RadioBanter - View Single Post

July 5th 05, 12:19 AM

wrote:
John Smith wrote:

N2EY:

Most of that is incorrect.

Most of what?

First you use "on the fly" encryption/decryption/"data
compaction" and
have it occurring in "real time." This has the effect of being
"transparent" and the user is not even aware that it is going
on.

That's what

"Convert the pictures and video into highly-compressed digital
formats for transmission." means, John. Whether it's done in "real
time" is just a detail.

An important detail indeed if the image is going to be transmitted in
real time. Presumably some time might be alloted if the image were to be
delayed by a second or so.

Next, forget the sn/noise ratio other than it has to acceptable for
transmission of understandable communication (however, this is
required no matter what the form of data--i.e., voice, ssb, cw, etc)

Signal-to-noise is an integral part of Shannon's thereom. It cannot
simply be "forgotten".

Forgetting signal to noise is what makes me believe that these claims
are more snake oil than substance.

Next, listen to digital signal occupying audio bandwidth (it is audio
bandwidth that is of concern here, NOT rf bandwidth,

No, that's not correct.

You are correct, Jim. That RF bandwidth must be capable of carrying
whatever "audio bandwidth" there is. Listening to the audible portion of
a digital signal is an interesting metric though! ;^)

The discussion is about transmitting pictures and video on the amateur
HF/MF bands. RF bandwidth is a very important thing there.

Very very important.

except
with the
possibility of fm and how you implement the data compression
and
transmission, i.e., just make it fit the existing rf bandwidth and NO
changes are needed--however, larger rf bandwidth will ALWAYS
result in
a drastic increase in transmission speed and wideband fm can
easily
offer itself to 1MBS and faster) a digital signal can be
treated just
like a analog signal if desired, the use of CRC checksums and
error
checking of the data is just more intense under these
circumstances
and there is NO standard established for this--so you MUST be
able to
make and use your own custom hardware and software. To avoid
this, just grab off the shelf digital hardware/software.

And the simplest way for hams to do that at HF/MF is to use an SSB
transceiver and a computer with a sound card.

Waaayy overly simplified. If it were just a matter of compressing the
signal until it fit into whatever bandwidth was desired/needed/mandated,
don't you think we would have gone that route, instead of inventing
faster modems, T-lines, Cable modems and DSL? There are limits which we
passed a long time ago, after which data MUST be thrown away. There are
finite limits that imagery or video cannot be compressed without
sacrifices in fidelity.

But that's not the only issue.

Hardly!

Next, for every patented form of audio video protocols there
are FREE
forms, usually the free ones are more acceptable, efficient and
suitable to ones needs, an example:
Use ogg vobis compression of audio as opposed to mp3
--in video--
Use xvid as opposed to divx 4-5

And make sure the folks at the other end are similarly equipped.

Oh, yeah. A lot of people forget about that one. We need someone to
communicate with.

However, any of this requires a sound and current education and
knowledge of the state of technology--and something which is
obviously lacking here.

Yes, John, your lack of a sound and current education about amateur
HF/MF communications is quite evident. Good to see
you admitting it.

But that can be fixed.

There's also the issue of FCC regulations. Of course those regulations
can be changed, and there are several proposals in development or
before the FCC to change them. But until they are changed, amateurs
will be constrained by the current rules, such as the 300 baud
limitation on HF. The vast majority of hams are not going to break
those rules, regardless of the available technology or their education.

The question raised by KB3EIA and N8UZE remains: How can video
be sent in a 2.5 kHz RF bandwidth on the amateur HF bands? I've
answered that question in a theoretical way. I don't think you
even understand the question and all its implications, John.

Some things for people to think about:

There is nothing preventing the use of digital imagery or video at HF
frequencies. NO majik involved.

But!

The bandwidth needed for the data being sent may very well approach or
exceed the frequency being used for the transmission!

This means that the data needs compressed or the time needs expanded.

Compression has well defined limits. Beyond these limits, data must be
thrown away. This is why original images (or video) should be performed
in as high a resolution and as low a compression as possible. It is
always possible to throw away data, but not get back data that has been
discarded.

So beyond whatever minimal image quality is agreed upon, time must be
expanded. 300 baud is exceptionally slow by today's standards. I worked
up some times on transmission of a level 5 640 by 480 jpeg earlier in
this thread. I don't consider those times acceptable.

So here we are.

- Mike KB3EIA -