[Cryptography] how to detect breakage -- lures etc.??

[Arnold Reinhold]

On Mon, 6 Jan 2020 20:45 Bob Wilson asked:

> In the photograph of a KL-7 device, link embedded in
>> https://www.cryptomuseum.com/crypto/usa/kl7 <https://www.cryptomuseum.com/crypto/usa/kl7>
> 
> I am surprised by the "Confidential" markings on several labels. I would 
> expect the device itself to be classified much more securely than that. 
> I realize those labels are not official classification markings, but 
> they might mislead someone into divulging details that were intended to 
> be more highly ranked.
> Or was all of the "more critical" stuff contained in removable parts 
> such as rotors, with some scheme to make sure their higher 
> classification was separately recognized?

Those are the official classification markings and yes, they only refers to "more critical" stuff contained in the removable subassembly, which if you look carefully is marked KLK-7/TESC, with a “Reg. No.” The base of the KL-7 does not have the Confidential label and its serial number is marked “Non-Reg No.” The U.S. practice of having removable rotor assemblies with a higher classification goes back to to at least the SIGABA, which was designed in the 1930s. Security is primarily provided by the keys, per Kirckhoffs.

One hint as to why the KL-7 rotors had such a low classification is in the NSA 1969 Cryptologic Damage Assessment from the North Korean’s capture of the USS Pueblo (https://nsarchive2.gwu.edu/NSAEBB/NSAEBB453/docs/doc25.pdf <https://nsarchive2.gwu.edu/NSAEBB/NSAEBB453/docs/doc25.pdf> ).  One of the crypto machines on board was a KL-47, a variant of the KL-7. The NSA concluded that it was highly likely that KL-47s had been compromised in the past since: "many thousands of equipments have been in use world-wide for more than a decade, including about 10,000 functionally identical KL-7 equipments held by NATO countries. The equipment, associated maintenance manuals, and operating instructions have been exposed to possible compromise a number of times over the years. In recent intelligence operations, the USSR has made efforts to obtain key lists and rotors, but have shown no interest in the equipment itself or its supporting documents. Damage to SIGINT interests as a result of loss of the KL-47 to the USSR is also judged to be negligible.” Given that the KL-7 was intended for widespread field use, it's low classification would seem to make sense.

The Cryptologic Damage Assessment has lots of interesting stuff, including accounting of the various key lists on the Pueblo, along with many that were offloaded before it left port, that give a sense of how complex the U.S. military communications networks were. 


Arnold Reinhold


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.metzdowd.com/pipermail/cryptography/attachments/20200107/c916789c/attachment.htm>