James Tucker to me to Andrew Farmer: > <snip> > > > - RF keyboards don't exist. Nobody's *that* unconcerned about security. <<snip>> > > In case you don't know, "typical" RF cordless keyboards and mice have > > been recorded to have effective ranges to around 150m (way beyond what > > the manufacturer's specs say): > > > > http://www.aftenposten.no/english/local/article427668.ece > > > > http://www.securityfocus.com/archive/100/374785 > > Amusing articles, but there are lessons to be learned here. > 1) RF ranges are based upon the environments the transmitters and > receivers are in.
What effect would a steel-reinforced concrete structure with mainly windows (and/or other non-ferrous material on the outside wall) have? My _guess_ is that it could act as something of a directional "horn", focussing more RF energy out the "soft" wall due to signal reflection off the other walls. This may explain the 150m result in the Norwegian apartment scenario from the first article. > 2) RF can be boosted (both ends effective) by adding GAIN to ONE END > ONLY. (Yes that means the coke can with a little wire out of one end > hanging out of the next door neighbors kids window is in fact a > wireless tapping antenna (joke, but this is not by any means > impossible)) 8-) True, but I see little relevance to above anecdotes for as far as we can tell, the "victims" in both cases were ordinary folk using off-the- shelf equipment. Of course, if the products being used had wide manufacturing tolerances, I guess there could be an issue??? <<snip>> > > Agreed -- these were stabs-in-the-dark by folk who forgot to understand > > the problem description before responding... > > Are the wireless device and speech recognition ideas really that much > more "likely", they are all "ideas" and are all possible in some way > or another. On balance, yes, the wireless and speech recognition suggestions are more likely. Of course, it may turn out in this case that they are not the explanation, but based on a great deal of experience and the event descriptions given, I'd say that those are the more likely of the suggestions made to date. Regards, Nick FitzGerald _______________________________________________ Full-Disclosure - We believe in it. Charter: http://lists.netsys.com/full-disclosure-charter.html
