things?

Industry Canada said they will act, if they get interference complaints from 
stations.  The grapevine seems to indicate that some have been filed, but I 
don't know for sure.  Unfortunately, this is not really a public process.

> If someone were to write the firmware, yes.  It would still require extra
> code and memory to include the selection algoritm, plus the decoder.  It
> probably is somewhat different than the hybrid version.  Extra cost, and
> hardly worth it for something that's at least ten years off.  I simply
> don't know what's under the hood.  I did ask one of the local engineers and
> he said they couldn't do all-digital mode.  So, I have no real way of
> testing my receiver.  If I run into the other IBOC station engineer, I'll
> ask him.

Some all-digital test results were filed with the FCC, so the code to 
implement it definitely exists.  Whether it was ported to the TI chipset is 
another question, but I'd suspect that the answer is yes.  Whether the 
receiver manufacturers were actually required to make the mode functional is 
yet another question.

> It may be on-channel, but the noise goes up as well.  European DXers aren't
> at all happy with their version of the Digital Disease.  Hiss-teria!

True, but as the IBOC proponents are fond of telling us, the broadcast 
services don't exist to serve DXers.

> No matter what they do, it will not cure the problem of poor building
> penetration, skywave interference, or higher ambient noise levels.  And
> this doesn't even address the horribly dictatorial business model IBOC
> users are forced to buy.  All in all, this is a really poor way to go
> digital.  They appear to have taken the worst choice at every turn.  How
> could it even be worse than this?

I can't argue with that!  IMHO, the real target for IBOC has always been the 
FM band, but they had to come up with this lame strawman AM system in order 
to appease the NAB and keep things going.

I'm not really arguing in favor of running digital in the AM band at all, as I 
think the legacy AM system will still be viable (especially in rural areas) 
for many years to come.  But there are some undeniable advantages to digital 
transmission, and I think could be some reasonable ways to accommodate it in 
the AM band, if common sense and good engineering practices are applied.  The 
hybrid AM IBOC system has neither of these attributes.

Barry

-- 
Barry McLarnon  VE3JF  Ottawa, ON
_______________________________________________
IRCA mailing list
IRCA@xxxxxxxxxxxxxxxx
http://montreal.kotalampi.com/mailman/listinfo/irca

Opinions expressed in messages on this mailing list are those of the original contributors and do not necessarily reflect the opinion of the IRCA, its editors, publishing staff, or officers

For more information: http://www.ircaonline.org

To Post a message: irca@xxxxxxxxxxxxxxxx