A User Authentication Scheme Not Requiring Secrecy in the Computer

In many computer operating systems a user authenticates
himself by entering a secret password 
known solely to himself and the system.  The system compares
this password with one recorded in a Password 
Table which is available to only the authentication
program.  The integrity of the system depends on 
keeping the table secret.  In this paper a password
scheme is presented which does not require secrecy 
in the computer.    All aspects of the system, including
all relevant code and data bases, may be known 
by anyone attempting to intrude.  The scheme is based
on using a function H which the would-be intruder 
is unable to invert.  This function is applied to the
user's password and the result compared to a table 
entry, a match being interpreted as authentication of
the user.  The intruder may know all about H and 
have access to the table, but he can penetrate the system
only if he can invert H to determine an input 
that produces a given output.  This paper discusses
issues surrounding selection of a suitable H.  Two 
different plausible arguments are given that penetration
would be exceedingly difficult, and it is then 
argued that more rigorous results are unlikely.  Finally,
some human engineering problems relating to 
the scheme are discussed.

CACM August, 1974

Evans, A. Jr.
Kantrowitz, W.
Weiss, E.

operating system security, security, authentication,
passwords, one-way encryption, cryptology

4.39 5.39

CA740802 JB January 17, 1978  10:19 AM

2622	5	2622
2622	5	2622
2622	5	2622