Re: Biometric Encryption: the solution for network intruders?

From: Walter Roberson (roberson@ibd.nrc.ca)
Date: 04/14/02 

From: roberson@ibd.nrc.ca (Walter Roberson)
Date: 14 Apr 2002 21:32:22 GMT

In article <uelhis3vw.fsf@earthlink.net>,
Anne & Lynn Wheeler <lynn@garlic.com> wrote:
:try and imagine a encryption technology where it was
:expected that only a random 30 percent of the bits matched between the
:encryption key and the decryption key.

That should not present particular difficulties. Apply
a Forward Error Correction coding (e.g., a convolutional coding) to
the core of the encryption key, parameterized to expect up to
70% bit loss. As a very -rough- estimate, this would stretch the
core of the key out by about a factor of 3 1/2.

Then, when presented with the lossy decryption key, run it through
the error correction algorithm, recover the core of the key,
and use the core to do the decryption.

Translated into the realm of Biometrics, what this means is
that if you use a biometric that extracts several times more
information than is required to uniquely identify a person, then you
could allow misreads on a substantial number of the items and
yet still decide upon the unique identity.

Note: this technique would not necessarily be of any value
against people who are deliberately attempting to pass themselves
off as someone else. One then gets involved with questions of the
amount of resources required to deliberately fool the reader, and
questions about the security of readings that have been taken.

If strong biometrics were to become widespread, then preventing
readers from being hacked up or compromised would become a big issue.
If you put a reader in every bar and licensed restaurant (for example),
then how certain are you that no-one *anywhere* has gone into the circuit
level and learned how to tap the raw signals closely enough to
be able to fake a reading?

Relevant Pages

  • Re: Blowfish Encryption Question
    ... >>SHA and then use that output as the encryption key. ... >>before decryption. ... original password. ... programs like PGP and my own Puffer program, ...
    (comp.lang.pascal.delphi.misc)
  • Re: PGP scripting...
    ... found in the script anyway). ... destroying the plain text after encryption. ... pool of data with a single trust (decryption keys), ... check the integrity of the encryption key. ...
    (SecProg)
  • Re: alt.computer.security
    ... decryption key for the file to become accessible. ... I'm considering the encryption key as ... The keyfile will *not* be kept on the same computer that it was used ... the floppy in order to decrypt the PC file, ...
    (microsoft.public.security)
Loading