perm filename SIGNAT[F83,JMC]1 blob
sn#729852 filedate 1983-11-06 generic text, type C, neo UTF8
COMMENT ā VALID 00002 PAGES
C REC PAGE DESCRIPTION
C00001 00001
C00002 00002 signat[f83,jmc] Computer terminals have analog signatures
C00005 ENDMK
Cā;
signat[f83,jmc] Computer terminals have analog signatures
A statement in a news story about the FBI confiscating
the terminal of a UCLA student who was breaking into computer
systems and committing sabotage led to the following idea.
A computer terminal transmits by keying between two
frequencies. These frequencies are maintained exactly and
neither is the timing of switching or the relative amplitudes.
Therefore, an analog recording of a terminal session can
identify the terminal as long as the signal hasn't yet passed
through a digitization stage such as those used in packet
switched networks.
If the interloper situation proves bad enough to warrant
it, TIPs and TACs could be provide with analog tape recorders
that could be switched on from the network headquarters when
a computer detected attempts to try large numbers of passwords
and complained. Requests for monitoring could also be made
manually from the computer under attack.
Reasonable privacy can be assured by requiring a request
from the computer being accessed before monitoring can take place.
Since the information is going to that computer anyway, the only
additional information the monitoring gives is the analog signature
of the terminal.
Naturally, the ARPAnet shouldn't install a lot of monitoring
equipment unless there is a continued problem of damage by
unauthorized access of sufficient magnitude to warrant it.
John McCarthy
Stanford University
1983 November 6