CWNP CWNA – Troubleshooting Part 2

11. 802.11 Coverage Considerations Part3

One of the recommended practices is that we don’t allow the access point to accept the slower speeds for connectivity. And that’s an important thing when we think about how to design your at least the 2. 4 ghz network. So by disabling those, you then don’t have this thing called the sticky client roaming problem. Sticky client look, here I am in the 1 minute megabit per second range and maybe there’s another access point over here that I’m closer to and I can get better coverage with if I were just to switch over, which would be roaming. But I’m being sticky.

I’m not going to give up on my favorite access point to connect to one that gives me better connectivity. But if we said, hey, one and two megabits per second, that’s just too slow, we’re not going to support it, well then you don’t have much choice. That also caused the hidden node problem that I just talked about and we don’t have to now contend for the medium contention. Meaning that I’ve got these people who are really slow but still trying to take their turn at the access point which by the way, you get only a certain amount of time, not a certain amount of bandwidth for your airtime fairness but nonetheless, you’re taking away airtime capability from those who could get a lot more done. 

12. 802.11 Coverage Considerations Part4

Other coverage considerations. Is this big thing roaming? If you talk to me about roaming ten years ago, I would have been thinking, I’m paying extra money on my cell phone. There was a day we actually had paid for that, but now we want this in our wireless networks and we say, okay, look, I want to be able to have my tablet, my byod online everywhere I go in the office, us, whatever floor I’m on. If I go over to the cafeteria, if I’m sitting outside at the park bench, I mean, I want it on. And so roaming is a method in which a client can basically have a handoff from one access point to the other, hopefully without losing any connections that they had to running applications. I mean, remember, somewhere in here is a wired network and actually let’s change what that is. And that wired network might be going through a router.

And if we’re not careful how we design this, existing applications might fail because you’re going from one subnet to another. But nonetheless, when roaming occurs, at some point, basically when the signal gets too weak from the original access point and the new access point with the same ssid, by the way, they have to have the same ssid for this to work gets stronger, then we’re going to send that reassociation request so we don’t lose our coverage. We do that with our cell phones, right? We go from one tower to the other as we’re driving down the highway, the freeway, I mean, wherever you’re at. So our goal is to have seamless communications for client stations that move between the coverage zones within what we would call the extended service set. Extended Service set just basically means that you have two or more access points using that same ssid for the purpose of extending the coverage area.

13. 802.11 Coverage Considerations Part5

Now to have seamless roaming. I’m going to tell you one more time, I’m going to write it down here. I’m going to keep saying it through this entire module. You need to have a site survey. I mean, to be able to know that you have seamless roaming is to have a proper design. Oh, I don’t even have to write it down. And a site survey, unfortunately, you might say, well, here’s my theoretical coverage, right? I know if my access point is here. I looked at the polar charts. I know what it says the signal strength will be. And we can figure it out based on power and distance. And I can say it just really looks good. But maybe you forgot that there’s a concrete wall right here as a part of the building that you’re in and you’re trying to cover. What does that do to your coverage? Or you’ve got a big kitchen area with a microwave oven over in the middle of this thing. I mean, what does that do to your theory?

So even though we try to get 15% to 30% overlap, the reason you do a site survey where you actually use you should use a real blueprint of your office. You should put the actual objects that are in there because most of the software is going to be able to say hey, this looks in theory, good, but look at what the real coverage is, right? It doesn’t look as nice where you might have thought there should have been some coverage here and some coverage here. Who knows what’s causing the interference?

And so this gives you a better plan on knowing what your coverage looks like and you could figure this out even if the software doesn’t do it for you, you could sit there and walk around. I used to do this both for security and for coverage and I’m going to tell you how long ago this was. I was using what they called a pda. I don’t know if any of you remember that. It might have looked like a tablet, but it certainly wasn’t smart.

But I had a wireless card, it had a linux distort for the operating system and I ran kismet and I would walk around and I would detect what the levels are in the coverage as I walk around a building. So one, I was doing it either for coverage or two, I was doing it because back then we didn’t have these wireless land controllers that helped us find rogue access points. So yeah, I was doing what was called war walking, looking to find access points that didn’t belong or I was trying to do my own survey to see where the actual dead spots were. So these are things that we need to do so we actually have a true idea of coverage because coverage is one of the big, again, considerations of how we design this network.

14. 802.11 Coverage Considerations Part6

One of the considerations you should have in coverage is, and this is when designing your wireless lan, is what happens if a client roams across a layer three boundary. Now, I already kind of gave you the idea. Layer three boundary just means that the access point that I was in, let’s say, was in the ten one one subnet. And then as I got to the other access point was on the other side of a router. Router. That’s a layer three device. And now the address there is 192, 168, 17, whatever. What’s going to happen is that when I do that roaming, I’m going to get a new IP address which is going to break all of the communications I had going on before I went from one access point to the other. Now, if I didn’t have to go from an access point on one side of a router to one on the other side of the router, then we could do layer two roaming. Which is great because think about it again, the whole wireless world that we’re talking about is all layer two. I hope you’ve very rarely heard me talk about IP addresses. That’s layer three. All we’re talking about are the Mac addresses, of which access point I’m connected to it’s all. Layer two works perfectly that way. So if you go from one layer three to another layer three and you have to get a new IP address, then all of those connection oriented applications are basically dead. If you were in the middle of a large file download let’s say you were downloading a 300 megabyte file and you’re at 97%, and then you go and you connect to the other access point while you’re walking. Yeah, forget it. You’re done. You didn’t get it all unless you took some extra consideration, iterations in your design, using something like a wireless lan controller to allow you to have that layer three roaming.

15. 802.11 Coverage Considerations Part7

So as an example, if I’m connected to an access point and the router, by the way, if you’re not used to the symbols, that’s the router. If I’m there and that router, every interface, by the way, on a router belongs to its own network, its own IP address. And that just means that there’s probably something out there, like a dhcp server that gives me my IP address anyway. So as I’m roaming, if I go across that boundary, I pretty much am going to end up with a new IP address from the other network that I just entered, unless we’re using something like a wireless lan controller. Now, a wireless lan controller here, they’re using generic terms like home agent and foreign agent. And I’ll try to describe that too, with even your cellular phone. I’ll draw a little quick picture of it. But the idea is once you make a connection, that wireless lan controller is the one that’s going to hand you out your IP address. And it’s also going to be talking to predetermined communications with the other wireless lan controllers in your network to be able to say, hey, I have this client who just connected to me. I’ve authenticated them, they’re good, whatever the case may be. And so when you come into the new access point and you get to that wireless lan controller, that second wireless lan controller is going to say, well, I’ll be the foreign agent, because your home agent, the very first one you connected to, told me that you exist and that you’ve been authenticated and that all is good with you. And instead of giving you a new IP address, in fact, what they’re going to do, they’re going to take all of your communications, right, that you normally do as they go through the access point. It’s going to have your original destination, well, whatever destination IP, your original source IP, and who cares what’s happening from layers four through seven. And they’re going to put it inside of a tunnel with the IP addresses for this new network. And so that’s your tunnel header, and it’s going to go to the wireless lan controller through this cap WAP tunnel to this wireless lank controller. That one’s going to rip the tunnel off and say, hey, okay, you’re good. I’m just going to send you traffic from here, not from where you were. Now I get it. That doesn’t sound like it’s a very efficient path in the world of IP routing. It’s not an efficient path. You’re not taking the shortest distance. You’re having to go back to this home agent to then get to wherever you wanted to go. And what if where you wanted to go was right over here? Right? Some computer just stuck over here. Yeah, you’re going to have to go all the way through and then make a uturn and come all the way back. But at the same time, you don’t lose your settings. We do that today in cellular communications, I’m drawing little trees because I think I mentioned some other time in this course that your cell phone providers make their cell towers look like trees so that people don’t think they look ugly. And I’m going to draw a picture of an iphone. And when you first connect to your cell tower, you’re going to connect to your own home agent, whatever your service provider is. And they have already determined, usually by a mutual agreement, to have tunneled communication. So I could be at verizon, this could be at T Mobile, but they already have contracts with each other to support each other. And so as you’re driving down this highway with your phone and you suddenly handoff from verizon to tmobile, your traffic is going to go that home, which will now be a foreign agent to you. Go back to your home agent, and then your traffic will leave from that location.

16. 802.11 Coverage Considerations Part8

One of the common mistakes when people were first putting out the wireless lans was to not worry about channel assignments. I mean, they said, okay, we want to have coverage. And by the way, these little hexagons are what they like to call cells or areas of coverage. And remember, technically, we wanted some of these to have some overlap so that that we didn’t have any breaks in the communications. And they might not have worried about what channel. I mean, at 2. 4 ghz, we got three channels and they would just turn them all on, maybe just took them out of the box, turned them on, and they all defaulted the say to channel one. So what we have happening is when we get somebody that gets in that area that’s in between, they’re going to have a tough time with their transmissions because they might be trying to send something to this channel one. And maybe there’s another computer over here that this access point is trying to talk to on channel one. And their signals are heard by each other and it causes confusion, it causes collisions. And so you don’t have as good a coverage.

17. 802.11 Coverage Considerations Part9

So the idea was, is to utilize different channels in each cell so there’s no overlap of channels. Now here they’re using these channels. One 4711. That, that’s fine. The goal is, is that from, say, this access point here, that the neighboring cells that I overlap with are not on channel one. One is on channel four or one’s on channel seven. Now, can I use channel seven again?

Sure, as long as that channel seven you’re using right. Is not overlapping with the other cell on channel seven. And so it gives you this ability to continue to grow your network here. Maybe this is where we need to have channel four again so that you have constant coverage. You can do roaming and you don’t have each of the access points interfering with each other.

18. 802.11 Coverage Considerations Part10

Another type of technology that came out, I’d like to say recently, because to me, it feels recent. I know 800 and 211 Inn has been out for a while, but that’s where we started using what they call channel reuse and channel bonding. Instead of using those 20 mhz channels, they decided to use two of them together, creating a larger 40 mhz channel. And yeah, that’s doubling the bandwidth, which really meant doubling the amount of data that we could send across that channel. So that channel bonding basically is going to double the frequency bandwidth, double your data rates that we can use to get those higher speeds like we’re seeing in 800 and 211. In and. In 800 and 211. AC.

19. Capacity vs. Coverage Part1

So we’ve talked a lot about coverage, but we don’t want to forget about capacity. When you design your wireless network, the two concepts that typically seem to compete with each other are capacity and coverage. In the early days of wireless networks, it was very common to just turn on the access point with the power set to the maximum level to get the largest coverage area that we could. And that was just deemed to be acceptable because there were usually very few wireless devices. But now we have a problem, don’t we? It’s all half duplex. The more devices that connect to that access point, the more devices that have to share the overall bandwidth. And so that becomes an issue with capacity.

20. Capacity vs. Coverage Part2

So in having said that, the solution would be is to have less power, smaller cells at each place where there’s an access point, so that we can have more user devices connecting to each one of these access points. Or it could be a phone or a tablet or whatever the case may be, so I can have more units. But they don’t all have to fight for the same access point because the access points coverage area would not be as big. If my access point coverage was really large, I’d have to take on all of the devices that were in that area, when in reality I’d like to have maybe kind of a load balancing. I would like to say, and yes, I get it, the more access points you have, the more expensive it is to set up your network. But again, what are the business needs? What are you trying to get out of this design? So we’re still trying to give you the coverage. We just don’t need to have everything on at full power so we have smaller cell sizes and more access points so that everybody’s not sharing the same actual wireless access point.