CWNP CWNA – Troubleshooting

1. Troubleshooting

So we’re going to talk about troubleshooting. And when we look at troubleshooting, we’re going to talk about all of the things that can cause issues with communications from a station to an access point. So look at things like what caused layer two retransmissions some of the coverage considerations, which, again, could be troubleshooting. Somebody says, hey, the wireless doesn’t work? Well, maybe it’s because there’s no coverage there. We’ll look at the differences between troubleshooting voice and data.

We’re talking about things like performance and capacity, how weather can affect your outdoor wireless communications. And then when it comes to the upper layer upper layer meaning the osi model, the IP addresses, the transport, the applications, how we can point the finger at them and say, it’s your application that’s broken. Okay, no, we won’t point the finger, but at least we can hopefully rule out the fact that it was our part of the network that was causing the error.

2. Layer 2 Retransmissions

So when we talk about retransmissions, 800 and 211 is not 823823, is ethernet, and in unicast 800 and 211 frames must be acknowledged if the frame was intact. In other words, if my PC sends a transmission to the access point, then the access point, if there were no problems, would send back to me an acknowledgment that lets me know that it was successfully received. If I don’t get that acknowledgment, the only assumption my computer can make is that something happened, something caused it to fail and I would have to retransmit that data again, wasting more bandwidth and time to hopefully get that acknowledgment.

So what could cause a problem? Well, there could be another computer out here that was also transmitting at the same time, causing this big collision, if you would. So that maybe when the access point receives it and looks at the crc, the cyclic redundancy check, some people also call that the fcs, the frame check sequence. And if that fails, which is an indication that there was some sort of corruption, then the radio just doesn’t do an AC. So again, that’s why the question mark. If I don’t get an AC, why not? And the assumption is going to be that there was some sort of collision that caused a problem. And if I have to keep retransmitting data, it’s going to adversely affect the way in which the wireless land works in two ways.

First, having to do retransmissions just means more overhead and subsequently less throughput. And if I have to keep retransmitting data at layer two, the delivery of the application traffic becomes delayed or inconsistent. And in fact, there are some applications that will not retransmit. So if I have a phone, by the way, that was a phone connected eventually, let’s say to a wired phone somewhere, if I send that 20 milliseconds worth of voice and I don’t get an app back according to that protocol for that application, I cannot retransmit it. What would it sound like to you if you’re listening to a phone call and somebody’s 30 seconds in the phone call, you’re suddenly getting part of the conversation that occurred 31 seconds ago. I mean, it wouldn’t make sense. It would just sound wrong. It would sound like they’re in a blender. So that can affect the application. And those are all bad as well if there is constant interference. We have other problems with applications like voice over IP. So let’s talk about voice over IP. Voice over IP from one phone to the other. Let me just put two more phones in here and you’ve probably heard me say this before, that this one way communication has to occur in under 150 milliseconds.

There are some things we cannot do about that time. In other words, if you are 4000 miles apart from one phone to the other at the speed of light, it’s still going to take a little bit of time for your voice transmission to get sent. Unless we have found a way to send communications faster than the speed of voice or speed of light, which we haven’t, then there are some things we cannot change. Each of the devices in between are going to take a few milliseconds to also be able to process and decide how to route and then turn them into ones and zeros and transmit it.

So there are some fixed times we cannot change. And if one of the times I’m having a problem with is just getting my packets to the access point, that’s going to affect what’s happening now. latency just simply means that if it takes too long, you sound like you’re on a walkie talkie instead of a phone. You speak, you say hello and you wait and you wait and you wait. And finally the other person hears the hello and then they answer you back and you wait and you wait.

And by the time it gets back to you, you’re probably saying, are you there? jitter is where parts of that message show up a little bit later than others. In other words, I said that every one of the packets I’m sending from one phone to the other is 20 milliseconds of your voice. I know it doesn’t sound like a lot, and it’s really not, but let’s say that I’m sending you these packets. By the time you process the first 20 milliseconds, you’re hoping the next 20 millisecond seconds is there ready for you to hear. If it takes an extra 30 milliseconds for you to get that next one, then it sounds a lot again, like that voice is in a blender. And that could also be a problem either through retransmissions or through other types of problems in our wireless design.

3. RF Interference Part1

Now when it comes to the types of rf interference that can affect the performance, it could be an interfering device that is preventing your regular 800 and 211 radio from transmitting, causing a denial of service. You know, you can go online. Anybody can go online and buy a jammer for different frequencies, at least for the unlicensed frequencies. There are at least, I hear reports of theaters that go out and buy jammers for the cellular frequencies so that people can’t use their phones while they’re watching a movie.

I don’t know if that’s really a good thing when it comes to public safety, but that’s what I hear. I could buy a jammer for the 2. 4, the 5. 0 keep anybody from having wireless connectivity while they’re inside of my house if that’s something I wanted to do. You could block specific channels or I could just have a really poorly made vacuum cleaner that emits rf out of the electromagnetic engine. That might also cause interference. Or maybe the kids that I have when I bought them remote control cars for Christmas are turning them on and by the way, they operate at 2. 4 and maybe those cars are causing interference. I mean, there are lots of devices that can create interference. The hope is that the strength of that signal is not greater than the strength of what my access point is trying to send so that we have a good snr there where we can still have our data communications. You might have a device that is using narrow band. Remember, narrow band means that in a frequency they’re using one channel. And so it’s a very small and finite frequency space and it might not cause a denial of service for the entire band, but for anything operating on that channel it could be a difficulty.

And if you were using the really, really old frequency hop or frequency hopping spread spectrum, that could cause a problem for that one particular channel that you would hop through. So the other thing with narrow bands is they’re usually very high amplitude. Now remember most of the narrow bands we talked about were way outside of wireless. That was the Am and the FM radio type of thing. But if you have it, it will absolutely disrupt your communications. Now, a source of interference, again, could also be a wide band. Wide band means that whatever is sending that interference is taking up the entire band and that could disrupt communications for everything within that band.

4. RF Interference Part2

So the term allband interference is usually associated with the frequency hopping spread spectrum, a technology we hardly ever use. But it would disrupt the communications because this is a technology that would hop from one channel to the other channel to the other channel. Good news, it was only at the 2. 4 ghz range, so we didn’t see it. The 5. 0 multipath, which today is a really good thing for high speed communications, but could, if you’re not running 811 N or AC, could cause what they call inter symbol interference isi, which could cause data corruption.

Remember, multitath is a signal bouncing, reflecting off of other objects and arriving at the receiver at different times. And that could cause a problem in understanding the signal. And that little difference in time from the primary signal and all those reflected signals, which we call the delay spread, could, as I said, cause that receiver some problems in how to demodulate that signal into ones and zeros. Like I said, today we use the mimo or mimo technology to our advantage, to have multiple paths, to send multiple bits of data. So today it’s constructive. It just depends on how old. Again, I said the equipment that you’re using is. 

5. RF Interference Part3

Do you ever remember the day when your car, or maybe your parent’s car had Am radio and you have this little chart of little places where the frequencies were? And believe it or not, I know this is really weird to think we had a little knob that we had to turn to move a little arrow across to get it to a channel. And of course we had volume. And as we’re moving that little arrow back and forth, we were looking for what sounded like the strongest signal for our favorite radio channel. I don’t know if I miss those days, I’m not being nostalgic, but the point of that discussion is that when we look at a frequency band of like 2. 4 ghz, there are multiple channels that are spread apart by 20. Problem we have is that if we didn’t regulate which bands or which channels, I might choose to use channel one. But a part of channel two, look at that part of channel two might interfere with channel one. And you would probably have noticed that because as you’re tuning it, as you’re moving that little dial

across, getting close to channel one, you might hear it, but it doesn’t sound so good until you move it just a little bit more. But then that would be difficult if there was another broadcast on the adjacent channel that was interfering. And so we’re trying to avoid what they call the adjacent channel interference that could degrade the performance by having an overlapping frequency space. And so that’s why when they chose the channels like channel one coming right here, they made sure there was enough distance that the next channel had absolutely no interference. In fact, they purposely put a 5 mhz separation here to make sure that there was no interference. And that’s why we have at least at this frequency, just the three channels, one, six and eleven. And so we didn’t have that problem. And that made sure that all of the channels were non overlapping so that we would not worry about one access point on channel one from interfering with another access point on channel six.

6. RF Interference Part4

The signal to noise ratio, S and R is also important for us as well. I know we talked about it, but now we’re back into the realm of troubleshooting. What in this noise floor or noise level is causing interference over our channels again, it could be kitchen appliances, almost anything with an electric motor. It could be doing it. It could be something that was designed to offer operated at that frequency, maybe just not as high of amplitude, like I said, the remote control cars. But it could, as background noise, cause problems for my receiver to be able to understand the signal that’s being sent by the access point. And if the noise background is too close to the received signal. In other words, if this noise starts getting up here, then my effective usable bandwidth here, when I look at the decibels that I have, might just suddenly become too weak to be able to understand.

And it gets worse, right? Because as we said, even through free path loss, as I get further from the access point, the signal gets weaker. And if I’m sitting over here and that noise floor is high enough, even though I should be in range, I might not be able to understand what’s coming by. You might remember I talked about this example that if you’re at a rock concert or a country music or whatever, you’re listening to some band playing live and they’re usually pretty loud.

The person next to you speaking to you so you can hear them has to make sure that their signal strength of their voice is louder than the music and high enough above it so you could actually hear them, rather than just thinking they’re mouthing the words or something like that. That could become a problem where even though the signal should be strong, the actual noise floor is taking that signal out. So it’s not actually a ratio, it’s simple subtraction. We figure out how much that noise floor is and we basically subtract it from the actual signal strength to see what’s left over.

7. RF Interference Part5

Another type of interference would be this one here where it says disproportionate transmit power settings between multiple clients can also cause communication problems, at least especially in a basic service set. Now, it says a low powered client station that’s a great distance from the access point could become unheard or an unheard client. If other high power stations are close, what does that mean to us? It means, look, we’re within range.

This one, because this laptop is closer to the access point, is hearing things at 100 milliwatts. But remember, as we move away, the strength of that signal through attenuation becomes lower. And if you can only speak at ten milliwatts and you’ve got this station over here at 100 milliwatts, that noise on the same channel might be so loud that nobody hears Station B talking. It would just be like, maybe, you know, people who love to talk over to the top of other people, and unfortunately, whoever’s loudest is the one that’s going to be listening.

So most your lower power do not because of anything other than distance from the access point, and it would become unheard. In other words, the access point would say, was there somebody else talking? I can’t hear you. Because the other signal is really strong. And maybe there’s even more devices that are really close that are also strong, keeping that weaker voice from being heard. So what in essence is happening from the point of view of this Station B that’s a long ways away, is that the noise floor caused by all of these other closer devices is so high that nobody can hear the ten milliwatt conversation. And so it would corrupt the fire stations incoming transmissions and basically prevent anybody who’s too far away from being heard.

8. RF Interference Part6

Now, the hidden node, that’s a tough one. So we talked a little bit about csma CA carrier sent multiple access, collision avoidance. And the idea was that if I sent my CA packet to get the attention of the access point, then it’s my turn to talk. And while I’m talking, my signal is hurting by all these other laptops. So they know not to speak. But maybe because of a thick wall or some other situation, I have somebody who is in range of the access point, but out of range of hearing the other computers talking. And so from that point of view, they would think, well, there’s nobody talking to the access point, therefore it should be clear to send and they send their packet during that other conversation. And what happens? We have collisions. That could be one example of a thick wall.

If I had enough room here, I’m going to try it anyway. So if I have a computer whose range is, let’s say this, and they’re in range of the access point, and I have another computer over here that is also in range of the access point, but not in range of the speaking computer, so what does that mean? No thick wall there at all? It’s just the fact that the antenna for that computer cannot hear what’s being sent by that computer because they’re to each other out of range, but yet both in range of the access point that also can act as the hidden or the hidden node. So basically the problem as assessed here is that the physical carrier sense is that all stations may not be able to hear each other. To make this opportunity of what do we call it in the dcf, the distributed communications. They may be trying to follow the rules, but they just can’t hear those other computers.

9. 802.11 Coverage Considerations Part1

Now, when we look at wireless and the coverage considerations, I really think we should look at two things. One is the coverage and the capacity to help solve a lot of our problems. So coverage could be roaming problems or interference that could we could do something to mitigate that or design in advance. And so the big thing I want you to take out of all of this is if you want to solve problems, do it before the installation. You should do a site survey so you can learn where the interference is. You can learn what the coverage areas are. You do not want to wait until you just, you know, maybe you have some legacy wireless network that was put in before you ever got to this company. And they’re saying, hey, this isn’t working for us. And what do they normally do? Normally they would say, okay, look, here’s my office building.

I’m going to throw in an access point and I’m going to turn it on to full power, and we’re going to get as much coverage as we can. And then maybe I’ll put another access point over here, turn it on full power, because we didn’t think about where it’s at, what the walls look like and the coverage. And then you might say, well, okay, let’s just put one more in here to fill the gap.

And now we don’t even have resiliency in this network, right, the ability for access points to go from a lower power to a higher power to fill in a missing access point if one were to go down. But their goal there really probably was just coverage, not capacity. They might not have designed their network for people to be able to roam through the office. And so these are things that we can solve ahead of time by doing a site survey and understanding what are the needs of the business.

10. 802.11 Coverage Considerations Part2

Another part of coverage considerations is, again remembering the distance that a client station is from the access point because as they move further away, they’re going to do what’s called dynamic rate switching, meaning that their total available bandwidth will continue to go lower as they move away and the signal gets weaker. Now, an access point and a computer can both support multiple data rates based on whatever type of spread spectrum technology that they’re using. But we have some problems. Remember, one of the problems I said was some interference by having some that have such a low signal that they’re being drowned out by those computers that are closer to the access point. And so we have to think about that because this does affect coverage and how far we want somebody to be from the access point.