kaos6565656
join:2005-12-11
Norfolk, VA
| Fiber Optic cable tapping?
I was once told that you can tap a fiber optic cable by bending the fiber optic cable at a 45 degree angle. This allows just enough signal to leak but not enough for either side to know. Has anyone actually accomplished this feat? Oh, Yes the person who told me had the knowledge the education and the security clearance to know. |
|
Edrick
Premium
join:2004-09-11
Orlando, FL
| I don't know much if anything about fiber optics but wouldn't bending the cable cause the light not to bounce inside of the cable and reflect the rest of the way down the cable expecially if your cutting the cable to "tap" into it? Well bend it 45 then splice it but doesnt any break in the cable or removal of the sheilding cause the light to stop reflecting
--
Ricky SmithVerizon FIOS User15 Mbit Down 2 Mbit Up |
|
53059959
Temp banned from BBR more then anyone
join:2002-10-02
PwnZone
| reply to kaos6565656
it is possible to bend it and let a small amount of signal leak out, if its a major backbone cable, the isp will probably notice. interruption could be detected by time domain reflectometry. for the most part tapping a fiber optic cable is expensive or impractical, but I mean a large government agency like the cia could do it.
»www.oysteroptics.com/index_resources.html |
|
LazMan
join:2003-03-26
Angus, ON | reply to kaos6565656
While it is possible to detect light without peircing the cable (we do it all the time) - the microbends introduced cause attenuation to go through the roof; and the losses are easily detectable.
Laz |
|
HerndonKeith
join:2006-02-03
Herndon, VA
| reply to kaos6565656
At a previous job, we introduced a network analyzer that would tap our fiber and check destinations for packet loss, latency, and excessive jitter. The fiber tap we used was from blackbox networks. I think the tap module costs roughly $900. Basically it works by taking the light, from the connector of the fiber, and reflecting the beam evenly between 2 output connectors. It was quite impressive to see it actually work. Even more impressive to find that there was very little loss. I think the most we saw was about -4db loss on the link. I am glad I did not have to deal with bending my cables at 45 degrees.
Just out of curiosity, When you say "bend at 45 degrees" are you talking about a sharp 45 degree bend or somewhat gradual? Is this technique, you are talking about, done on multi mode or single mode? I cannot picture a standard patch cable bending at a sharp angle without a noticable degrigation. Also, How does one get light to "leak out" without having to pierce the outter jacket and the buffer coating and what about the kevlar getting in the way?
--Keith |
|
kaos6565656
join:2005-12-11
Norfolk, VA | You'll have to ask the US Navy fiber tapping group. LOL
However, I was given a preview in a networking class I took. |
|
fiber_man
Things Happen For A Reason
Premium
join:2001-01-27
Port Saint Lucie, FL
 ·AT&T U-Verse
| reply to kaos6565656
As Lazman said you can put micro bends on fiber to detect the light signal and the direction of the traffic. We do it every day to "tone"(2khz) fiber for new jobs. Just to make sure we have the right fibers. You can see every one of them on a trace with an OTDR.
PON networks will actually split the fiber bandwidth. 1 to 32 ratio or higher. I.E. 1 fiber to a pedestal with 32 fibers going to 32 homes.
--
GO NOLES!! |
|
MacLeech
The one and only
Premium
join:2001-07-14
SoCal
2 edits | reply to HerndonKeith
said by HerndonKeith  : I think the tap module costs roughly $900. Basically it works by taking the light, from the connector of the fiber, and reflecting the beam evenly between 2 output connectors. It was quite impressive to see it actually work. Even more impressive to find that there was very little loss. I think the most we saw was about -4db loss on the link. That sounds like a standard 2-way splitter/coupler.
It terms of calculated numbers, dB (without any reference units, i.e. dBmV, dBm, dBc) are logarithmic ratios:
9dB loss = 87.5% signal loss (1/2/2/2)
6dB loss = 75% signal loss (1/2/2)
3dB loss = 50% signal loss (1/2)
0db = nothing gained, nothing lost
3dB gain = 200% signal gain (1x2)
6dB gain = 400% signal gain (1x2x2)
9dB gain = 800% signal gain (1x2x2x2)
2-way splits usually lose about 4 dB in real life.
--
For official Adelphia support, contact Adelphia. I'm just here for advice... |
|
njellis
join:2001-05-12
Glendora, CA | Show-off! |
|
soliton
join:2006-02-21
Australia
| It is both possible and easy to tap a fibre. The main issue is that you cannot encrypt data at Optical network speeds...yet....Also, what to do with 10s or 100s of Gigabits of data afterwards...Decoding 32 X 2.5 Gbs wavelengths from a dWDM system is physically possible and not that hard, and if a government puts the resources and money into it then it's quite achievable.
Can anyone think of any Governments with vast resources who would like to get their hands on information at any cost? hmmmmmm...
There is equipment available that will detect any fibre tap, but only if used correctly. However, most companies who have private dWDM optical networks do not use this equipment (to frugal) despite it's low cost, as they have been convinced (wrongly) by the dWDM vendors that fibre is secure.
I can think of 3 major global financial institutions who have unprotected fibre dWDM systems in Asia Pac alone..
BTW, you can get 1X2 couplers that will split as little as 0.5-1% of the light from the main fibre core, thats 0.02 -> 0.04 dB loss...hardly noticeable. |
|
EricJRW
join:2001-12-19
Fort Worth, TX
| This makes for an intersting read:
"Secrecy Power Sinks Patent Case"
»www.wired.com/news/technology/0,···,00.html
--
Two wrongs don't make a right, but three rights do make a left. |
|
soliton
join:2006-02-21
Australia | Nice article.
Anyone got anymore? |
|
davef139
@direcpc.com | reply to soliton
A good way to know if stuff is hackable is if your in the US, that if the Goverment rolls it out for there own use its hackable, and yes the Pentagon is wired with fiber and they will not install somthing they cannot bypass themselves. |
|
