Re: Digest for opendoorhack-a-thon@googlegroups.com - 2 Messages in 2 Topics

[Al Jigen Billings]

Wireless controllers would seem to be an invitation to have your door hacked by someone outside of it.

[Spenser Gilliland]

Somewhat true, the main problem would be a DOS attack against the
network by causing interference on all channels. However, this is
very similar to if someone cut off the hard line in a wired network.
It's an error condition that should be accounted for and cause alarm
bells to ring.

Snooping wise it's not reasonable to hack the AES128 encryption that
is standard on all 802.15.4 implementations. Assuming that the
implementer doesn't use a default or weak key.

Anyways, that's my opinion on 802.15.4 sensor/embedded networks.

Spenser

--
Spenser Gilliland
Computer Engineer
Illinois Institute of Technology

[james p sweeney]

We need to keep this in perspective. Yes, wireless would be hackable (be it DOS or unauthorized access). But not any more so than the physical lock that is in the door. If someone wants cause problems for any of these spaces, wireless network circumvention is unlikely to be the easiest way.

So, is the goal here to design a maximally secure system or to design something that is at least as secure as the physical locks that we're supplementing.

Sweeney