Searching \ for '[PIC]: 32bit block cipher' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/devices.htm?key=pic
Search entire site for: '32bit block cipher'.

Exact match. Not showing close matches.
PICList Thread
'[PIC]: 32bit block cipher'
2001\01\08@044752 by Germain Morbe

flavicon
face
Hi,
does anybody know a short 32bit block cipher routine for the pic. The keylen
need not be more than 64bit. I already saw TEA but its blocklen is 64 bit
and key is 128. Therefore i think one could save codespace and even ram. My
requirement is not extremely high sercurity but rather good mixing from one
input bit to more (best 50%) output bits.

Germain Morbe

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\08@052117 by Roman Black

flavicon
face
Germain Morbe wrote:
>
> Hi,
> does anybody know a short 32bit block cipher routine for the pic. The keylen
> need not be more than 64bit. I already saw TEA but its blocklen is 64 bit
> and key is 128. Therefore i think one could save codespace and even ram. My
> requirement is not extremely high sercurity but rather good mixing from one
> input bit to more (best 50%) output bits.

Warning!
It is a federal offence for USA firms or citizens to
release cipher technology/information to other countries.
It is considered weapons technology or something and
it is one thing the Feds DO keep an eye on.
I think it might even be illegal for a US citizen to
discuss methods with non-US citizen.

However I'm Australian and not limited by US federal
laws in this regard. :o) I have not done this on a PIC
but have done it in an application on the PC.

Most encryption systems use a transpose/substitute
system, where the key is used to determine the
position the data is transposed to in the block,
could be a simple rotate by x bits for example.
With x determined by data in the key.
Then you follow with a substitution where the
same key, or a different part of it, is used to
substitute one data unit for another. If each data
unit is a byte (like an alpha character), that byte
would be incremented by a value determined by the
key. Or each nibble, 2 bits etc is substituted.

When this transpose/substitute cycle is carried out
for only a few cycles the encryption is pretty secure.
With a fast PIC you can do it many times, even have
the total number of cycles determined by the key also.
This becomes very difficult to solve, and is one
of the reasons the Feds don't like it!
It gets more complex but you said you didn't need
max security and it sounds like ram/speed are issues
(aren't they always!)

The big issue is always key security, much easier
to crack the delivery system and get the key, or
crack the encryption algorithm within the product,
than to crack a message itself. Have you read up on
public/private key systems?

-Roman

PS. Although the US cannot export encryption products
to your country you can export your product to the US.
:o)

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\08@063746 by Germain Morbe

flavicon
face
Roman,

thanks for your warning, i see the reasons why the feds think as they do but
i do not share their point of view because of the following facts.

As you ask me, YES i read some excellent works about cryptography. I found
all of them on the internet and i never had been adviced that the
information on such sites is for US citizens only. Further i found a posting
anywhere some days ago, saying that its not a federal offence for US
citizens to post crypto-mathematical background on websites. But it seems to
be an offence to post an assembly routine.

That way the federal restrictions would only apply to persons not able to
code such a routine. People who are able to understand the math behind
(which i consider to be the real dangerous people) are free to get such
information. And they are almost definately able to write any crypto routine
they ever need themselves.

Germain Morbe

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\08@085444 by Mike Bakula

picon face
Germain,

I don't have any information on implementations for the PIC, but I can
recommend to you Bruce Schneier's "Applied Cryptography".  It has
extensive, easy to read discussions of protocols and algorithms, including
examples of a  number of cryptosystems in C.

Applied Cryptography Second Edition: protocols, algorithms and source code in C
Copyright 1996 by Bruce Schneier
Published by John Wiley & Sons
ISBN 0-471-110709-9

The book is exportable, but you'll have to type in the examples
yourself....  :)

-- Mike Bakula

At 10:49 AM 1/8/01 +0100, Germain Morbe wrote:
>Hi,
>does anybody know a short 32bit block cipher routine for the pic. The keylen
>need not be more than 64bit. I already saw TEA but its blocklen is 64 bit
>and key is 128. Therefore i think one could save codespace and even ram. My
>requirement is not extremely high sercurity but rather good mixing from one
>input bit to more (best 50%) output bits.
>
>Germain Morbe

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\08@092801 by Alan B. Pearce

face picon face
>The book is exportable, but you'll have to type in the examples
>yourself....  :)

I have always wondered about buying a copy of one of these export limited things
while in USA, and carrying it out in my luggage, CD included. Or on the other
hand posting the CD seperately...

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\08@180026 by mike

flavicon
face
On Mon, 8 Jan 2001 07:37:37 -0600, you wrote:

>Germain,
>
>I don't have any information on implementations for the PIC, but I can
>recommend to you Bruce Schneier's "Applied Cryptography".  It has
>extensive, easy to read discussions of protocols and algorithms, including
>examples of a  number of cryptosystems in C.
..then read his recent book 'Secrets and Lies' to get the whole
picture, i.e. the actual encryption is often the least of your
problems.
{Quote hidden}

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\08@181526 by Mike Bakula

picon face
At 11:04 PM 1/8/01 +0000, Mike Harrison wrote:
>On Mon, 8 Jan 2001 07:37:37 -0600, you wrote:
>
> >Germain,
[...}
>I can
> >recommend to you Bruce Schneier's "Applied Cryptography".  It has
> >extensive, easy to read discussions of protocols and algorithms, including
> >examples of a  number of cryptosystems in C.
>..then read his recent book 'Secrets and Lies' to get the whole
>picture, i.e. the actual encryption is often the least of your
>problems.

That would be:

Secrets and Lies: Digital Security in a Networked World
Copyright 2000 by Bruce Schneier
Published by John Wiley & Sons
ISBN 0-471-25311-1  (It's not out in paperback yet)

Also an excellent reference, particularly on the broader subject of
operating relatively securely in networked environments.  Not quite on the
topic of the question though....

-- Mike "and neither, I fear, is this" Bakula

{Quote hidden}

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\09@065820 by dr. Imre Bartfai

flavicon
face
Hi,

you can use also the "pseudo-one-time-pad" technique using a good
pseudo-random number generator, and XORing the result with the text to be
coded.

Regards,
Imre


On Mon, 8 Jan 2001, Mike Bakula wrote:

{Quote hidden}

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2001\01\09@210248 by Russell McMahon

picon face
>you can use also the "pseudo-one-time-pad" technique using a good
>pseudo-random number generator, and XORing the result with the text to be
>coded.

I think that "pseudo one time pad" encryption could be very risky because
you are then at the mercy of the true randomness of your pseudo-random
generator AND the ability of its algorithm to resist detection.  I suspect
that most real encryption crackers (NSA, NBS, anyone with a spare Cray in
their basement) would not have too much trouble with most pseudo-random
algorithms.

TRUE one time pad is uncrackable. If you don't mind the inconvenience of
transferring the arbitrarily large key it will give you more security than
anything else around.

I am by no means a cryptologist but I suspect that using a very large one
time pad with some extra simple rules based on the pads contents as well
would allow you to use a finite length pad with fair security.

Depending on how much hands on input you can tolerate you could perhaps use
a true one time pad to send a message advising which mutually accessible
internet document to use as the current one time pad. This could be
something like a picture file posted by some completely independent 3rd
party on one of the "here is my photo albumn" type sites. If it was evident
to the "cracker" that this was your method and if they had access to all
your incoming internet data the method may be less secure. If you need
security against that level of attack then you should probably be in a
witness protection program :-).

regards


Russell McMahon
_____________________________

What can one man* do?
Donate food daily free !!! -  http://www.thehungersite.com/
Donate Vitamin A!  http://www.thechildsurvivalsite.com/
http://www.rawa.com  - one perspective on Afghanistan
http://www.changingourworld.com    http://www.easttimor.com   http://www.sudan.com

(* - or woman, child or internet enabled intelligent entity :-))

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


More... (looser matching)
- Last day of these posts
- In 2001 , 2002 only
- Today
- New search...