CWNP CWNA – Troubleshooting Part 3
Another concept, especially when it comes to channels. At the 2. 4 ghz frequency we only have the three channels. In the United States, we have eleven channels at 5. So what we want to do is try to get through what they call a band steering. That means we want to try to move people, steer them from 2. 4 to 5 ghz where we have more advantages, more channels, more coverage, better speeds than the rest of it.
And so if we can convince a device to do that move, then you would say we have steered them from one band to the others. So yeah, the 5 ghz uni bands have wider frequency space. Like I said, they have more channels. We could use channel reuse and bonding to get better speeds. Lots of benefits by moving there. And so that’s what we’re going to try to show you is that your regular access point can help convince a client which frequency to use.
One of the ways we can do that is if I have an access point that can operate at 2. 4 or 5. 0. And it’s listening to these probe requests coming from a client and it says, oh, I heard you on 2. 4, I heard you on five, I’m only going to respond to you from 5 ghz. So the poor client says, oh, there must be nobody out here at 2. 4. And so in a way, we done the band steering to get them into that 5 ghz range.
Load balancing a lot like capacity, again, is where and this, by the way, depends again, on the vendor who’s making your wireless devices. We do have methods to manipulate the Mac sublayer to try to balance clients between multiple access points. Now, because every vendor might do it a different way, I don’t have anything specific to say. This is the technology, the way they do that without getting away from being vendor neutral. But the goal is to say, look, this access point had 60 clients. And so capacity, I mean, 60 clients, they’re all sharing the same bandwidth.
And if I have the ability to take some of them and move them to other access points that are at least in range, then we can try to get some type of load balance where maybe the people over here, the three clients, are like, hey, we got a lot of bandwidth. nobody’s using this thing. And suddenly we move those new clients over there. And so the bandwidth might not be as good originally for those three clients, but it’s not as bad as it was for everybody that was in that third access point. So that might be useful.
Well, wireless lans were usually in the old days, put together without any type of site survey. Have I said site survey before? I’m sure I have, because it’s important. I’m going to try to make you all hate the fact that I say site survey. Anyway, they were designed mostly for coverage. Can I cover all of my office with as inexpensive access points as I can? And we were more worried about data applications. It’s like, who’s got a telephone that’s going to connect to a wireless access point? Well, guess what? We do. We have a lot of them. Or even softphones. Your tablets could be running something that uses voiceover IP. We call it a soft phone because it’s an IP phone, but it’s software on your tablets or your computers. And so, anyway, at that point, people were happy. The enterprise has said, hey, that’s not a big deal, we’re not doing voice that way. And the data applications are running just fine. But not everything will run as optimally as you would like. So, I mean, let’s take a look at it in voice packets. I don’t know how many times I’ve said this. You have 20 milliseconds of voice. Then they have a layer four transport protocol. They have the layer three, they have the layer two. But every one of these packets are the same size.
So they’re uniform, they’re very easy, and they’re not very big at all to be able to get through the network. But in the world of data, that packet could be up to, well on ethernet, up to 1500 bytes, unless we’re doing some data center stuff where you might have to have jumbo frames the voice, because it’s 20 milliseconds of voice. These packets are predictable. They’re going to come one after the other, after the other, after the other. We’re on data, it’s bursty. I send you a big file, you don’t need any more for me. And then I’m not using that connection anymore.
If something happens in the delivery of data because of tcp, there’s an error correction I can retransmit. If you miss something, eventually you get the file. Doesn’t happen over here with voice over IP, right? They’re using udp, which has no error connection or correction. Excuse me. And on top of that, they’re running a protocol called the Real Time Protocol rtp. And so the packets have to be consistent. Like I said, they have to have very low latency.
And in fact, believe it or not, if you miss a packet, if one gets dropped, we don’t want it back because we don’t want the voice to sound crazy. So really, the saying better never than late is true. With IP data, you’ve got to send me all of the data so I can reconstruct it. And so they have a better late, meaning it takes a while, then never getting that packet.
Performance is something else we have to look at when we are thinking of troubleshooting. So what do people complain about? They can’t connect. Maybe that was the rf interference we’ve talked about. Maybe there’s not enough coverage. We’ve been talking about that. The next thing people complain is, boy, the network is slow. And so we want to look at what we can do do to try to improve that performance. So one might just be the transmission power rates. That means that the original amplitude or power that you use is going to have the effect of the range of a cell.
Hopefully you remember, or if you haven’t, you’ll go look at that module where we did the mathematics, so you can understand what I’m getting at. But an access point that transmits at 30 milliwatts is going to have a much larger coverage zone than one that’s transmitting at one milliwatt if they’re using the same type of antenna. Now, if you have too much amplitude, you can cause a lot of problems with these radio frequencies. In fact, if you have the ability and I don’t know of a vendor that you could go to and buy an access point where you could go over the legal amount of power, but if you could, and you could turn that power up as high as you want, you could actually affect the health of the receiving transceiver. In other words, you could almost burn it out if you wanted to. But besides that, too much power and amplitude would maybe be too large of a coverage area and that could then be interfering with some other access points. But we can still add the perception of power because we talked a lot about the transmitter sending an AC signal that is going to be turned into radio frequencies, and that some of those antennas can passively increase what they see as the gain by maybe directing that radio frequency in a certain direction.
So if I have still at 30 milliwatts from the transmitter and I have an antenna that adds six DBIs of power to it, then I’m going to have a greater range than if that antenna only added three DBIs. So the idea here is if you want to increase the range for a client and change it by changing the type of antenna that you’re using, not by upping how much AC current is going through. Now, remember, the different types of antennas have different coverage areas. If you had one of these antennas that was omnidirectional, you see it going in every direction. Remember, you could have some like even a parabolic dish that may focus that radio frequency in a certain beam.
So it’s going to have more power coming out of it, more distance to cover. And the reason distance is important to the word performance. Remember, the stronger the signal is, when I get to see that signal on my antenna, the greater my bandwidth is going to be, which means the greater my performance. Now remember, a wavelength of higher frequency signals have a smaller wavelength. And that means that they’re going to grow weaker, faster, and may not penetrate through other objects like walls, as well as a lower frequency signal that has a larger wavelength. Again. Did you know a site survey could help you predict this? Again, I’m just throwing that out there. So all things being equal, 2. 4 ghz access points have a greater range. It doesn’t mean they have a better performance, but a greater range than a 5 ghz access point just based on the length of the waves. But it’s all part of planning, because to me, performance means that the stronger the signal that I get, depending on the design of the room, is where I’m going to have better performing networks.
So the free space path loss is an environment where I don’t have walls and chairs and all that kind of stuff. I’m just sending my signal out. Maybe it’s more outdoors than anything else. And so the signal really is a function of the distance and the frequency that I’m using. And in one of the other modules I described, it like throwing a pebble or a rock into a lake. And when you do, you see these ripples of water coming out. And as those ripples, the circumference of the circle that’s being made gets larger and larger, right? The waves get smaller and smaller until you almost can’t see it anymore. And that’s what’s happening in radio frequency in that environment. And that’s why they call it free space path loss. Now you start throwing in walls and other obstacles, they’re going to weaken that signal because of either absorption or they’re going to bend things. We, again also talked a lot about what happens to a signal and that can all affect the performance, how far away you are or what type of structure you have to go through with that radio frequency. If you had a building made of all concrete walls, you’re going to need more access points than if your building only had dry walls, because concrete is denser and it’s going to attenuate that signal faster than drywall. So if you’re looking for performance and you have a room that like at schoolrooms I talked about in classrooms made out of cinder block, I’m going to need a lot more of those access points to cover that school than if they were just in a regular office building that might not even use drywall to some aspects in between. They might just have really tall cubicle walls that you have to get through.
We also want to think about the applications that we use when it comes to performance. Obviously, different types of applications have a variable effect on bandwidth consumption. If you’re streaming movies on Netflix, hulu, or if I forgot your company, that does it. I’m sorry, I’m not a stockholder in either one of those. But those are going to affect bandwidth consumption, aren’t they? As opposed to you trying to make a phone call, would you uses very little packets and so we need to understand what the applications are so that we can better decide what the performance is going to be. wifi and data collection, scanning really don’t need a lot of bandwidth for you to be able to get that kind of information.
In other words, I can have something that collects the type of packets and the amount of the bandwidth being used so I can see if people are spending a lot of bandwidth on Facebook or Twitter or some other social media, all of these things. And we can analyze what the needs are for that bandwidth. Like I said, some might just be doing file transfers if it was getting information off of a database. High definition video streaming is also something that takes up a lot of bandwidth. So you have to remember it’s a half duplex medium that’s shared by everybody. And so the more connections and your knowledge of the type of traffic is going to affect the performance of that network.
When we look at problems that we want to troubleshoot. For outdoor wireless communications, one of the biggest factors is weather. And so whether it’s a point to point over a long distance or a wireless mesh network, the thing we have to worry about is what the medium looks like. If it’s clear and there’s no fog, there’s no rain, no snow, things are great. But we do have to think about that, because what, what can hurt electronics? Water. At least not if you told my four year old who one day decided to water my laptop. But I’m sure he meant it well. But nonetheless, it does seem to be an issue. Now, one source of big interest is lightning, whether it’s direct or indirect, right? When you have that electrical charge in the air, so it doesn’t have to be a direct strike, but it can surge or do power surges and cause problems. So they do make lightning arresters that you can put in between the antenna and the transmitter to try to take care of that problem.
I mean, at home, if you have expensive electronics, I’m sure you have some sort of power strip that it was designed to prevent those surges from destroying your in home equipment, your big screen tvs, everything else like that. So you could also put in lightning rods or have copper fiber transceivers. If they’re fiber that’s light, it’s not electrical. All of these can help you have some way of ensuring yourself against lightning strikes. Again, we have long distances, narrow beams to go those long distances, highly directional antennas. And if you’ve ever had cable TV or satellite TV at your house, you might have noticed as that dish, right, is on the little pole. If it’s really windy, that dish begins to shake. And when it’s shaking, it doesn’t take very much of an angle shift to make the beams not talk to each other. And so that can also be an issue if you have really strong winds. And so all of these are kind of what we should worry about.
Now rain and snow and fog, that’s going to make the medium of air harder to get through. So most by the way, not only that, but the rain and the snow that moisture can also destroy electrical equipment. So have your outdoor equipment should be protected from the water damage and the exposure to the water. That can be a very serious problem with your cabling and connectors. So those, those connectors should be protected with things they call drip loops or coax seals to prevent the water from getting in, as well as your radio frequency signal is going to attenuate.
Now there’s nothing you can do about that. If it’s raining, your signal is not going to be as strong getting through. If you have a torrential downpour like they had almost every day that I was in Nairobi a while ago, I couldn’t even see across the street, let alone wonder what my signal looked like. They tell us that in whatever you define a torrential downpour, that your signal might degrade as much as 0. 8 DBS per mile in both the 2. 4 and the 5 ghz frequency range.
Well, we were trying to talk about issues we would look at for troubleshooting in this module. We talked about layer two retransmissions coverage considerations, capacity, voice and data, the performance of your network, weather.
And the only thing really about upper layer troubleshooting is once we get those successful frames sent, at that point we can hand it off and say, now it’s your network infrastructure structure or it’s your servers that you have to troubleshoot.
Popular posts
Recent Posts