home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!ukma!cs.widener.edu!dsinc!netnews.upenn.edu!netnews.cc.lehigh.edu!news
- From: chess@watson.ibm.com (David M. Chess (863-6665))
- Newsgroups: comp.virus
- Subject: re: Math Models of Polymorphic Viruses
- Message-ID: <0001.9301281842.AA17847@barnabas.cert.org>
- Date: 18 Jan 93 20:54:41 GMT
- Sender: virus-l@lehigh.edu
- Lines: 39
- Approved: news@netnews.cc.lehigh.edu
-
- >From: ygoland@edison.SEAS.UCLA.EDU (The Jester)
-
- >The question is:Given functions VX() and VY(), can I determine if
- >they are both members of the same tree?
-
- Yes, you can: they are! *8) Consider V0, where V0(V0,K) produces a
- completely random string of bytes, using K (derived from the current
- machine state, or whatever) as a key. Any other VX and VY are in the
- tree leading from V0 (in fact, they're both on the very first level of
- the tree). That's a limiting case, of course, and rather silly, in
- that most of the offspring of V0 aren't V's at all in any useful
- sense. But clearly for any VX and VY you could write a V<X,Y> which
- produced either VX or VY depending on the parity of K. So again any
- pair VX VY are in the same tree.
-
- A related problem which doesn't have a trivial answer is something
- like: given VX and VY, is VY anywhere in the tree leading down from
- VX? That is, is VY a descendant of VX? For the VXs that are existing
- viruses, answering this question is trivial. It's interesting to
- think about how hard the question can be (I mention this in passing in
- the beginning of my "Virus Verification and Removal..." paper, which I
- think is available on VIRUS-L).
-
- In Hoffman's "Rogue Programs", there's a paper by Len Adleman called
- "An Abstract Theory of Computer Viruses". It contains what seems to
- be a proof that it's possible to design a virus so that given one
- instance of an infected object, it's not decidable whether or not
- another object might be a descendant of it. However, I don't
- understand the proof; I'd love to hear from someone that does! The
- next proof in the paper softens the blow: it seems to be a proof that
- you can come close enough, by deciding whether or not another object
- is EITHER a descendant of the captured virus OR an element of the
- "germ set" for the virus, where the "germs" of a virus are the set
- (roughly) of droppers of the vius.
-
- - - -- -
- David M. Chess \ Nothing moves;
- High Integrity Computing Lab \ where would it go?
- IBM Watson Research \
-