CWNP CWNA – WLAN Topologies Part 2

12. ESS Part1

Now, an extended service set is where you have two or more basic service sets connected by what they call a distribution system medium. That’s just a nice way to say that your company has two or more access points that are connected to your wired network. And that wired network would be the distribution system medium or the DSM. Now, usually an extended service set is really just two two or more access points and their associated client stations all united by that same DSM.

The most common of that might be one where you have multiple access points with maybe or maybe not partially overlapping coverage cells. Now, unless you do some extra work on the management side, what you’re going to find happening is that people will lose their connection when they go from one one access point to another one and have to reestablish their connection. And if they had anything else running on the network, that, of course, would fail.

13. ESS Part2

So when we look at this, this is an example of having two access points with overlapping coverage, so that as somebody who might be on the WiFi begins to move, they’ll still have coverage, but at some point, the signal strength from, let’s say, access point one is going to become weak and access . 2 will become strong. And so they’ll basically switch, switch over. Now, in this case, they almost always have the same SSIDs.

So that if you have an operating system that says always connect or automatically connect in the reconnect. The problem is, in this particular situation, the handoff is going to cause this access point, which might be on one IP subnet, then, although I’ll have to go to a new one that’s on different IP subnet. And as I said, even though they have constant coverage, it doesn’t mean that they have constant connectivity. They’ll have to free-associate with the next access point.

14. Roaming Part1

Now, when we are thinking of the idea of roaming, we cannot do it with an extended service set unless we have something else running for us to be able to make that connectivity happen. And of course, you also need to have overlapping coverage. So to get seamless roaming, which is in today’s world, a key aspect of wireless land design, then there is no requirement that an extended service set is going to do that unless you have something else in between. Usually, and we’ll describe this later, something like a wireless LAN controller that can help you with the handoff.

And again, a handoff is just what it sounds like. You’re going to go from one access point and as you move to the range of the other access point, you can connect to it. And because of the central management, it would be able to re that all you did was switch from one access point to another and everything else will be constant. But to get a really truly roaming experience, those access points do need to have that overlapping coverage for you.

15. Roaming Part2

So the idea that I’ve expressed so far without that central management is something called colocation. And Colocation is just that. It just means we have the ability to be located on one access point or located on another access point. So, again, as you roam from one place to the other, I’m going to draw a laptop because I know how to draw those.

You’re still connected to the network. It’s just a lot of your settings are going to be gone. You may have to reassociate with the new access point, which also takes time. So basically, if you were, let’s say, like this picture of somebody holding a wireless WiFi phone voice Over IP they’re going to lose their phone call while they’re going through that change.

16. Roaming Part3

Now, the DSM, as I said, is a way in which all of these access points are connected to the same wired network. It’s almost always going to be an 802 three Ethernet backbone. I say that because not very many companies today are running things like Token ring. This is not going to be a wide area network connection because if it was, you’d have to have one super access point to cover thousands, thousands of miles.

But in order to make it work, and with that overlapping coverage so that you don’t have any complexities with your communication, it’s important that even though you have the same SSID with each one of these, you have to make sure they’re on different channels, so that in the middle, where the overlapping occurs, that you’re not having competing signals. If you had competing signals, not only do we have the problem of reassociating, but we also have the problem of having less bandwidth because there are two competing signals in the same area.

17. Mesh BSS Part1

Another type of topology that you can have is what we call the mesh basic service set. So it’s kind of a hybrid of what you might think was an extended service set by having more than one access point. In reality, all those extra access points are handing off your attempt to associate with the network to just a single kind of what we’d call a home based access point. But the reason for all the other access points is to extend the availability of wireless. In other words, if you had, like, a large park or a large outdoor area at your office and you wanted to make sure everybody had connectivity, you can put access points around at different locations where they are actually.

Connected back to the access point that is connected to the wired network so that you can have coverage, but you still technically have 510 15 access points, all representing a single access point. So that’s why we call it a basic server set, but it’s in a mesh. Now, one of the problems we have with that is that your traffic could start going in loops. If you had a lot of access points all over that outdoor area, how do you know your traffic is just not going in circles instead of going to the actual location where the real access point is? And so they have a protocol that’s called the Hwmp protocol.

It’s proactive and reactive and it basically kind of acts like a layer two routing protocol to make sure that the path you take from the access point you’re connected to to the legitimate access point. I say legitimate, meaning it’s connected to the wired network is taking not only the shortest path, but but a loop free path.

18. Mesh BSS Part2

So as an example here, I have in fact, in this case, I actually have two access points that are connected to the wired network. But I might be out here in the park grounds or whatever it is that you’re complex sitting outside, and I want to connect. So we have these other access points that maybe I can associate with the one that’s closest and has a stronger signal to me. And what his job is to do is to basically route my traffic traffic to the access point that is connected to the wired network. And the goal, like I said, is to take the shortest path. So they do all communicate with this Hwmp protocol. So let’s say if I was over here, the question would be, when I connect to this access point at the bottom, am I going to go to this direction, or am I going to take this direction? And whichever direction I take, what’s the next top after that? When I talk about a loop, I’d hate for it to send me to this one that sends me to this one and sends me back here. Then I don’t get anywhere. I don’t get to the wired network. And so that protocol is designed is to find what looks like the shortest path to get to the actual access point that is connected again to usually that distribution medium, the DSM, which is my Ethernet or 802 three type of network.

19. WLC Roaming (eNotes)

Now, at some point there has to be a distribution service. The distribution service is how we are able to get the data that comes into the access point sent basically to the eventual destination, whether it’s on a basic service set or even an extended service set. And often we call the medium that we use for that distribution the DSM or the distribution system medium for almost every single one of our deployments. That medium is going to be Ethernet.

20. Access Point Modes

Now your access point can act in many modes. One of them is the bridge mode. Bridge mode really just means that I have an access point, let’s say in one building, an access point in another building and their job is just to talk to each other so that everything happening in the local area network at one building can talk to everybody at the other local area network. It’s not really designed to have clients connect to those access points. We’re just trying to span a great distance between the buildings to be able to have connectivity without stringing a cable.

Now to do it, there are going to be some extra Mac layer intelligence and what that basically comes down to if you think about Ethernet frames, if I were to draw an Ethernet frame, it has the preamble and then it has the destination Mac and it has a source Mac. I forgot to start a frame on there, but it doesn’t matter, has the type of frame format and then it has all this other stuff that’s not Ethernet. Well, destination and source Mac are great for a wired network and a switch to be able to figure out where to send things. But we’re actually going to add two more Macs and those are the Macs of the transmitter and the receiver. So you actually have a frame that kind of looks like Ethernet, but it has four Mac addresses so that we can identify the transmitters and the receivers as well as the actual computer that’s sending the traffic and the computer that it’s supposed to go to. Now on the workgroup bridge mode, which basically is transformed into a workgroup bridge, it gives you a wireless backhaul for your connected 802 three clients, which is kind of what we’re used to right here. I’ll have clients connecting to that access point and that will hopefully be connected into a wired network repeater mode. Kind of what we saw in the mesh network is where the access point just extends coverage of the real access point connected to the network. So you might have seen that picture already where I might have the local area network switch that’s wired and connected to an access point, but that access point is going to connect to another one, maybe to a couple of them. So that as you get further away from the main access point, all you have to do is be able to connect to one of them and it will forward or repeat your connection to the other access point to get you into the wired network. And what I began setting up was what eventually will become the mesh mode. And the mesh mode is where you really have a large area that you want to cover following kind of the same mode of the repeater mode. And then we have some that are in scanner mode. Somebody might also call it monitor mode. That means you might have an access point. And I’ve seen this in a lot of even like military installations where up on the ceiling I might see two access points within feet of each other. And at first glance you might say, well, why do I need them to be 4ft apart or 3ft apart? It shouldn’t need that much coverage. Well, that’s because one of them is an access point where all the computers connect to that are in that building and the other one is basically acting as a spy. It’s going to be looking in its area for other access points that don’t belong there. So it can report back to a management system like a wireless Lang controller that it might have found a fake or a rogue access point. Now that mode, as I said, scanner or monitor mode is a part of what a lot of vendors call the wireless intrusion detection system. Now it might, as a spy, see an access point that belongs to another company that’s in the office next door. And that doesn’t mean that it’s doing anything bad, it just means it was able to see it. But it’s really bad if it’s somebody who creates an access point and tries to mimic your network to steal your traffic. So hoping that computers would connect to it instead of the real one. And then we can actually, depending again on the vendor, ask that spy to begin to do a delivery denial of service attack against that rogue to make sure everybody drops off of there and connects to the real access point.

21. Client Station Modes Part1

Now, as far as the client, the computer that is connecting the access point, there are two modes typically that they can use. The one that is used 99 point however many percentage of the time is the infrastructure mode that means I have a computer that is wirelessly connecting to an access point. But, you know, if you have two computers that want to talk to each other or a tablet that wants to talk to a computer or whatever, whatever the case may be, there may not be a wireless network that you can use. And so they can go into what we call the independent service set or the ad hoc mode. That simply means that they can use their wireless cards between each other to be able to make a connection and to exchange data.

22. Client Station Modes Part2

So the second one I just mentioned was the ad hoc mode. Some people might call it peer to peer mode. The thing we have to worry about, though, is an access point has intelligence built into it to prevent things that we might call collisions or crashes where we have interference. But nonetheless, that’s just two systems talking to each other over wireless connectivity to be able to have a connection. Maybe they want to share a file. Maybe there’s playing some sort of online game, whatever the case may be. But that is a part of the independent basic service set, which ad hoc mode. I like that better. But the IBSS and that’s basically what they do. The problem, like I said, though, is there’s nobody in there to control if I’m sending traffic at the same time this one’s sending traffic, and then we have these big collisions into radio waves. And so it’s not as an effective means of communication.

23. Section Review

So in this module we talked about the wireless networking topologies what we can see out of the eight or 211 topologies and what kind of configuration modes we can put each of the access points into.