Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site petrus.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!gamma!epsilon!zeta!sabre!petrus!karn
From: karn@petrus.UUCP (Phil R. Karn)
Newsgroups: net.ham-radio
Subject: Re:  Ham Encryption
Message-ID: <479@petrus.UUCP>
Date: Sat, 24-Aug-85 16:28:27 EDT
Article-I.D.: petrus.479
Posted: Sat Aug 24 16:28:27 1985
Date-Received: Sun, 25-Aug-85 12:59:23 EDT
References: <860@brl-tgr.ARPA>
Organization: Bell Communications Research, Inc
Lines: 24

We are fortunate that the encryption prohibition is worded the way that
it is, with the key point being whether there is intent to obscure the
meaning of a message. This allows us complete freedom to experiment with any
sophisticated, nonstandard encoding method we choose without worrying
about difficult it might be to monitor. As long as it serves some valid
communication purpose other than hiding the message, we're OK.

A case in point is the 9600 baud FSK modem design by Steve Goode, K9NG.

Steve's biggest problem was to get rid of the DC components in a data stream;
this simplifies the design greatly, since you can use AC coupling and not
worry about things like op-amp offsets, crystal tolerances, etc.  One
approach is Manchester encoding, but this effectively doubles the bandwidth
required for the signal. Steve's solution was to add a pseudo-random stream
to the data at the transmitter, removing it at the receiver. In the
commercial world, this technique is commonly known as "bit scrambling", a
term with unfortunate connotations. (The technique DOES resemble an
encryption system, although one based on a linear feedback shift register
would be very easy to break.) However, since Steve's intent is not to obscure
the meaning of a message but rather to simplify the hardware design, it is
acceptable. (Of course, it also helps that he published the algorithm and
encoding/decoding circuit at the last ARRL packet radio conference.)

Phil