Re: IBOC "HD" radio in USA
- From: Robert Orban <donotreply@xxxxxxxxxxxxx>
- Date: Wed, 07 Mar 2007 17:40:31 -0800
In article <1173263892.328825.286910@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>,
On Mar 7, 9:55 am, "Robert A. Staton" <rsta...@xxxxxxxxxxxxxxxxx>
"Boltar" <boltar2...@xxxxxxxxxxx> wrote in message
On Mar 6, 1:26 am, Steven Stone <xxspfl...@xxxxxxxxxxxxx> wrote:
|Thats interesting. Since digital takes a short while to encode you'd
|always expect at least a small delay
The Sangean HDT-1 runs both analog and IBOC thru the TI DSP chip.
This produces no delay switching between analog and digital as long as
the radio station has their end in synch, however you will notice a
delay playing an traditional analog radio alongside the Sangean HDT-1
tuned to the same station.
How does it know how to synchronise the signals though? Unless it does
some serious processing on quite a large section of the broadcast (so
it doesn't get caught out by constant notes/tones that are the same
waveform for a long duration) which is probably asking a bit much from
a consumer radio, I'm not sure how it could do it unless they've
embedded some sort of timing signal into both the digital signal
(easy) and the FM signal (in the RDS data?).
I don't know the exact answer to this, but it's not too difficult to make a
good guess. The encoding process is going to take x amount of time, giving
a delay of x. The receiver knows that the encoded stream took x amount of
time for the encoder to do its thing, so it delays the analog audio by
exactly the same amount of time.
That would work but only if the broadcasters exactly sychronise the
analogue and digital signals (which means they'll have to delay the
analogue for enough time to allow for the digital encoding process) so
the radio can assume that any delay is just down to its own decoding.
This is exactly how the system works.
iBiquity has tightly specificed the relative delay between the receivers'
analog FM and IBOC decoders. Broadcasters set an exact time offset for the
analog FM and IBOC transmissions to make them line up precisely in time at the
receiver. The typical delay required is about 8 seconds (the analog FM is
delay with respect to the IBOC). This delay is typically trimmed to be
accurate within a few tens of microseconds. For example, in our Optimod-FM
8500 audio processor, the granularity is 15.625 microseconds, corresponding to
one sample at 64 kHz sample rate. The processor is designed so that this delay
remains sample-accurate, regardless of how the processor is adjusted.
Broadcasters often use a receiver for this adjustment, set so that the analog
FM and IBOC decoded signals are summed. They then adjust the transmission
delay to minimize audible flanging in the summed signal. Fortunately, there
are now several instruments available that can measure the delay (typically by
cross-correlation techniques) and read it out. Needless to say, this process
is not nearly as difficult as the "by-ear" method.
- Re: IBOC "HD" radio in USA
- From: Steven Stone
- Re: IBOC "HD" radio in USA
- Prev by Date: Re: DIGITAL RADIO TECH
- Next by Date: Re: DRM coverage & schedules
- Previous by thread: Re: IBOC "HD" radio in USA
- Next by thread: Re: IBOC "HD" radio in USA