Daxon Scrambler Scam?

Hello DSP wizards,

I am still not convinced of Daxon's "FFT Scrambler" shown here:
http://www.daxon.ru/eng/daxon500.html

They say that they implement a band splitter scrambler using an FFT/inv
FFT without any need of synchronization whatsoever. Is that really
possible, even though they claim the audio quality is "excellent"?

Im just wondering if theres something fishy about this.
Thanks :),
--Shafik

Hi.  You asked this before and so I'll repeat my answer:
The web page you listed does NOT say "without any need of synchronization
whatsoever".  It says "Doesn't require the synchronization burst".  (Not being
familiar with this type of device, I don't know all the details, but it would
seem that this is a very important distinction.)

If you are really so concerned about this product, why not buy one and try it
yourself?  Or try to find others who have used it?

"Shafik" <shafik@u.arizona.edu> wrote in message
news:1102709269.597886.246750@c13g2000cwb.googlegroups.com...
> Hello DSP wizards,
>
> I am still not convinced of Daxon's "FFT Scrambler" shown here:
> http://www.daxon.ru/eng/daxon500.html
>
> They say that they implement a band splitter scrambler using an FFT/inv
> FFT without any need of synchronization whatsoever. Is that really
> possible, even though they claim the audio quality is "excellent"?
>
> Im just wondering if theres something fishy about this.
> Thanks :),
> --Shafik
>

I have actually called them and asked, and theres no need for synch at
all.

I wanna make sure it's not a hoax before I buy one to test it out.
--Shafik

On 10 Dec 2004 17:14:09 -0800
"Shafik" <shafik@u.arizona.edu> wrote:

> I have actually called them and asked, and theres no need for synch at
> all.
> 
> I wanna make sure it's not a hoax before I buy one to test it out.
> --Shafik
> 


It sounds like an implementation of:
"A speech security system not requiring synchronisation" by Lin-Shan Lee.
IEEE Comms mag, Vol23 No 7. It uses circulant matrices to allow a frequency
domain technique to be applied in the time domain and removes the need for
synchronisation.

I have an implementation of this algorithm. It does work. I would, however, only
rate it as voice _privacy_, not secure. Although the number of keys can be 
arbitrarily high, there is only a subset of these keys that allow real security.

The algorithm does have some nice features, however....

--ted

Is there a place where I can access that paper for free?
--Shafik

Don't know. Try your nearest university library, they often
have journals such as these (and photocopiers) - but beware of
copyright laws :).
 
There are about 3 or 4 papers altogether.

--ted


On 14 Dec 2004 17:34:42 -0800
"Shafik" <shafik@u.arizona.edu> wrote:

> Is there a place where I can access that paper for free?
> --Shafik
> 


-- 
ted bullen MEng                     Inspiration Technology Pty Ltd
1/341 King William St, Adelaide SA 5000         www.instech.com.au
phone: +61 8 8211 9668  mobile: 0418 843 953  fax: +61 8 8211 9658