CWNP CWSP – Module 03 – Encryption Ciphers and Methods Part 2

4. WLAN Encryption Methods

Now let’s take this into the world of the wireless local area network, the WLAN. So remember that 800 and 211 is defined really for layer two of the OSI model. And so what we’re seeing here is that we’re still going to go through a lot of encryption. Not once did I say anything about having to have an IP address. I really didn’t, because we’re really talking right now about encryption between that laptop and its connection to our wireless access point. Here, I’ll put a third antenna. So now it’s an 800 and 211 N. Okay, that was a little bit of wireless jokes for those of you out there. Sad thing with the video is I can’t tell how hard you’re all laughing at my jokes. So we’re doing the encryption here through the communications, which is basically our radio frequency.

And so we’re talking about doing this at layer two. And the technical name that they use for this is the Mac Protocol Data Unit or the Mpdu. So anyway, the technical name. So encryption is going to make the ciphertext portion, what we call the MSDU or the Mac Service Data Unit and encapsulate it with the Mac address and the frame check sequence. And I think I talked about that before. Now, notice, of course, there is still a physical layer and a Mac layer. And what we just focused on here was the Mac layer, which, again, if we think about it, the physical layer information, the ones and zeros and everything that we’re sending is what layer one is all about.

Layer two is what we’re interpreting as the ones and zeros that were being sent out. So even though they might call us the physical layer SDU, the Psdu or Data unit, it is the physical portion, the actual what we’re sending over radio frequency, all the ones and zeros and how it’s formatted and what it means to us. It has a preamble start, a framed limiter, has a physical header. And all of these are just describing what’s that layer one for the back and forth communications so that we can recognize it from maybe garbage that’s being sent from some other type of interference.

So we take that. And basically, like I said, what we’re going to do is we’re going to take that Mac layer SDU, again, as I said, the service data unit and put in it in front of it unencrypted. So remember, this is the encrypted portion here, but we still need to have the address fields because right, the actual Mac addresses that the access point and the client use to identify themselves.

And we still have the Frank correction sequence, something that, again, sometimes you saw already referred to as the cyclic redundancy check. I don’t care what you call it, but it’s the way of making sure that there was no interference or change of information. And that’s what we’re sending. And so anybody can interpret this portion that has the ability to read or listen with the radio and that’s unencrypted information. And as I said, the rest the real meat of what’s being sent is that MSDU.

5. WEP Again

Now, just as a reminder when it came to Web that it used the RC Four algorithm, again, that was a block cipher. It was still considered a very good algorithm, but we don’t suggest that we use it now because we have better ones with better keys. So it was considered, again, a legacy encryption system. And the strongest thing I can say is it’s not suggested for use. Unless you just really want to say, oh, I turned on encryption, but really don’t care if it’s safe. Remember what it did is it actually used with WEP 40, a 40 bit key and a 24 bit initialization vector as the keystream. With WEP 104, it was 104 bit plus the 24 bit initialization vector to give you a little bit longer key. But they still use the same algorithm, RC Four, for the encryption and decryption. And everything we talked about when we were talking, talking about the weaknesses of WEP at a different part of a course are still there. Right. It’s just a reminder that this is where we came from and why it’s better to go with stronger algorithms that have stronger keys.

6. TKIP Part1

One of the initial improvements we saw is TKIP the temporal key integrity protocol. As you can see here, it’s designed to replace WEP. And really it was important that we had something, especially after WEP was so easily broken, and it was something that we were using before we came up with some other standards with the 800 and 211. See, I can write it right, but not say it right, but you got it. 800 and 211 I. And so it was kind of an intermediary, kind of a temporary solution, but it’s not a bad solution because we’re actually going to see that with many of our encryption standards today, that TKIP is designed to be an extra addition or an option that you can use to be able to even further enhance the type of information or type of encryption that you have. And so what it did for the short run is it allowed us to replace Web without having to replace equipment. Usually it was just a firmware upgrade that you had to do to be able to have that temporary solution. And like I said, don’t forget it eventually is going right here, optional. Now, as a solution with WPA or WPA two, I should say, how it differs. What was different about it than WEP? What made it different? Well, number one, temporal keys. Now, if you ask me, that sounds like something right out of Star Trek with talking about temporal portals. Okay, I won’t get too geeky. Here’s the big thing. It was a dynamically created key. It wasn’t a static key. And that was one of the problems with Web, is that it used a static key.

And so after enough exchanges were captured with the initialization vectors, we could eventually glean what the actual static key was. But if in this case, I’m using a different key every time I’m making a new connection and then adding on those 24 bits of IVs, you’re not going to be able to do your replays and figure out what the keys are, because we’re doing it different each time. Now, what it does is it uses what we call a four way handshake to create these dynamic unicast keys. And those are going to be two keys or keys that are created that are unique between two stations or two options or two people with radios. So that would basically be your client and the access point.

And so if I had another person connect, they’re going to come up with a different temporal key. So I would have TKIP number one. But I’m not going to reuse it like I would with Web because these two are going to come up with a different key. That means there’s going to be in a security association. That means the access point is going to store all these keys and associate them with these clients that have their own Mac addresses. So, again, we’re making it a lot easier to keep more security and making it harder for people to be able to guess what the keys are.

So that was an important one. The other problem with the static key, too, is if it was static sometimes, let’s say I was a social engineer or a hacker, I might call up or talk to somebody sitting next to me who’s connected to the access point, and I can say, oh, I just had to reset my laptop. And so I lost the key. Can you tell me what your key is?

And if I do it right, they usually would tell me their keys. But if it’s a dynamic one, then it doesn’t matter. They can give me their key all they want, which maybe I can use to decrypt their stuff, but they actually won’t know it because it’s dynamically set up. It’s not stored in settings like it is in Windows with a static key. It also added sequencing. So sequencing basically is making sure your packets are coming in order. And remember, one of the weaknesses I said with Web was that I could capture the initial exchange between the two points and then just replay that attack to get more IVs and replay it to get more IVs. And then basically I can figure out what the IV was versus the password.

But to replay means that now my packets are coming out of order. So we wanted to defeat these replay attacks by using the sequencing. TKIP, I should say, uses what they call a TKIP sequence counter. I’m not going to spell out the whole world word. And that TKIP sequence counter called the TSC is where we can basically drop any of those packets that are being reinjected into us. So again, one of the weaknesses with Wet was the sequencing or the lack of sequencing. Another strength with TKIP is what they call key mixing. So key mixing is well, I like the way they say it, if you look at the general definition, a complex two phase mixing process to create stronger seeding material.

And so basically what we’re trying to say is the stronger seating material is a stronger initialization vector IV. Again, we’re trying to stop, like the IV collisions or even the weak key attacks that came against us. It used what we called the enhanced data Integrity. Often called some people call it Michael, some people call it Mick. It stands for the Message Integrity Code. Or some people might call it the message integrity check. Don’t you like it when we can’t all come up with the same definition of an acronym? And its job through Mick was to defeat the bit flipping and forgeries. Remember I said that with the WEP integrity that it might be you could change 10 to a one and another one to a zero, and the message looks like it’s right, but it’s not. It comes out as a forgery. So it’s a better enhancement to the integrity. And TKIP countermeasures.

All right, so it is, unfortunately, still easy for somebody to do a type of an attack. One might be like attacking Michael the mic. Here’s kind of the idea. If I can’t steal your data, then I try to make sure nobody else can get your data. So I might try what’s called a denial of service attack. And so if I start injecting packets to make the mic or the mic look wrong, then it’s going to drop all your packets. It’s just something we have to worry about. But that was one of the design constraints. So in order to fix that, we had to come up with some other types of countermeasures to stop or defeat the chance of a forgery or even a denial of service attack.