ronpin
Imagine Reality
join:2002-12-06
Nirvana
 ·AT&T Southwest
 ·Charter Pipeline
| reply to Fraoch
Re: Anonymously Track a PC Anywhere on Net
From the cited PDF paper...
...and we show how one might use a Fourier transform on packet arrival times to infer a device’s clock skew. ...
They almost had me believing it. Packet arrival times have random influences that no "Fourier transform" could account for. This is bullshit -- the paper is a fraud -- but I"ll keep reading it just to make sure. Besides, I'm pretty sure that TCP does not waste 32 bits on a time stamp unless there is a real-time/ordering requirement (but that could have changed in the last 5 years since I dealt with it). ICMP request are mentioned -- but don't most router firewalls block those anyway?
--
Lord protect me from your followers |
|
teambnet
Team B Group
Premium
join:2003-05-06
Chicago, IL
| I agree: total BS. It would be impossible for a single deployed solution to parse traffic from a constantly evolving number of arrangements behind customers' public interfaces- especially if LAN side hacks appeared that were designed to overwhelm and not just obscure. |
|
tquade
join:2000-10-14
Regina, SK | Concur, although, it could be done with a laplace transform, a bit of convolution and a sprinkling of negative phlogiston.
Ted |
|
MoeDumb
"America Si, Obama No."
Premium
join:2002-09-23 | It's crackers to slip a rozzer the dropsy in snide.
(I don't know what the hell I'm talking about either.)
--
"tick...tick...tick..." »www.jtf.org/ |
|
jaxjaguar
join:2001-05-29
Northridge, CA
| reply to ronpin
On the BS bandwagon myself. I'm no TCP protocol expert, but doesn't a NAT router change the time stamp when it NAT's the packet? And if it currently doesn't, I'm sure it's just a simple firmware change to add that feature and make the ISP's efforts worthless.
So there's nothing to worry about here. |
|
