Army signals security & "Clansmen" series radios

[Owen Lewis]

----- Original Message -----
From: "David Howe" <DHowe@Hawkswing.demon.co.uk>
To: <ukcrypto@maillist.ox.ac.uk>
Sent: 06 September 2000 17:23
Subject: Re: Army signals security & "Clansmen" series radios


> ..You
> can then factor in your groupings by using the largest possible key that
> doesn't contain a non-recipient (so, you could send to two flotillas, and
to
> three ships that aren't in a flotilla that is effected but should know
> anyhow)

You surely could but PK is not an essential to such ad hoc 'you and you but
not you' encryption. an

> The downside is traffic analysis - if you suddenly notice three ships and
> two flotillas getting shared encypted traffic, you might start looking for
> connections.

But traffic analysis should not help. It should detect the broadcast but
cannot tell how many recipients there were and even less who those recipents
were. If you assume that the recipients must in turn transmit
acknowledgement and that these will be detected:

            1. The acknowlegements will not (should not) identify the
recipents by name/unit/formation etc.

            2. Such acknowledgement does not need to be transmitted in a
timeframe or by a means that that will label it as an acknowledgement.

Would one expect that traffic analysis is likely to help identify the
strength, composition and location of a nuclear strike force from a homeland
broadcast half a world away?


however, if you lose the key material on a given ship, you
> compromise (at worst) the groupings that ship is a member of, not the navy
> as a whole.
>
>
>