Posted on 04/09/2005 12:23:55 PM PDT by quidnunc
American intelligence agencies have less trouble getting information, than in finding ways to store it. This situation changed in the 1990s as fiber optic cable became the cheapest way to move huge amounts of electronic data at the speed of light. One of the major money pits during the dot.com boom was companies laying hundreds of thousands of kilometers of fiber optic cable across land and under the sea, replacing the older, much lower capacity, copper cables. The U.S. has a modified nuclear attack submarine that could tap into those undersea cables, and did so successfully several times. But these new fiber optic cables moved enormous amounts of data, with five gigabytes a second being a common throughput. To give you a sense of what that means, consider that some iPods have 60 gigabyte hard drives on them (using tiny, one inch hard drives). But a fiber optic cable can fill up that 60 gig drive in 12 seconds. It can fill 300 of those drives in an hour, 7200 in a day, and 216,000 in a month. Even the U.S. Navy’s newest and largest attack sub, the USS Jimmy Carter, which is specially equipped to tap into fiber optic cables, can’t hold enough hard drive or tape drives to hold more than a week or so worth of data. The problem is no longer one of grabbing the data, but of quickly finding what you need. The USS Jimmy Carter is receiving hundreds of millions of dollars worth of new equipment, including some very powerful computers. Like it’s smaller, Cold War predecessor, the USS Parche, the Carter has an underwater “joining room” for splicing, and tapping into, fiber optic cable. This is a very tricky task, considering the high voltage running through the cable, and the need to tap in without interrupting service, and alerting the cable operator.
Then again, some 90 percent of transoceanic fiber optic cables eventually cross American or British territory. So getting into the cable is not impossible, or very secret. What is kept very secret is any news about the software, and other technology, that would be used to scan the data stream coming through the fiber optic cable. But that raises another question. How long are you going to park the USS Carter over that tap in order to filter its throughput? All the more reason to believe most of the taps are being made on land, or close to land so another cable can be run to a land station containing computer equipment to handle the filtering. But the USS Carter would have its hands full tapping all the new undersea fiber optic cables out there that don’t cross friendly territory, or planting other sensors. Whatever the USS Carter ends up doing, it will be decades before the general public knows the details of what is inside that sub, and what exactly it does.
-snip-
An Oversimplified Overview of Undersea Cable Systems
Power on repeatered cables When repeaters are needed they must be powered. The standard approach is to send a constant current of about 1A from one end of the cable to the other, along a copper sheath, which lies outside the Fibers and inside the armour (if present). Each km of cable offers a resistance of some 0.7 ohm, and the voltage drop across each repeater is typically 40V (on four Fiber-pair cable), leading to a requirement of close to 10 KV across a typical 7500 km transatlantic crossing with 100 repeaters. In branched cable systems the power management becomes somewhat more complex, and the branching units incorporate very high reliability relays to cope with the power reconfiguration needed in case of repairs.
In a slightly different take, while moving and storing data rapidly is necessary, it is not sufficient. Where we fall very short is in the analysis end, where collected data often sits for days / month / years wanting for analysis and interpretation.
Can you imagine sorting through a data base of every cell phone call in a day in a metropolitan area. Given my reaction to being forced to listen to one unwanted cell phone call, I would go stark-raving mad trying to listen to a whole cities worth.
For that matter, I don't think one taps a fiber line like a conventional phone line. You can cut and splice fiber, but you can't apply a 'T' tap.
We've concentrated on tapping undersea cables for some time.
We used to take subs inshore to spy on cable transmissions as far back as when the USSR was still in existance.
Here's a short description of the USS Jimmy Carter and its capabilities, insofar as they're known publically.
http://www.deagel.com/pandora/?p=pm00116002
Within the same sheath. Parallel right next to the fiber itself.
Sad...but probably true.
Look at Paracel. Look at Mercury. Now think of how many BILLIONS the NSA can afford to spend on toys even more advanced than these. General-purpose speaker-independent speech recognition and sophisticated filtering are probably applied to EVERY phone call into and out of the Middle East.
"We" were also going to mine manganese nodules off the sea floor. I'm sure you know what they were actually fishing for.
The "noise" is not a problem wihen it comes to conventional voice calls. There are only so many international calls live at any one time. I have no inside information on how the NSA does it, but an educated guess would be:
Speaker recognition to identify calls by people they want to listen to. This eliminates the need for tapping specific lines (that's for those lAm3rz at FBI, at the NSA, they listen to wholesale quantities of traffic).
Speech recognition to turn calls into transcripts. They can apply many times the power of the most powerful PC to this task.
Lots of sophisticated filtering to find the interesting transcripts.
Lots of sophisticated filtering to find the interesting transcripts.
I saw a CIA analyst on a news show a few years ago. He explained that the problem almost always requires human interpretation. Evil people will never talk openly about their plans. They'll use ever-changing "street" lingo or say simple things like "did you receive your order as promised?" Stuff we all do every day. Plus for voice recognition, we'd need a sample. Do we even know what Zarqawi sounds like? I don't mean some scratchy tape, but well enough to feed a sample for voice-recognition over a 4kHz limited voice line? And you can bet Bin-Laden isn't giving direct orders. Ultra-heavy processing power is a great assist to those who need to protect us, but it still requires the "man on the ground" to keep the system fed with up to date information.
For speaker ID, you need a sample. For speech recognition, you don't.
There are not an overwhelming number of international calls. I'd bet there are no more than couple hundred lines in and out of Afghanistan - if that. Perhaps a couple thousand in and out of Iraq. We probably p0wneD Iraq's voice and data networks a few months before the war started, and we probably still monitor those networks closely.
Satellites have much lower capacity than the terrestrial phone network, and can be comprehensively tapped at the downlinks. VOIP calls are probably hard to monitor, and encrypted IMs are probably quite hard. But conventional voice calls between countries and within a few interesting countries are not too much to monitor.
this is, of course, known as "jimmying" the cable
The cables carry electricity to power repeaters.
Technology isn't good enough to shine the light in on one side of the ocean and see it on the other side without amplification in between.
Would you even need to "splice into" the electronics directly? Wouldn't there be enough RF leakage to pick it up just by parking an appropriate antenna next to the repeater?
If I had to lay money on it, I'd still bet that Bin-Laden was vaporized early in the war, and the CIA has been giving out orders in his name as a "sock puppet" in order to sucker in more terrorists. "Abdul, meet your contact at 123 Falafel Street at 11:00pm". Then poor Abdul "just happens" to get nabbed by the Marines as he pulls up to the parking lot. Of course, we have to put out a "Bin Laden" tape every few months to keep up appearances.
The body of the repeater is pretty heavy-duty metal, and any RFI that could be used to pick up the signal on the cable won't travel through salt water. Possibly possibly the power conductor in the cable could pick this up and carry it out of the repeater. Maybe something could be glued onto the cable?
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.