HOLOCOMM: Secrecy by Delocalization

[Ed Gerck]

HOLOCOMM: Secrecy by Delocalization

Campinas, March 30th, 1998 -- Dr. Ed Gerck announced today a new
digital communication and encoding system that uses quantum
mechanical principles to provide for data privacy and reliability,
called Holocomm. As one of its main characteristics, information
encoded with Holocomm becomes fully delocalized and can be read only
with the proper decoding parameters. This affords a holographic
property: any part of the encoded information can be used to recover
the whole information to a degree.

The encoded information can be static, as in a certificate, e-mail
message, contract, etc, or dynamic, as in a real-time communication
channel.

Holocomm is neither cryptography nor steganography, even though both
share properties with it. It is not cryptography because the encoded
information cannot be localized. It is not steganography because it
does not depend on another information to hide the original
information, while it can use another information such as an image,
if so desired. It can provide strong security as with cryptography
while keeping the data hidden as with steganography. Moreover, since
the information is delocalized but not merged, no part of it can be
removed without removing the message itself while any part of it can
reproduce the message to a degree, which has no parallel in
cryptography or steganography.

"Holocomm fully delocalizes information, while strongly encoding it.
Delocalization means that all information syntax is free from the
limitations of locality, while all information semantics is
transformed into a global property. Thus, delocalization is not a
loss of identity but an omnipresence of it, for each part of the
message. A Holocomm message is a macro quantum mechanical wave
packet, where information is not stored in specific places, like
characters in a plaintext, bits in an encrypted message, bytes in a
data packet or as a time-dependent signal in a modulated-carrier wave
system, but is effectively distributed as an indivisible property of
the wave packet as a whole. Then, documents become truly holographic
in the sense that a rather small part of a document can be verified
to afford origin authentication of the entire document. Even data
integrity authentication for the whole document can be derived from a
small sample, to a degree. Holocomm documents can also be smaller in
size than the original documents." explains Dr. Gerck.

Regarding its strong encoding techniques, Holocomm can be a carrier
for any known encryption system such as RSA, DES, Idea, Blowfish,
RC4, etc. or for its own secure quantum mixing and encoding modes.
"Holocomm can interoperate with known, tried and secure encryption
techniques while also offering new avenues for strong data privacy
that can become vitally crucial, for example, if integer
factorization suddenly becomes viable for large key sizes. Further,
as world e-commerce needs are in opposition to export controlled
cryptography, Holocomm offers secure and open alternatives -- now."
declares Dr. Gerck.

Holocomm allows several problems to be solved, that have either no
solution or only partial and difficult solutions. For example:

- diverse encodings and media: Holocomm can produce messages with any
transport encoding, such as Base-64, ASCII, binary, etc. and for any
media such as e-mail, WWW, voice, film, etc.

- resistant transport: MIME or ASCII-armor may not be necessary in
the majority of cases for e-mail because text which is privacy
encoded and/or signed in a Holocomm system can be recovered to a
large degree even when mangled by mail transport systems. Localized
errors can be compensated because the transported information is
delocalized.

- digital signature recovery: today's digital signature systems need
an integral and errorless copy of both the document and the
signature. With Holocomm, even rather small parts of a document can
allow the document's origin authentication to be verified with
negligible error, also providing for data integrity authentication of
the whole document, to a degree. This recuperates a 3D-world
property legally known as a "holograph" (not to be confused with
hologram) and which was entirely lost for digital signatures.

- digital signature legislation: by allowing the usual legal concept
of a holograph to be applied for digital signatures, Holocomm
reduces the risk of document repudiation, tampering, etc. Since the
data and its signature become securely intermingled, the signature is
not localized and cannot even be separated from the document --
making the digitally signed document akin to a document wholly in the
handwriting of the author.

- strong transparency: A Holocomm message may be strongly encrypted
and yet such encryption may be undetectable within Complexity Theory
limits, when Holocomm works in privacy-transparent mode.

- resistant encryption: a Holocomm message that uses its own quantum
encoding modes can resist tampering attempts that may try to change
it or render it unreadable or unusable.

- Intrinsic Certification: Since quantum packets can be added without
loosing their identity, Holocomm allows independent secure
multichannel messages to be transported in the same certificate or
message, offering new tools to implement the Intrinsic Certification
method being developed by the MCG -- Meta-Certificate Group. Dr.
Gerck has granted the MCG worldwide rights to apply Holocomm in the
MCG developments and APIs.

- mandatory key-escrow: Holocomm can use Intrinsic Certification to
allow independency from CAs and TTPs, thus being legally independent
of any key-escrow or key-control legislation that may be imposed on
CAs and TTPs.

- privacy encumbering: Holocomm allows its privacy modes to be
undetectable within Complexity Theory limits, thus rendering useless
any attempt to encumber privacy. The same applies for certification
encumbering, by allowing undetectable Intrinsic Certification.

- rights management: Holocomm allows a non-invasive, indelible and
secure stamp to be applied to copyrighted materials such as software,
.gif or .jpg images, text, film, music, video, voice, etc.

- e-commerce: by providing for strong document non-repudiation,
secure signatures, privacy and independent secure multichannels,
Holocomm closely resembles traditional legal documents used in
age-old commerce, which can be hand annotated and signed by several
parties, all independently targetting any desired portion of one
document, without any need to expose the document to a notary (eg, a
CA or TTP) in order to achieve security.

- product and environmental security: Holocomm can securely store
complex chemical fingerprints of products (such as crude oil) and
allow undeniable product tracing and identification. Thus,
commercial and environmental aspects of different chemicals and
products can be securely certified to combat tampering, contraband,
unidentified spillovers and environmental damage. The same can be
applied to complex animal or plant biometric-data and other identity-
or capability-related data.

- cross-delocalized data: Holocomm allows for a set of data to be
delocalized in different places and not just in one place, i.e.
cross-delocalized data. This can be useful to implement secure
fail-safe procedures because not all data have to be accessible all
the time.

- export free: Holocomm was entirely developed to be export-free for
any degree of privacy and security. It can be freely exported and
re-exported to any country in the world. Holocomm does not depend on
any other patented system.

Holocomm can be applied now to worldwide security applications.
Companies, individuals and investors are invited to contact Dr. Ed
Gerck at ege...@novaware.cps.softex.br in order to discuss proposals:

"The opportunity is now open for any interested party to lead or
participate in commercial efforts that may deploy Holocomm in
applications that can profit from origin authentication, data
integrity authentication, privacy, security, non-repudiation
properties, encryption transparency, resistant encryption,
product security, etc."

The Holocomm system can and will be fully disclosed to the public, as
its security rests on freely chosen encoding/decoding parameters with
strong computing penalties to prevent parameter-space search, not on
method secrecy. The need to follow patent legislation procedures
clearly prevents a full disclosure at this moment but a technical
summary is provided below.

REFERENCES:

Dr. rer. nat. Ed Gerck is the CTO of Novaware
(http://novaware.cps.softex.br), Coordinator of the Internet open
group MCG (http://www.mcg.org.br) that has representatives from 25
countries and visiting Professor at UNICAMP, Brazil.

The Holocomm system is derived as a combination of free-standing
and/or bound-state wave packets in a large quantum mechanical system
that works as a thermal bath, which provides for the necessary state
mixing and strongly penalizes parameter-space search. Holocomm
depends on concepts published by the author more than 15 years ago
[1] [2], when the question of how to represent quantum states in a
fully analytical form was shown to be solvable with high accuracy
also as a function of a reduced representation of the quantum
mechanical equation itself and not only as a function of a reduced
representation of the wave functions for the full equation.

[1] E. Gerck et. al., "Solution of the Schroedinger equation for
bound states in closed form", Phys. Rev. A, vol. 26, p. 662, 1982.

[2] E. Gerck et. al., "Scaling laws for Rydberg atoms in magnetic
fields", Phys. Rev. Lett., vol.50, p.324, 1983.

This document is hereby released for public information, which does
not constitute a right to use such information for any other purpose
or for a product. Republication is allowed with copyright and author
citation. The Holocomm system, and the "Holocomm" name are Copyright
(c) Ed Gerck, 1998. All rights reserved worldwide. Patent
applications reserved worldwide.

-----------------------------------------------------------------

-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/ Now offering spam-free web-based newsreading

[John Savard]

Ed Gerck <ege...@novaware.cps.softex.br> wrote:

>Holocomm
>depends on concepts published by the author more than 15 years ago
>[1] [2], when the question of how to represent quantum states in a
>fully analytical form was shown to be solvable with high accuracy
>also as a function of a reduced representation of the quantum
>mechanical equation itself and not only as a function of a reduced
>representation of the wave functions for the full equation.

I thought this post had been cancelled, but I probably couldn't see it
again because I had read it - the newsreader on my other account is
tin.

Anyways, I hope this isn't an attempt to present the Fourier transform
as a revolutionary new invention. If not, it may be of interest.

John Savard

[Ed Gerck]

In article <35210b9...@news.prosurfr.com>,
jsa...@teneerf.edmonton.ab.ca (John Savard) wrote:

>
> Ed Gerck <ege...@novaware.cps.softex.br> wrote:
>
> >Holocomm
> >depends on concepts published by the author more than 15 years ago
> >[1] [2], when the question of how to represent quantum states in a
> >fully analytical form was shown to be solvable with high accuracy
> >also as a function of a reduced representation of the quantum
> >mechanical equation itself and not only as a function of a reduced
> >representation of the wave functions for the full equation.
>

>

> Anyways, I hope this isn't an attempt to present the Fourier transform
> as a revolutionary new invention. If not, it may be of interest.
>

The Fourier transform (eg, in cartesian coordinates) can be represented by its
eigenfunctions, eg the Gauss-Hermite functions, which indeed represent a
solution base used to approximate wave functions which obey a particular form
of the Schroedinger equation.

However, that is not the case here ...

The document becomes a snap-shot of a quantum system in evolution, after the
states have been sufficiently mixed. However, and this is important, state
mixing still maintains the identity of each state. So, information is
delocalized but does not loose its identity for each part -- which makes it
possible to decode it of course.

This is fundamentally different from any encryption or steganographic
system. This reflects a basic quantum property. If you take an electron,
you can't localized it -- the electron can effectively be anywhere.

Note: This means that, contrary to what might be understood, the
bit's probability function is NOT the probability to find the bit at
a point, but in a region. In fact the bit is quantum mechanically
delocalized and it has no sense to talk about finding the bit in one
position of the message.

Cheers,

Ed Gerck

[Ed Gerck]

> Ed Gerck <ege...@novaware.cps.softex.br wrote:
> >
> > a new digital communication and encoding system that uses quantum
> > mechanical principles to provide for data privacy and reliability,
> > called Holocomm. As one of its main characteristics, information
> > encoded with Holocomm becomes fully delocalized and can be read only
> > with the proper decoding parameters. This affords a holographic
> > property: any part of the encoded information can be used to recover
> > the whole information to a degree.

On the subject of Holocomm, I received several concrete questions as
well as several messages of "utter surprise" (to put it mildly)...

While the "utter surprise" messages are indeed a sign of no prior
art for Holocomm and further confirm its importance, let me now
address some of the most common questions, for the list's benefit.

1. DELOCALIZATION

Of course, it can be very surprising to learn that it is possible to
have in your hands a snapshot of a macro quantum mechanical wave
packet!

However, information "delocalization" is nothing new. For example,
SHA-1 can reflect in each 160 bits of its output a change made in one
bit in the output. And that has been around for a long time.

But, and herein lies the significance of Holocomm in this aspect,
because "delocalization" for SHA-1 is actually information merging
where the fine difference is that Holocomm's delocalization still
keeps the identity untouched for each part of the message while
SHA-1's merging mixes the different parts so that they fully loose
their identity.

Hence, a SHA-1 message cannot be decoded (and so should it be) while
a Holocom message can be fully decoded.

This also means that a Holocomm's message -- what you can call a
Holocomm-blob -- can be syntatictally divided without semantic
division, to a degree. Which may sound also strange but can be used
as a tool to also increase communication reliability, not only
secrecy.

2. PATENTING

Some also question whether we are trying to patent the Fourier
transform or something similar.

Indeed, Fourier transforms can be used inside Holocomm but Holocomm
is not based on Fourier transforms, and that can be easily proved
because Fourier transforms are always lossy for random finite
messages that are transformed into finite code-spaces. Even for FFT
of finite sequences.

However, we are neither patenting Fermat`s theorem (which has already being
patented) nor Fourier transforms or any other French technology :-)

Holocomm represents the first practical and viable application of
Quantum Mechanics to obtain increased signal secrecy and reliability,
covering the objectives of cryptography and steganography with one
method.

3. INTEROPERATION

Holocomm produces what is being called a Holocomm-blob. The input to
this process can be plain text or the output from any encryption
system, such as RSA, RC4, RPK, DES, Idea, Blowfish, etc. Of course,
such outputs will all equally result in a Holocomm-blob with
delocalized information-- which can be so transported and then
properly decoded (ie, relocalized) at the other end, and then
respectively decrypted.

Cheers,

Ed Gerck

[Walter Eric Johnson]

Ed Gerck (ege...@novaware.cps.softex.br) wrote:
:
:
: > Ed Gerck <ege...@novaware.cps.softex.br wrote:

: > >
: > > a new digital communication and encoding system that uses quantum

: > > mechanical principles to provide for data privacy and reliability,
: > > called Holocomm. As one of its main characteristics, information
: > > encoded with Holocomm becomes fully delocalized and can be read only
: > > with the proper decoding parameters. This affords a holographic
: > > property: any part of the encoded information can be used to recover
: > > the whole information to a degree.

:
: On the subject of Holocomm, I received several concrete questions as

: well as several messages of "utter surprise" (to put it mildly)...
:
: While the "utter surprise" messages are indeed a sign of no prior
: art for Holocomm and further confirm its importance, let me now
: address some of the most common questions, for the list's benefit.

There have been holographic-type codes before. They weren't designed
to hide information, but to make sure the message gets through by
enabling the receiver to rebuild the message despite numbers of
random errors. I suspect this was used for transmitting pictures
from space probes to earth since they would not be able to save the
image in case it needed to retransmit it. Better to send it once,
albeit a longer version, and then dump it from memory.

Eric Johnson

[John Ioannidis]

In article <6fpa74$ao6$1...@nnrp1.dejanews.com>,
Ed Gerck <ege...@novaware.cps.softex.br> wrote:
>
>
> HOLOCOMM: Secrecy by Delocalization
>
>

Great April 1st posting... but isn't it a day early?

/ji

[Ed Gerck]

In article <6fuj2k$50p$1...@news.tamu.edu>,

wej...@fox.tamu.edu (Walter Eric Johnson) wrote:
>
> Ed Gerck (ege...@novaware.cps.softex.br) wrote:
> :
> :
> : > Ed Gerck <ege...@novaware.cps.softex.br wrote:
> : > >

> : > > a new digital communication and encoding system that uses quantum

> : > > mechanical principles to provide for data privacy and reliability,
> : > > called Holocomm. As one of its main characteristics, information
> : > > encoded with Holocomm becomes fully delocalized and can be read only
> : > > with the proper decoding parameters. This affords a holographic
> : > > property: any part of the encoded information can be used to recover
> : > > the whole information to a degree.

> :

> : On the subject of Holocomm, I received several concrete questions as
> : well as several messages of "utter surprise" (to put it mildly)...
> :
> : While the "utter surprise" messages are indeed a sign of no prior
> : art for Holocomm and further confirm its importance, let me now
> : address some of the most common questions, for the list's benefit.
>
> There have been holographic-type codes before.

Exactly and good point. Even though the analogy is not complete. The
holographic codes you mention do not represent the major part of the present
case because those codes produced messages which were lossy and large, did not
allow compression and did not target delocalization of information. So,
Holocomm includes those (and other) points for which there are no prior art.
For example, Holocomm can selectively allow some information to be
well-localized while fully delocalizing other parts, all in the same
Holocomm-blob and with zero-loss. The degree of delocalization is also a
variable, as well as the intensity (ie, redundancy) of it.

The best analogy here is not so much holography but the quantum paradox of
Schroedinger`s cat.

Of course, holography can be (and has been) seen as a quantum process (being
basically a wave phenomena) and indeed holographic codes can be seen as a
particular case of a lossy Holocomm message that does not favor
selective delocalization. This is fine and very reassuring because it shows
that Holocomm reproduces known results for a particular set of parameters.

The use of Holo in Holocomm is exactly there to express the holographic
property, which word and meaning dates back to 1623 -- more than 120 years
before Gabor`s theory. And the posting made a point of stressing that, so that
people could better relate to it.

> They weren't designed
> to hide information, but to make sure the message gets through by
> enabling the receiver to rebuild the message despite numbers of
> random errors. I suspect this was used for transmitting pictures
> from space probes to earth since they would not be able to save the
> image in case it needed to retransmit it. Better to send it once,
> albeit a longer version, and then dump it from memory.
>

Exactly, they aimed more at reliability -- even though the codes were lossy
and reliability was achieved mainly by coherent en/decoding, because noise is
incoherent. However, much of that was dropped in favor of faster and better
(in that approach) use of homomorhic processing and other DSP techniques.

Further, Holocomm's "delocalization" feature can be seen also in SHA-1, where
*all* output bits change when one changes a *single* input bit. However, SHA-1
hopelessly mixes and merges all the data (as it is intended to do), while
Holocomm allows for reversible and selective delocalization.

Thus, in two contrast points to former pure holographic codes, Holocomm aims
at (1) non-lossy reversible (2) selective delocalization -- which also allows
interoperation with all known cryptography algorithms (that require exact data
for decoding). The reliability feature is also further enhanced by the
non-lossy aspect of it. As mentioned, Holocomm can also work in lossy modes,
including lossy compression -- which can be quite useful.

Holocomm is the first example of a practical quantum mechanical communication
and encoding system that affords privacy and reliability, to a high degree,
while also offering compression and selective information delocalization.

As such, it naturally has many parallels in several things that are based on
wave functions or on the Schroedinger equation .. which essentially defines
wave phenomena ... as the theoretical basis of Holocomm, as stated.

Thank you,

Ed Gerck

[Mike.Andrews]

In article <6fuj2k$50p$1...@news.tamu.edu>,
wej...@fox.tamu.edu (Walter Eric Johnson) writes:

>There have been holographic-type codes before. They weren't designed

>to hide information, but to make sure the message gets through by
>enabling the receiver to rebuild the message despite numbers of
>random errors. I suspect this was used for transmitting pictures
>from space probes to earth since they would not be able to save the
>image in case it needed to retransmit it. Better to send it once,
>albeit a longer version, and then dump it from memory.

If I recall correctly, at least some NASA probes (e.g., Voyager)
used a pseudonoise sequence sent straight-up for a "1" bit and
inverted for a "0" bit. Inefficient, perhaps, but it certainly
makes recovery easy at the receiver, even with a really abysmally
bad SNR.

--
Mike Andrews
Mike.A...@fd9ns01.okladot.state.ok.us

[John Savard]

wej...@fox.tamu.edu (Walter Eric Johnson) wrote:

>There have been holographic-type codes before. They weren't designed
>to hide information, but to make sure the message gets through by
>enabling the receiver to rebuild the message despite numbers of
>random errors. I suspect this was used for transmitting pictures
>from space probes to earth since they would not be able to save the
>image in case it needed to retransmit it. Better to send it once,
>albeit a longer version, and then dump it from memory.

My understanding is that the error-correcting codes used from space
probes were not holographic, but merely doubled the length of blocks
from 16 to 32 bits or from 32 to 64 bits using conventional Hadamard
error-correcting codes.

John Savard

[Peter Pearson]

Now that April Fool's Day has ended in all timezones,
this thread will dry up ... right?

- Peter

[Jayant Shukla]

Ed Gerck <ege...@novaware.cps.softex.br> writes:

>Further, Holocomm's "delocalization" feature can be seen also in SHA-1, where
>*all* output bits change when one changes a *single* input bit. However, SHA-1

^^^^^^^ ^^^^^^^^^

>hopelessly mixes and merges all the data (as it is intended to do), while
>Holocomm allows for reversible and selective delocalization.

Are you sure?? If "all" output bits change then what you have is just
the complement of your previous output. This means that your method only
gives two possible outputs and one is a complement of the other (because
any n-bit number can be transformed to another n-bit number by changing at
most n-bits ;^) ).

regards,
Jayant

[Ed Gerck]

In article <Eqpo5...@research.att.com>,

;-)

Thank you for your declaration of no prior art for Holocomm. Is that
also the opinion of your employer or just your personal technical
opinion? I may need to cite it in the future, hence the question of
its exact context...

Indeed, it can be very surprising to learn that it is possible to
have in your hands a snapshot of a macro quantum mechanical wave
packet!

But, it will not delocalize you... have no fear!

Thanks,

Ed Gerck

[Ed Gerck]

In article <6g3c8n$fkn$1...@sunnews.cern.ch>,

jsh...@rsplus03.cern.ch (Jayant Shukla) wrote:
>
> Ed Gerck <ege...@novaware.cps.softex.br> writes:
>
> >Further, Holocomm's "delocalization" feature can be seen also in SHA-1,
> >where *all* output bits change when one changes a *single* input bit.

> > ^^^^^ ^^^^^^^^
> >However, SHA-1 hopelessly mixes and merges all the data (as it is intended

> >to do), while Holocomm allows for reversible and selective delocalization.
>
> Are you sure?? If "all" output bits change then what you have is just
> the complement of your previous output. This means that your method only
> gives two possible outputs and one is a complement of the other (because
> any n-bit number can be transformed to another n-bit number by changing at
> most n-bits ;^) ).

You misread twice.

First, I was NOT mentioning Holocomm`s properties, I was comparing it with
SHA-1 in order to allow some understanding of delocalization by way of a
metaphor. I was mentioning SHA-1`s properties -- as can be understood by the
sentence "also in SHA-1, where all output bits change..." and the word
"where".

Second, bits can change by shifting, rotating, etc. -- which are much more
useful than just changing by reversal. Please note that I wrote "bits".

What I stated, without making much fuss about it is just what everyone knows
about SHA-1. And, therefore, not such a surprise and possibly a useful
metaphor for delocalization. However, the surprise is the reversibility of
delocalization. The novelty also lies on the use of delocalization, an
essentially quantum property, to achieve security and reliability in messages
that can be reversibly encoded and decoded by delocalization.

Cheers,

Ed Gerck

[Walter Eric Johnson]

Ed Gerck (ege...@novaware.cps.softex.br) wrote:
: holographic codes you mention do not represent the major part of the present

: case because those codes produced messages which were lossy and large

I can see that they were large, but do not at see that they must necessarily
be lossy. Can you explain just why they were lossy?

Eric Johnson

[Ed Gerck]

In article <6g94bu$gru$1...@news.tamu.edu>,

As you commented, those codes were not designed to hide information but to
decrease transmission errors. They were not designed to eliminate losses
completely, either. In fact, small losses had to be tolerated in order to
reserve more transmission power to the most significant signal window, with
adequate signal apodization. If apodization had not been used, then
transmission losses would decrease but reception losses would increase.

However, if you want to verify a cryptographic signature, one bit of
difference will make the test fail and transmission losses cannot be
tolerated.

Cheers,

Ed Gerck

[Ed Gerck]

In article <ppearsonE...@netcom.com>,

ppea...@netcom.com (Peter Pearson) wrote:
>
> Now that April Fool's Day has ended in all timezones,
> this thread will dry up ... right?
>

;-)

Thank you also for your declaration of no prior art for Holocomm.

Ed Gerck

[Walter Eric Johnson]

Ed Gerck (ege...@novaware.cps.softex.br) wrote:
: In article <6g94bu$gru$1...@news.tamu.edu>,

: wej...@scully.tamu.edu (Walter Eric Johnson) wrote:
: >
: > Ed Gerck (ege...@novaware.cps.softex.br) wrote:
: > : holographic codes you mention do not represent the major part of the
: present
: > : case because those codes produced messages which were lossy and large
: >
: > I can see that they were large, but do not at see that they must necessarily
: > be lossy. Can you explain just why they were lossy?
: >
:
: As you commented, those codes were not designed to hide information but to
: decrease transmission errors. They were not designed to eliminate losses
: completely, either. In fact, small losses had to be tolerated in order to
: reserve more transmission power to the most significant signal window, with
: adequate signal apodization. If apodization had not been used, then
: transmission losses would decrease but reception losses would increase.

But that's not what "lossy" means. Every definition, description,
analysis, ... using the work "lossy" used a definition of something
like "any information encoded does not decode to the exact same
message." If there are no transmission errors, the methods we were
discussing would provide exactly the same message as was encoded.

Does anyone else use the term "lossy" with your definition?

Eric Johnson

[Ed Gerck]

In article <6gf08j$oh8$2...@news.tamu.edu>,

wej...@fox.tamu.edu (Walter Eric Johnson) wrote:
>
> Ed Gerck (ege...@novaware.cps.softex.br) wrote:
> : In article <6g94bu$gru$1...@news.tamu.edu>,
> : wej...@scully.tamu.edu (Walter Eric Johnson) wrote:
> : >
> : > Ed Gerck (ege...@novaware.cps.softex.br) wrote:
> : > : holographic codes you mention do not represent the major part of the
> : present
> : > : case because those codes produced messages which were lossy and large
> : >
> : > I can see that they were large, but do not at see that they must
necessarily
> : > be lossy. Can you explain just why they were lossy?
> : >
> :
> : As you commented, those codes were not designed to hide information but to
> : decrease transmission errors. They were not designed to eliminate losses
> : completely, either. In fact, small losses had to be tolerated in order to
> : reserve more transmission power to the most significant signal window,
with
> : adequate signal apodization. If apodization had not been used, then
> : transmission losses would decrease but reception losses would increase.
>
> But that's not what "lossy" means. Every definition, description,
> analysis, ... using the work "lossy" used a definition of something
> like "any information encoded does not decode to the exact same
> message."

;-) I assume I was not clear enough, in my reply above -- as I meant exactly
what you said. Let me be less concise, since redundancy can help communication
efficiency.

When I mentioned signal apodization at the transmiting end, I said that signal
apodization implied losses and that such losses were minimized as a whole
--ie, not only at the transmiting end but also taking into account what could
be received (ie, as a function of available resources and conditions).

However, what is signal apodization? Simply a convenient tappering off of the
signal (eg, with gaussian borders), or in other words, an intentional signal
loss. First I note that there is a sharp distinction between the concepts of
apodization and cut-off: apodization is not sharp ;-) The reason is based on
the Fourier transform. The objective is to avoid ripples which would be caused
by sharp transitions in the work domain (ie, frequency, time, position, etc.)
when calculated in the transformed domain -- since the Fourier transform of a
gaussian is a gaussian, both smooth, this objective can be obtained by a
gaussian-like taper.

Apodization is usually done (as it was) to optimize the use of transmission
power and bandwith for the most significant part of the signal. This can be
interpreted (and used) in the time, frequency or space domain representation
of the signal. For example, analog phone lines rely on frequency apodization
in order to better use the available "copper-line" spectrum for as many voice
channels as possible, while allowing the most significant part of the voice
spectrum to be transmitted.

Back to Mars ;-) Apodization then implies "engineered loss" in the sense that
losses are *inserted* into the system in order to improve the system's
performance, by convenient window and tappering selection. Thus, apodization
in holographic signal encoding is necessary to *improve* overall performance
and implies losses -- so that "any information encoded does not decode to the
exact same message."

A related question is: what would happen if apodization were not used? Could
it be lossless? The answer is that losses would be **larger** than with
apodization! That's why I wrote: "If apodization had not been used, then

transmission losses would decrease but reception losses would increase."

A further related question is: but, if I had more transmission power or the
signal would have a digital encoding such as TCP/IP packets? The answer is
that you can never exactly represent a Fourier transform of an arbitrary
signal in a finite domain. Thus, Fourier transform pairs are always lossy in
finite domains -- meaning that "any information encoded does not decode to the
exact same message." They can be fairly close approximations though, as
Gaussian laser beams prove, when used with apertures that have large Fresnel
numbers.

> If there are no transmission errors, the methods we were
> discussing would provide exactly the same message as was encoded.

As above, no. Further, Shannon's Tenth Theorem -- in a more general framework
-- says that it is always possible to reduce noise to an arbitrarly small
error ....but does not say to zero....

BTW, the assumption of zero transmission errors is funny... since you do not
know what was transmitted, by hypothesis.

>
> Does anyone else use the term "lossy" with your definition?

Everyone, including you ;-)

Cheers,

Ed Gerck

PS: I am always amused by the "holier than thou" attitude that can be seen in
some Internet postings, when people assume that a certain technology branch is
the exclusive domain of a group or country. It is perhaps one of the great
merits of the Internet to level off such understandings and reduce such
barriers.

[Walter Eric Johnson]

Ed Gerck (ege...@novaware.cps.softex.br) wrote:
: In article <6gf08j$oh8$2...@news.tamu.edu>,

So we are talking about different things entirely. I'm talking about the
message encoded, broadcast, received, and decoded in such a way as to
minimize the chance of losing the original message. You seem to be
exact shape of the broadcast signal. The point is that the codes were
supposedly used so that major parts of the broadcast signal could be
noisy and yet all or nearly all of the message could be recovered.

Eric Johnson

[Ed Gerck]

In article <6glolg$mvd$1...@news.tamu.edu>,

wej...@scully.tamu.edu (Walter Eric Johnson) wrote:
>
> Ed Gerck (ege...@novaware.cps.softex.br) wrote:

> : In article <6gf08j$oh8$2...@news.tamu.edu>,

> So we are talking about different things entirely. I'm talking about the
> message encoded, broadcast, received, and decoded in such a way as to
> minimize the chance of losing the original message. You seem to be
> exact shape of the broadcast signal. The point is that the codes were
> supposedly used so that major parts of the broadcast signal could be
> noisy and yet all or nearly all of the message could be recovered.

I would say we are in non-lossy (ie, 100%) agreement if you use "nearly all"
above.

Back to Holocomm, one interesting feature of using delocalization for secrecy
is that you do not introduce any *new* information into the stream -- which
seems to provide an a priori protection against subliminal channels and
weak-padding attacks. Block ciphers and stream ciphers do not do that, of
course.

Other interesting properties of Holocomm seem to derive from the fact that the
stream is "wholly signed" by the delocalization feature. For example, you may
not need intermediate MACs as in SSL, to make sure that the stream was not
subreptiously changed in some bit values.

Cheers,

Ed Gerck
___________________________________________________________________________
Dr.rer.nat. E. Gerck ege...@novaware.cps.softex.br
http://novaware.cps.softex.br
--- Meta-Certificate Group member, http://www.mcg.org.br ---