NSA Approves First 802.11b Product for Secret Data

joehoya writes "I realize this is a couple of days old, but the National Security Agency recently certified the Harris Corp's Secnet-11 as the first 802.11b system permitted to carry US SECRET level data. See press release. The system integrates NSA crypto with commercial chipset based 802.11b PCMCIA cards and access points to create a secure wireless LAN. Unfortunately, you and I won't be able to buy them, as they are only available to organizations with an NSA COMSEC account."

Comments

Yeah but

[Chanc_Gorkon]

By the time we can buy them, the encrypted brain implants would be the hottest thing. Start looking in the military surplus bins in 2020.

How is this unfortunate?

[drinkypoo]

It's already possible to "leverage" "existing technologies" in order to do secure communications using "commodity hardware".

Or, in English (and not marketdroidspeak) you can have perfectly secure communications over existing 802.11 as long as you encrypt at the protocol level rather than the hardware (link? I need to study my OSI seven layer network burrito) level. So why do we care about this anyway?

Re:How is this unfortunate?

[DAldredge]

You used perfectly and secure in the same sentence. That is all the proof needed to show that you do not know what you are talking about.

Re:How is this unfortunate?

[Anonymous Coward]

real basic encryption is done in hardware its just easy to crack. But who cares you should always use your own higher level protocal encryption schemes anyway. Yes wireless communications are easier to "grab" out of the air, then say grabing your ethernet from your company. In either case you should always use hire level protocalls like ssl and ssh to handle encrypted data or access to accounts.

Haveing "hardware" only encryption is not and will never be a solution.

I look on the current encryption scheme that 802.11b uses as a simple mesure to make recording or watching communication harder. It in NO WAY is a means of totel security. That is always better left to higher level protocals then the link layer in 802.11b, or any networking protocal for that matter.

Yes I only use/allow encrypted connections to all of my wired, and wireless systems.

Re:How is this unfortunate?

[isorox]

PLNTSPA

Physical (wires and stuff)
Link (Ethernet protocols, mac addresses, clever wires etc)
Network (IP)
Transport (TCP)
Session
Protocol
Application

WAP would fit in the bottom 2 layers, with encryption in layer 2 I believe. You personally encrypt in Session, so your encrypted packets get sent over a non encrypted network (routers have to be able to read the destination IP address, dont they!)

Re:How is this unfortunate?

[Openadvocate]

Yes but this is where most people fail.
First you could start by "securing" the net using the "security" available today in 802.11, something all too few companies does.
Then instead of connecting it to your network, you could connect it to the outside of a VPN box, so that you would need to run VPN over it.
If setup right it would work well for those on notebooks, since use the same method to connect to the company network when you are on site or remote using the internet. The difference is that on site, you would use your 802.11 card and remote you would use a ethernet/modem connection to the internet to connect.
I have tried this and it can work, you can even make it work so that the people in the sales dept. can understand it.

With that said, I am still amazed by amount of companies who install a 802.11 net without securing it at all. I have tried it many times, I open my notebook connect to the network and ask them for a account so I can login. Then they ask me how I got connected to their network and I tell them that I am just using their wireless net.
After that I normally can sell a few hours extra to secure their wireless net. And recommend that if they want that extra security, they sould do something like I mentioned above.
And so I end the day with selling a few extra hours and maybe some VPN boxes.

Re:How is this unfortunate?

[Sheetrock]

SSL and SSH are great when you can use them, but there are circumstances when software-level encryption is prohibitive for one reason or another (too costly, unavailable, breaks things). Windows file shares are a pretty good example of the latter, as are NFS shares or a system that just doesn't support it.

You can establish an SSH session to a Linux system rather easily, but maybe the six-year-old AS/400 sitting on the internal corporate network doesn't. Upgrading the AS/400 is an expensive proposition. Implementing a VPN solution, perhaps at the border router or with another internal system, is probably the best method with current 802.11 hardware. But if the hardware supports encryption, everything is transparent.

Hardware-level encryption certainly doesn't absolve the end user of the responsibility of encrypting Internet communications. However, on an internal network, I think you should be able to trust your wireless connections to the same degree you can trust your wired ones. At worst, hardware-level encryption is a wasted step, but it would give some protection to the average user who expects the internal network to be protected.

Re:How is this unfortunate?

[Cadre]

I have a question that's related: how do I make sure that nobody unauthorized is connected to my network?

IPsec

Re:How is this unfortunate?

[sczimme]

You seem perfectly secure in the knowledge that he does not know what he's talking about.

Oh, wait a minute.

Dang.

Re:How is this unfortunate?

[drinkypoo]

I know you were only being clever, but I do wish to emphatically state that I believe that any system put together by humans can be taken apart by humans without destroying it and reverse engineered. The amount of time it takes only goes up and down with the amount of time put into protecting it, and the ratio of intelligences of the engineer and the reverse engineer.

You might say they enjoy a reverse proportion.

(you may now throw tomatoes)

Re:How is this unfortunate?

[doc_side]

already discussed on slashdot

Re:How is this unfortunate?

[crucini]

According to the press release, this equipment allows a network of nodes (not just 2) to communicate, prevents traffic analysis (presumably it always maxes out the channel) and does not yield information about source and destination addresses.

How would you accomplish that with protocol-level encryption?

Re:How is this unfortunate?

[arkanes]

There is no such thing as an internal wireless network.

Make NSA crypto Open Source!

[Lieutenant_Dan]

I think it would be best for national security if we made the NSA crypto algorithm Open Source. The military could benefit greatly from the vast experience and dedication of the Open Source developer community. With their steadfastness, courage, and discipline they could be a great aid in these of needs where we face numerous international threats.

Only when we harness the power available in the Open Source developer community can be achieve fully secure e-mail communications.

Re:Make NSA crypto Open Source!

[joe_bruin]

well, the nsa not too long ago standardized on an open source, patent free digital encryption algorithm for their 'advanced encryption system' (aes), to be used in many forthcoming applications, and replace the aging 'data encryption system' (des). the algorithm they've chosen is called rijndael. here is the source for one implementation.

is this kind of like what you were asking for?

Re:Make NSA crypto Open Source!

[istartedi]

Let's say that the quality of the code is roughly proportional to QN, where N is the number of developers and Q is the quality of each developer.

The alleged value of Open Source is that it allows you to increase the value of N by a dramatic number. Even if the developers are merely average, you can get a higher QN with Open Source than with closed source for many projects.

Of course, if the number of half-finished projects on Sourceforge is any indicator, simply opening up is not enough. You have to have some appeal to developers or you aren't going to raise your N much.

Then of course there is the other factor, Q. Even if you have something really cool, there is no gaurantee that those interested will be any better than average, and you will also have to expend some effort "managing" those who are below average or who are just plain crackpots.

Something tells me that the NSA has no trouble attracting developers with a very high "Q" and in sufficient "N" to do an excellent job.

Yes, I know about the "mythical man month" and that you can't just add up developers as I've suggested. That's why this is just an approximation.

Frankly, I think your post borders on Trollish because you've got "only" and "fully secure" in there; but there are probably plenty of people on /. who will eat up your post, just as there are plenty of people who think that obscurity==security. Of course neither side is right; Open Source isn't a panacea, but giving up obsccurity isn't always such a bright idea either.

Re:Make NSA crypto Open Source!

[khafre]

There is a declassified crypto algorithm, designed by the NSA, and available to you. It's Type 2 (good for sensitive but unclassified) called Skipjack. Available here.

hum....

[tadheckaman]

When will someone take one apart and find out its a and figures out how rip the firmware out of it for use in standard cards?

Re:hum....

[Syncdata]

When will someone take one apart
Excellent Question, especially given the well publicized trouble government employees have in holding on to their laptops. Just cause it's technically secure doesn't mean the laptop itself can't just get picked from an unnattentive employee.

Public or private key?

[thirty-seven]

I wasn't able to find this in the press release. Does anyone know if the encryption algorithm would be public key based, or would it be DEC or something like that?

Re:Public or private key?

[JUSTONEMORELATTE]

I wasn't able to find this in the press release. Does anyone know if the encryption algorithm would be public key based, or would it be DEC or something like that?
If you mean DES, not DEC, don't hold your breath. Evi Nemeth at the University of Colorado had effectively reversed DES in 1991, and the NSA has her work.
--

Re:Public or private key?

[Ayanami Rei]

It's called "Baton" and it was developed by the NSA, the details of the algorithm are Top Secret/Propreitary. It's a Type-1 encryption algorithm, the kind that can be used to encrypt Secret/Top-Secret information, for example, on SIPRNET. Harris/Intersil was licensed to create a security module that implements the algorithm.

Baton is a symmetric key cypher, by the way. I read somewhere it's a 160 or 320-bit key and of course it has various chaining modes. So it's definitely strong. It uses the SHA-1 hash in the protocol too.

Re:Public or private key?

[JUSTONEMORELATTE]

My memory isn't the best, so I may well have a detail wrong.
Evi did an analysis of the algorithm. The point on which she concentrated her efforts was the pairs of large factors(? primes? Pairs of numbers, but I don't recall why they were paired)
Normally, one of these numbers is present in the key, the other in the cyphertext. By making use of large numbers of cycles (ran her code on the "burn-in" floor for Prime supercomputers for several months) she developed a large set of these pairs.
Given this large list, one can take one number from the cyphertext and simply look up the key. No, she doesn't have all possible values. No, she technically hasn't broken the algorithm. Practically speaking, her presentation consisted of taking a 5000+ /etc/passwd file, and decrypting more than 99% of the passwords in a few minutes on a Sun3 workstation.
The NSA has her code. The NSA has her database of numbers, and certainly knows how to run her code to increase the size of the database.
Do I know that this is true? Of course not. I believe these things because Evi told me how she did her work, and that she gave her data and source to the NSA.

--

Re:Public or private key?

[karlm]

Any useful research involving DES most certainly is not corelated with prime numbers. DES consists entirely of xors, register rotations (for the key schedule), bit duplications, bit swaps, and a set of 8 6x4-bit optimally non-linear lookup tables (S-boxes). There are no additions, multiplications, exponentiations, or thier inverses, involved.

Now, it wouldn't surprise me if she used diffential cryptanalysis (DES is known to be quite weak against diffentiacryptanalsis) to do a lot of precomputation to make cracking crypt (whic is based on DES) passwords much faster. Big deal. Anyone still using crypt passwords deserves to get thier passwords compromised.

Correction:

[dj28]

That should be nsa.GOV, not nsa.MIL.

Re:Correction:

[fobbman]

Sure, but .mil seems to be more accurate.

dnetc

[Flamesplash]

Yeah then dnetc could brute force hack the crypto....

Re:dnetc

[timeOday]

Well, unless military satellites are dropping film canisters or tethered with fiber optic cables, I think it's farily safe to conclude that classified data is already being transmitted through your person at this moment.

Re:dnetc

[Flamesplash]

...or tethered with fiber optic cables...

Hey! that would be a great way to keep them from drifting off into space.

preview and submit too close

[tadheckaman]

When will someone take one apart and find out its a Cisco card and makes a firmware upgrade for other cards?

/me hopes this will make it out to the market

[EvilOpie]

I hope that at some point technology like this makes it out to the hands of the average consumer. It's good to see that at least someone is trying to make wireless access more secure. It would be nice to be able to pick up a secure wireless product at some point, and use it out of the box without worries of it being insecure.

But until then, there's always VPN or SSH tunnels. And as an added bonus, you can impliment SSH tunnels for free. (even for web and other traffic... not just SSH data)

Proprietary crypto is lame

[BalkanBoy]

... Bruce Schneier has said this over and over again - it will be a cold day in hell before a proprietary cryptographic algorithm is going to be nearly as scrutinized as a publically available one. I don't see that the algorithm the NSA's using has been disclosed (in the article), and I doubt it will be. Granted, the NSA has probably more cryptographers on staff than anyone else, but that is no guarantee for the quality of the algorithm they are using. This way they may be potentially running on borrowed time until someone figures out a way to attack it...

Re:Proprietary crypto is lame

[Rick the Red]

If the NSA opened their crypto to the public, we'd all see that they're just using PGP like everyone else. By keeping what they use a secret, they keep us guessing if they've cracked what we're using.

Re:Proprietary crypto is lame

[photon317]

However, the NSA is somewhat of an exception to this rule. It is widely known that they are the largest employer of mathemeticians worldwide, compared to any other governmental or private organization, including universities. Therefore, widespread solid peer-review of cryptography can actually happen *inside* the NSA without making anything public to the outside world and they would still get decent results. Add on that the NSA's cryptographers and mathematicians tend to be about a decade ahead of the public/academic world, and it all adds up to the NSA not needing to follow the conventional cryptography peer review mantra.

Re:Proprietary crypto is lame

[Dillon2112]

One of the biggest aspects of military security lies in not revealing what technology they use. Any information given gives a potential attacker a clue where to start...a lack of such information greatly increases the time to even ascertain whether an encryption is even worth spending time on. As anyone who has hacked or tried to find security vulnerabilites knows, one of the nicest things you can hope for when trying to gain access to a server is what software and version its running. If you don't know that, it kind of makes it hard to know where to start.

Re:Proprietary crypto is lame

[nrjyzerbuny]

"it will be a cold day in hell before a proprietary cryptographic algorithm is going to be nearly as scrutinized as a publically available one."

The NSA is the largest employer of mathematicians and cryptographers in the world. World-class peer review is possible within the NSA. How many people peer review crypto? Honestly? This is the same argument used for Open Source software, and the same thing applies, plenty of people use it, and a few actually look over the source, if they break it, or find something they don't like. I would bet that more people look over NSA internal crypto than have looked over most public source crypto. In addition, the people looking at NSA source are all qualified individuals, people who know an S-Box from their asshole.

The NSA is consistantly 10-20 years ahead of the private and scholastic sector. The NSA for example was involved in the creation of the S-boxes for DES. While many people argued that the NSA would weaken the algorithm in an attempt to make it more easily crackable, only later was it discovered that the original boxes were vulnerable to an attack that had not even been discovered by the non-government sector.

You may not trust the NSA, but their in-house review is as good and better than anything you will find elsewhere, even in the much-vaunted open-source community.

Re:Proprietary crypto is lame

[3waygeek]

More like 200 years ahead of the rest of the world in maths theory.

Re:Proprietary crypto is lame

[jareds]

The academic community today would have had a hard time breaking Enigma (assuming the wirings were not known) even with the computing power available to us.

Uh, the Enigma had a few billion keys, tops. You could brute force it in minutes.

Re:Proprietary crypto is lame

[TerryAtWork]

Yes but the NSA is a special case.

Fact is, they are where open source divides by zero.

Software out of the NSA is BETTER than OSS....

Too many good people wail the tar out of it before it goes out the door.

Only NASA nakes better software.

And remember - the NSA knows math they don't teach at the universities....

Let's start the pool now....

[JUSTONEMORELATTE]

How long 'till:

• One of them gets detected with a pringles can across the street from an NSA office
• That same cantenna manages to sniff enough packets to crack the keys

My money is on Friday, November 22, 2002
--

Re:Let's start the pool now....

[sakeneko]

How long 'till:

• One of them gets detected with a pringles can across the street from an NSA office
• That same cantenna manages to sniff enough packets to crack the keys

My money is on Friday, November 22, 2002

D*mn, I thought the Kennedy conspiracy theorists had finally gone dormant....;>

If it isn't a Pringles can, it'll be some other low-tech, widely available object slightly modified by a bright teenage kid showing off for his buddies.

Re:Let's start the pool now....

[djrogers]

I'd like to see you get line-of-sight to NSA headquarters with a pringles cantenna... Heh, watch out for the men in the black uniforms ;-)

Re:Let's start the pool now....

[JUSTONEMORELATTE]

AC Wrote:

Misspelling the word "too" in your .sig is a capital offense.

Gack... The perils of cut-paste when your source is /.
Hey CowboyNeal, Taco, Anyone -- when you select "Freaks" from your homepage here, the cute message misspells "too"
--

Re:Let's start the pool now....

[JUSTONEMORELATTE]

You only posted this comment so you could use the "word" "cantenna."
Ok, you caught me. <grins>
Guess who's smarter than you. Yup, the NSA.
I have no doubt of that. And I'd wager huge sums that the person to crack the new encryption won't be me. (Mess with the NSA? That's the last thing I'd do! Yes, there are two ways to interpret that statement. Yes, both are true.)
But I also expect it will take about a week after the first network goes live before it's broken.
--

link may need changed?

[doc_side]

nsa.gov maybe instead? and not nsa.mil?

But it only works with Windows..........

[jcrb]

who is fooling who here? None of the OSes (only Windows versions) it works with are certified for TOP SECRET data.... guess its pretty useless till someone does the linux port eh? :)

Re:But it only works with Windows..........

[nihilist_1137]

Isn't there a NSA secure Linux distro?

Re:But it only works with Windows..........

[Zordak]

It's more like an NSA secure Linux kernel hack that adds some kernel level authentication layers. Also, their diclaimer states that it's more of a conceptual thing they were doing to prove out the concept, and that they don't guarantee it will actually work or anything.

Re:But it only works with Windows..........

[Hal_9000@!!!@]

None of the OSes (only Windows versions) it works with are certified for TOP SECRET data

Yeah, and Trusted Solaris, and Trusted Irix, and a bunch of other OSs you've probably never head of. Look at this if you don't believe me.

Re:But it only works with Windows..........

[Dragon213]

And if you would see the software that secures them for TOP SECRET data, you would be amazed...
That particular drive is not used for any other processing, nor is it removed for the secure COMSEC vault. It is coded and numbered, and is not used in any other computer. The computer itself has an encryption algorithm that I've never seen (not GOSH, BLOWFISH, or PGP algorithms) based upon a 1024-bit rotating key that not even the user knows. It is completly random (insofar as a computer can be random) and based upon a random seed. The user's login and password is also encrypted, and typically the computer is not connected to an ethernet network, but rather a dial-up connection through STU-3 or -4 secure modems.
And yes, the government uses Windows because of a licencing deal with the ever-pervasive MS.

Just something to think about....

Re:But it only works with Windows..........

[kir]

isolated probably means behind a firewall. not inside a bombshelter.

Isloted DOES NOT mean behind a firewall. All classified networks (networks processing SECRET, TS, etc.) are isloted in the sense that they're not, in any way, connected to the internet. And... if the computer is processing TOP SECRET information, it's more than likely in a bombshelter... well... not REALLY a bombshelter, but something similar, like a secure basement office or something.

Re:But it only works with Windows..........

[WatertonMan]

Unless they changed it recently, the above isn't true. We used both Macs and PCs in top secret areas. We used removable drives and most of the work to keep things "classified" were the trust of the employees. The really secure stuff were in a safe with no cables at all, but we had plenty of cabling actually running into quite a few areas.

Actually truth be told it never seemed that secure to me. Pro Force would actually let you into the buildings at night unaccompanied to do work. You then left on your own. Admittedly we were just doing analysis of nuclear explosions and weren't working on the main models or anything. Those were up the road. But it always freaked me out how little real security there was.

It always seemed odd that we had these rules about no non-optical cable beyond a certain length but the staff (including summer interns) was given free reign.

As most security lectures point out, the typical way a hacker will crack your network isn't a direct brute force way. Yet those more "primitive" approaches are what are typically left unguarded. (Although of course with recent WiFi networks not putting security on at all was always silly - but the government was much smarter than that)

speak for yourself

[tps12]

you and I won't be able to buy them

While you're correct that most citizens (including Slashdot editors, I'd guess!) won't be able to buy these babies, please remember that a large portion of Slashdot's readership is in IT, some of us in positions where we may, in fact, purchase equipment through an NSA COMSEC account. Industries and corporations deemed "essential to the National Security" under conditions set forth in the NPHG Protection Act have been given this priveledge since its passage in 1973, in response to the Viet-nam War. I work at a major corn distributor (food being an essential supply during potential siege or embargo, and breakfast being the most important meal of the day), and I can tell you that I hope to have my hands on these sometime this month, before Christmas or President's Day at the very most. It should speed up our processes considerably to not have to be tied to "wired" networks. It's a fun time to be in IT, and this cloak-and-dagger stuff just makes it better.

Re:speak for yourself

[treat]

It should speed up our processes considerably to not have to be tied to "wired" networks.

Then why didn't you just run ipsec over conventional 802.11? It will be just as secure as this, and can be done on commodity hardware and with free software.

Re:speak for yourself

[Anonymous Coward]

cloak-and-dagger stuff just makes it better

you use an algorithm designed by some freak at the NSA and suddenly you are cloak-and-dagger?

cloak-and-dagger is snowboarding down a mountain in front of an avalanche while helicopters fire rockets at you. get with the program here.

Re:speak for yourself

[treat]

No, the original poster is right on this one. IIRC, NSA regs prohibit such workarounds.

Do you really believe that his corn processing is DOD classified?

Re:speak for yourself

[treat]

During a national emergency or war, food production plants may need to keep aspects of their operation secret. Since some of these operations will be in some way coordinated with the government, they will use the security products sactioned by the government (that is the NSA).

I challenge you to produce evidence that the US federal government imposes information security requirements on food production plants.

Re:speak for yourself

[Bandman]

the identity imparing ski-goggles

Relax, my man

[ekrout]

joehoya writes "I realize this is a couple of days old..."

Hey, this is Slashdot, my friend! We post things two or even three times just to ensure that our faithful geek readership doesn't miss a beat on the latest and greatest technology news of the past 48 hours^H^H^H^H^H^H^H^Hweek^H^H^H^Hmonth^H^H^H^H^H year!

Re:Relax, my man

[bogie]

It's not flamebait its true. That and "teh suck" are totally lame and overused.

Possible Use for detecting detecting software?

[lpret]

In a recent article we discussed the futility of implementing a detector detector in a network. This seems that this would be one use that would actually help as an extra layer of defense.

Re:Possible Use for detecting detecting software?

[sakeneko]

In a recent article [slashdot.org] we discussed the futility of implementing a detector detector in a network. This seems that this would be one use that would actually help as an extra layer of defense.

Stratum8 Networks , perhaps? (Disclaimer -- I work there, so I'm not unbiased.) :)

why not in software?

[mocktor]

impressive stuff... from what the datasheet says this all looks to be implemented hardware on the card - but given the low-level facilities of the chipsets on consumer-grade 802.11 cards is there any reason why some bright coder can't do a similar thing in driverspace?

Re:why not in software?

[pVoid]

Remember, what can run, can be reverse engineered. Them making a software driver is an invitation for people to reverse engineer the stuff going on in the card.

Eventually, yes, a smart person will make a software version of this (that's the outcome of it all). But the reason they use hardware is to make life harder. Maybe even impossible (if enough effort were to go into the hw design).

In other news

[kcbrown]

MELBOURNE, Florida, November 4, 2002 -- Harris Corporation (NYSE:HRS) announced today that its Type 1 Encrypted, Secure Wireless Local Area Network (SWLAN) product, ClipperNet 11(TM), has been certified by the National Security Agency's (NSA) Commercial COMSEC Endorsement Program (CCEP). CLipperNet 11 is a revolutionary new product that enables civilian users to securely communicate multimedia information, including data, voice and video, via a wireless network at an unprecedented 11 Megabits-per-second (Mbps).

"ClipperNet 11 is an innovative new product that allows us to provide our civilian customers with the advantages of secure wireless communications," said an NSA spokesperson. "With Type 1 Encryption, NSANet 11 meets the Department of Defense's stringent requirements for wireless transmission of both classified and unclassified information."

When asked whether the product had any relationship with the Clipper chip proposal of the mid 1990's, the NSA declined to comment. "Er, emm ... we don't have any comment on that", said one NSA spokesperson, who was last seen leaving hastily.

"Don't worry", a Harris spokesperson said. "We would never even think of embedding any technology into our products that would make it possible for secret government agencies to read the encrypted data stream, and we would certainly never use any information gained in that way for marketing purposes. Trust us!"

Harris shares were up 2 3/4 points today.

Re:In other news

[filmnorthflorida]

It would be even funnier if it weren't so true.

Anyone want to guess?

[Neck_of_the_Woods]

Anyone want to guess what the WarChacking Symbol for this would end up being? Mabye a secure network symbol a barbed wire? Condom? gun? Handcuffs?

Re:Anyone want to guess?

[unicron]

You sure are uppity for someone that eats bugs all day.

Re:Anyone want to guess?

[karlm]

Anyone want to guess what the WarChacking Symbol for this would end up being? Mabye a secure network symbol a barbed wire? Condom? gun? Handcuffs?

As long as you don't use the condom and handcuffs symbol simultaneously, that means something completely different.

Secrecy

[Shadow Wrought]

From the press release:

...said an NSA spokesperson.

So even their spokespeople are unidentified?

Warfighter?

[Anonymous Coward]

"In the 21st Century, information is the warfighter's ultimate weapon, and for the first time ever, the U.S. Government has access to a wireless network solution that is Type 1 encrypted," said Bob Henry, president, Harris Government Communications Systems Division...

Warfighter? Holy Doubleplusgood Newspeak, Batman!

Because, you know, it's important to distinguish between the warfighters and the warsitontheirassesbitchingaboutcivillibertyers.

Or maybe this is some sort of subtle dig at wardrivers. "Those bandwidth-thievin' pinkos DRIVE around with their wireless rigs! Real men use new technology for FIGHTING!"

Interesting

[drhairston]

Given the common practice of 'WarDriving' that most young people seem to be 'in to' these days, it is probable - nay, inevitable, that these wireless points will be detected by someone and posted at a site such as cryptome. The interesting question, of course, is whether the publishing of data about the presence and location of these acccess points will be considered illegal, and whether the 'War Drivers' will be arrested for detecting the signal.

One would hope so, but you never know.

Do it at higher level anyway

[Goonie]

Better still, don't bother with encryption at the hardware or driver level at all - do it at the application level where the algorithm can be changed without too much hassle if it is discovered to be insecure.

Sounds like...

[sheWhoWalksWithToesL]

Security via obscurity. I wonder how long THAT will last.

expensive network kit

[Indy1]

Ouch, i just found the price list. This stuff is $$$$$$$$$$$$$.

The pc card's are $2500.

Wap's are $1000.

I think I'll stick to VPN over 802.11

Source of pricing is www.govcomm.harris.com/secure-comm/support/priceli st.html

...it is...

[Eric_Cartman_South_P]

http://www.gnupg.org/

;)

Ok, so what can WE use....

[Eric_Cartman_South_P]

Listen, this is ON topic, and just 'cause I'm not 733t in all areas doesn't mean I (other readers) are not interested! Help me! What would someone using Linux or Windoze use to get simialr security. With 802.11, what can I run on my box so my networking, web browsing, etc, is secure (as can be) between me and the AP. Ideally, we're talking open source solutions here. I'm not talking about securing my e-mail, but making things secure between me and the AP so I can have get an IP on my laptop, and use the network "normally" with encryption in the background over the air. Anyone? Anyone?

Re:Ok, so what can WE use....

[afidel]

Cisco wireless gear using LEAP is probably the closest to this setup. Of course for really paranoid customers they still suggest using IPSEC on top of LEAP. Their cards have drivers for Windows (all versions from 95 on), linux, and mac's.

Re:Ok, so what can WE use....

[Eric_Cartman_South_P]

Thanks!

I would be *extremely* surprised

[myowntrueself]

if the NSA approved of something that they didn't *KNOW* they could hack into...

There's that secure wireless oxymoron again

[kbielefe]

At my work we deal with a lot of secret and/or export controlled material. There are areas at my work set aside for foreign customers that we aren't even allowed to run a wired LAN connection to. If you want to run some software over there, you have to put it on a floppy or CD and carry it over from your desk. This can be a real pain when trying to find an elusive bug. Maybe it was just easier than getting the security measures approved to connect the LAN.

If they have good reason at all to be that paranoid about a wired LAN, I think it won't take long for this "secure" wireless thing to come back and bite the NSA.

MAIN NSA COMSEC ACCOUNT

[Istealmymusic]

The main NSA COMSEC Account is 880099, and its address follows:

Middle River Facility
Building A-W Dock 2
2800 Eastern Boulevard
Middle River, MD 21220

PCMCIA still good??

[myowntrueself]

Isn't it about time that PCMCIA were replaced so that people have to buy new laptops etc?

(I imagine it wont be long before you won't be able to buy a MB with PCI; VLB started out as a purely graphics bus (VESA local bus) and it wasn't long before it was used for SCSI, Multi IO and probably others. Were there ever VLB NICs?

With this history it is a little surprising that manufacturers arn't producing multi-AGP boards and SCSI cards etc on AGP, eventually replacing PCI.

I know its not an exact match, and maybe theres something about the AGP standard that makes this impossible, but you get the picture;

Market saturation requires forced obsolescence and upgrade fever to achieve constant economic growth. Any stability spells doom for the market for some reason; its a self destabilising system. Any trends of economic stability as opposed to economic growth causes instability and either growth or shrinkage, thereby producing instability again.
I dunno about the commas in those sentences. Feel free to rearrange them to taste.

Re:PCMCIA still good??

[afidel]

PC Card which is physically identical to PCMCIA is what laptops actually have these days. PC Card -> PCI as PCMCIA -> ISA.

Re:PCMCIA still good??

[man_ls]

One reason that there are no other devices than graphics cards for AGP is that the data throughput rates are highly asymmetrical.

We're talking gigabits/sec of "push" bandwidth (textures, etc. to the graphics card) and sometimes as few as megabytes or even kilobytes of "pull" data, reading back from the AGP card. It wasn't designed with sending data back, it was designed to take data as fast as the host system can throw it off and process it.

Additionally, a dual-AGP controller wouldn't work because it is connected too directly to the CPU and Northbridge. NICs branch off the PCI bus, attached to the Northbridge; AGP has its own lines on the Northbridge. More AGP slots, the Northbridge is going to run into bandwidth problems, and it is frequently already the bottleneck in the highest of performance gaming machines.

NSA Press Release

[Newer Guy]

For Immediate Release: NSA to use Navajo "Code Talkers" for 802.11b encryption. 11/12/2002 The National Security Agency ("NSA") of the United States announced that effective immeidately they would be using 'code talking' technology based on the language of the Navajo Nation to encrypt all their 802.11b links. "We feel that this is an approriate encryption for these type of links" says Hugh G. Peter Head of NSA Encryption. "Besides, it will put many unemployed Native Americans back to work". The move was immediately commended by Microsoft Corp., who pledged to use this radical new encryption system in all it's new wireless products.

This is great!

[LittleLebowskiUrbanA]

My unit sets up networks in the field (I'm a Marine) and most of the work involves running a fiber backbone and running CAT5 to each and every computer in the fieldHQ (tents). We had looked into running wireless but of course the security was non-existent. Maybe now, we can spend more time training the junior Marines on real networking, not running a CAT5 drops to some officer who "has" to check his email. My platoon will be looking into this tomorrow, I can assure you.

We're gonna get it now...

[TerryAtWork]

I just checked the NSA web site and it's /.'d ...

What about system accreditation?

[jinx90277]

I work in the defense industry, so I have to deal with security issues on occasion. Even though they got someone to sign off on the security of the wireless transmission, it will be interesting to see how they actually implement this technology as part of a larger accreditable system.

In my experience with security organizations, they tend to overemphasize the role of physical safeguards in designating a system as "secure," especially when it comes to COMSEC. How will they feel about accrediting a system in which multiple COMSEC units can be moved outside of a secured perimeter?

So what? Even the phone book is classified!

[pvera]

This is a non-event. And secret is not a life-or-death classification level, as anything that is considered remotely important will automagically get tagged with TS + keyword.

Most readers missing the point...

[drunkrussian]

To get something approved for processing at the SECRET level is a moderately big deal for those who work with such data. For the outside world, it's not the last word on the quality of the system.

You can't, for example, get a Linux box approved to process SECRET information (at least, last I checked). Windows is approved, however. Yet, for the commercial user, I would say that Linux is more secure than Windows. What matters is how the system is set up. I'm kind of surprised that there's any demand for wireless networking at the SECRET level. With few exceptions, a classified box has to be physically disconnected from all other machines and operate only from hard drives with no communications software on them. There was an article on cnn.com today about a hacker who got access to sensitive but not classified information on military networks. The reason he didn't get access to classified information is because of the way it's protected.

And forget about anything at the TOP SECRET level or above. We have a room at the office that does work at the TS level. If you bring a disk in there, you can't leave with it. If you bring a hard drive in there, it can't leave the room. Once a computer goes in there, it can't leave either. Well, that's not entirely true...security chops them up into little tiny pieces, waves magnets over them, and does some other magic to make them completely clean before they can leave. They're certainly never useable again. They even destroy the monitors before removing them from the room, in case an image might be burned into them.

Anyway. People who deal with SECRET information will probably be interested in this article, and I'm sure life will go on with no change for those who don't.

Re:it's not worth the money b/c...

[afidel]

Other than the insane price for the PC Card (maybe this is a large pack, say enough for a small office?) These are in line with the list prices for Cisco's stuff not too long ago (they have since come down some). Unless you are talking about SE Linux then your first comment are totally off. As far as the NSA is concerned there are only a few OS's this should run on, SE Linux, Trusted IRIX, Trusted Solaris and possibly a few other trusted UNIX variants that I am not aware of.

not inevitable

[crucini]

I don't think it's inevitable. 802.11 is basically spread spectrum, right? This system probably keys the SS from a totally different algorithm. So it will probably look like widespread RF noise. Unless they chose to preserve some channel discovery mechanism from civilian 802.11.

Pst... it's sooo secret...

[Cheese Cracker]

one of the nicest things you can hope for when trying to gain access to a server is what software and version its running.

Hmm... most crackers use Netcraft to see what sites like the NSA website uses...

Re:Pst... it's sooo secret...

[jareds]

And I'm sure the NSA's public web server isn't physically separate from any networks used for classified data.

Re:it's not worth the money b/c...

[afidel]

Hmm Solaris will run and be supported on x86 in the next iteration. It will no longer be a free download but you will be able to get it. I guess Solaris based on the hehawing Sun gave for Solaris 9 x86 support and This page

I hope they did their homework.

[small_dick]

I'd hate to see people getting killed or the economy damaged just so the NSA can claim to be "hip and with it" on the wireless side of things.

SECRET information might not be as dangerous a loss as the higher level stuff, still, going wireless, to me anyway, means broadcasting...and sniffing...and recording...and analyzing...lots different from copper or fiber in a shielded, isolated, locked down environment.

Sure!

[LinuxGeek]

First fill out forms:

GSA-1132-4
GOV-3321-11-23
MI-33241-A
FL-31-S4 T

Then provide your security classification and reason for needing the information and it will be provided via secure channels if approved.

I can say that I know people that work for several organisations that produce and grow huge amounts of food and the larger companies do keep different government departments informed about production and supply chain problems. That said, I really assume if you needed to know the information you challenged the AC to provide for you, then you would already have it.

AES is coming to WLAN

[hedley]

Don't worry the IEEE is voting on 2 competing methods for WLAN encryption: OCB (offset code book) and CCM (I don't remember the acronym break down). CCM will most likely win since OCB wants ~100K$ per company implementing it in firmware. Both approaches use AES with 128bit key and 128bit block data. This is a big change since WEP (wired "equivalency" protocol) uses a stream cipher, RC4. There is no IV any longer that is sent with a monotonically increasing #, instead the "nonce" as it is called is AES encrypted.

Each of these methods rely on the fact that you won't be able to reverse a known packet back to its plaintext. (relying on the fact that AES is not easily reversable).

That article was complete marketing speak too. "11mbs!" the effective rate of a WLAN these days is maximally in the high 6's or 7's if you all use short preamble. With long preamble, the effective rate is in the 5's to 6's.

Hedley

Some highlights

[mdecerbo]

I googled around. The site at www.secnet11.com is actually pretty informative, and there's some other information floating around out there too.

Some highlights:

• The card sticks out of the computer with two antennas poking up.
• It uses an NSA encryption algorithm called BATON (from various stuff on the Web, I get the impression that BATON is a 64-bit block cipher with 128-bit keys that is designed for very fast operation)
• the message address is encrypted to prevent traffic analysis (this is a big selling point against VPN technology)
• Each packet has an 80-bit IV (it's rare to learn even that much about a Type 1 encryption system)
• Cards cost over $2500 each. That's 30 times the price of a commercial WiFi card, but cheaper than traditional NSA encryption data products which seem to run around $5K per node.
• "Red keys" are loaded via a special cable that connects to a data transfer device such as the CYZ-10.

I wonder how much work it would be for someone to implement a commercial version of this using Rijndael, or AES, or something unclassified. With a larger market than the government, maybe it could be cheaper, and the development costs made up on volume...

Let's face it, it's a pain to set up IPSEC on all your boxes...

Re:Some highlights

[karlm]

I wonder how much work it would be for someone to implement a commercial version of this using Rijndael, or AES, or something unclassified. With a larger market than the government, maybe it could be cheaper, and the development costs made up on volume...

First of all AES is a subset of Rijndael. (Rijndael can be defined with multiple block sizes, AES is Rijndael with a 128-bit block).

Second, the IEEE is working on a WiFi standard using AES.

Third, it appears that AES is no stronger than 100-bit effective key size due to the XSL attack. (Serpent is reduced to a maximum of 200-bit effective key size. Twofish and RC6 appear to be safe against XSL.) I'd go with Twofish or RC6, but everyone is going to use AES b/c it's the NIST standard, despite it's weaknesses.

When the NIST decided on Rijndael, everyone knew it ran the fastest and was probably the weakest, and now we're stuck with 100-bit (maybe even weaker) equivalent keys.

Clarification on military networks

[Boka Dominigo]

I speak only for myself, not as an official representative of the U.S. Government.

I decided to write this because I often see misconceptions of military networks on slashdot.

I have been a network administrator in the U.S. Air Force for 5 years. I have administered classified networks in Asia, Europe, the Middle East and the U.S. I have worked on Air Force and Army networks.

(1) The basic levels of classification are:
Unclassified
Confidential
Secret
Top Secret

There's some gray areas between and above but those are the basics

(2) You can process classified information on almost any platform you want. Top Secret on DOS, no problem. Windows 95, every day. Linux, sure. The big restrictions come when a computer is connected to both classified and unclassified networks. In that case the machine must be trusted to differentiate between the classifications. It must make sure that only Unclass was writted to the disk you're going to carry over to the unclassified network.

(3) Classified information, once properly encrypted, is no longer classified and you can pretty much do you what you want with it (put it on your t-shirt, print it on a flag and wave it, blast it in to space, send it over the internet, whatever)

(4) Because of the above, wireless and classified are nothing new. Radios, wireless networks, satellite phones, all of the them are used to transmit classified information.

(5) Moving classified information over unclassified networks is old news and several devices already exist. Devices like the NES (Network Encryption System) and the TACLANE are used to plug in to a classified network, encrypt and encapsulate the data, then move that data over an unclassified network.

http://www.fas.org/irp/program/security/_work/kg -1 75.html

(6) What this new device offers is conveniance. Previously to run a network over a wireless link the procedure went something like:

Connect computer/network to DTE/DCE device
Connect DTE/DCE device to crypto
Connect crypto to wireless transmission medium

These steps needed to be completed for both sides of each link. It is slow, complicated, and expensive.

(7) Why not use IPSEC? It's complicated and not NSA certified. You should be able to give crypto to a user and only explain three things to them; in, out, power. Nothing to misconfigure, either it works or it doesn't, no chance of classified spillage.

(8) Why doesn't someone with access just take this thing apart and figure out whatever? This product is likely a CCI (controlled cryptographic item). Opening CCI without certification/authorization is illegal. Besides, without disecting the chips, how much are you really going to learn?

(9) The NSA must have a back door built in, right? No. A back door built in for them would be vulnerable to anybody. I highly doubt we would move national security information over a wireless network with a back door. If you're using their encryption keys, they have a copy and can read the info anyway. If you're not using their encryption keys, then you don't have one of these devices.

(10) Isn't someone going to crack this in a week? No. NSA certified encryption is good and well tested. We still routinely send Top Secret information over 10 year old encryption devices. If they had been compromised, we wouldn't be using them. The information sent from this device is encrypted. Without the same encryption key, you can't communicate with the device. Period.

(11) What about sniffing packets and breaking the key? Go ahead and try. Encrypted information has been floating around in the air for years and years. Multimillion man armies have been sniffing and recording and trying to break for decades. They keys change often. Sure, someone might (if they were lucky) break one key in ten years, but many devices get a new key every day.

I'm sure I left some stuff out and there are faults in my knowledge and spelling. If you have any questions, post and I will try to answer them.

Point to Point or Multipoint-Capable

[Effugas]

My primary question about this system is whether it creates a secure domain, within which everything may be sniffed and trusted, or whether communications follow the principle of minimum exposure, i.e. a connection between hosts A and B cannot be read by hosts C and D, while a broadcast packet from A can be read by B, C, and D but not by anyone else.

It's much easier to create a shared domain than it is to create a dynamic key mesh (presuming there's no pubkey stuff at work, and even then things get tricky). One shortcut is simply to provide keys to the upstream router, and let the router sniff all traffic (and experience the cost of routing traffic between endpoints). My bet is that this is what's done.

Anyone know?

--Dan
www.doxpara.com

Correction: Linear cryptanalysis

[karlm]

The s-boxes in DES are maximally resistant to differential cryptanalysis. However, they are weak against linear cryptanalysis..