Slashdot Mirror

I am not a security expert. Just a laid off SysAdmin.

What you are reffering to is called "Air Gap Security" and is used by the Government on all of their most sensitive computers. Do you think that the President's laptop that he uses to connect to the Internet is the same one that he uses to compose high-level briefings on?

No, of course not. Any computer that you are TRUELY paranoid about keeping the data secure would be:

1. In a seperate room with no possible network connection. The power outlet and the lights are the only wires coming near the room. No wall in the room is an external wall in the building.
2. That room is shielded against stray EM frequencies. Slashdot had an article on Tempest a while back on it.
3. You should frequently search for bugs in that room and the surrounding rooms.
4. Have some form of security system in place.

It seems to me that if you went to all of that trouble to secure a computer system from monitoring, you either have a business that has really outrageous security needs, an organization that really wants it's members to remain confidential, a citizen who is just freakin' fed up with our rights getting trampled on, or someone who is up to no good.

Notice how in the above paragraph only ONE situation warrants a warrant? Without having to go to that extreme of measures to protect our privacy, what reasonable solution can be provided to the average citizen to ensure theirs?

Well I have my own Cisco based [1 2 3] information which sums up networking to a tee. Security Focus, Packet Storm, SpyKing, and Cryptome all cover the other areas for information when I need it. Is it me or in the past 2 years did everyone jump on the "Hacker" bandwagon writing books on information that's already a point and click away? Not taking anything away from the book, but Information Security Management Handbook 2001, Cisco's Routing TCP/IP, and other security books in my library have done me justice. Makes I guess a nice intro for newer users, but personally I don't like books with "Hacker" in them, they tend to be geared for those with little clues, and who are often too lazy or dumb to find information and study it on their own.

Spyking.com has some interesting information on how to build your own TEMPEST Interceptor. I haven't tried it so I don't know if it works or not. Has anyone tried building one of these?
You can also find more information about TEMPEST in an article at Spyking.com

Spyking.com has some interesting information on how to build your own TEMPEST Interceptor. I haven't tried it so I don't know if it works or not. Has anyone tried building one of these?
You can also find more information about TEMPEST in an article at Spyking.com

Spyking.com has some interesting information on how to build your own TEMPEST Interceptor. I haven't tried it so I don't know if it works or not. Has anyone tried building one of these?
You can also find more information about TEMPEST in an article at Spyking.com

For those of you that care, here is the real link:
http://cryptome.org/nacsim-5000.htm

also, here is a really neat site with an analysis on what this stuff really means:
http://eskimo.com/~joelm/tempest.html

and yet more great reading:
http://www.austinlinks.com/Crypto/tempest.html
http://www.thecodex.com/c_tempest.html
http://www.spyking.com/datascan.html

Got this from here. It's interesting to read if you know the basics.

The general idea is to set up an L-R-C circuit and destabilize it. For example, we know the change in emf (E) is related to the ke quantities (L = indcutance, R = resistance, C = capacitance) by:

L (d^2 I(t) / dt^2) + R (dI(t) / dt) + I(t)/ C = dE(t)/ dt

where I(t), and E(t) are time-variable

Under suitable conditions (cf. Spicer, 'Solar Physics', v53, p. 305, 1977) it is possible to induce a voltage 'spike' which is the equivalent of a HERF burst. In his Fig. 4, Spicer shows a profile of such a burst, generated using an ST-tokamak.

Of course, ST-tokamaks are not needed to accomplish this. It is fairly well known that *all* electrical circuits containing an inductance (L) are intrinsically explosive (cf. Alfven, 'Cosmic Plasma', p. 34, Dordrecht-Reidel, 1981).

Thus, if such a circuit is somehow disrupted, there is an explosive release of magnetic energy, of order:

W = 0.5 LI^2

Where L is the inductance (say of the inducting coil in the circuit) and I is the current. Say, for example, that L = 10 H (10 Henries) and I = 20A, then:

W = 0.5 (10H) (20A)^2 = 2000 J (Joules)

This is a fair amount of energy.

In practical HERF circuits one would incorporate a 'double layer' say by using a capacitor across which some maximum value of voltage V(D) might be applied. As the current is made to exceed a threshold value I(D), a voltage drop V(D) occurs across the capacitor up to the current I(ex) when it explodes and disrupts the current (cf. Alfven, ibid.).

When switched on, the current increases at rate:

dI/dt = (V(b) - R I/ L)

(V(b) is intrinsic emf of circuit)

Without the 'double layer' (capacitor) it reaches saturation at:

I(s) = V(b)/ R

If I(D) I(D) is given by:

dI/dt = [V(b) - V(D) - RI]/ L

The current will tend towards a saturation value:

I(s)' = [V(b) - V(D)]/ R

If I(ex)
The explosion will generate a high energy radio frequency pulse.

This pulse, if proximate enough to an electronic device (i.e. computer) will render it inoperable

• Destructive Energy Devices
• Electromagnetic Pulse (EMP) & TEMPEST Protection for Facilities
• Hardening your Computing Assets against HERF
• How to build a HERF Gun