CWNP CWNA – Wi -Fi Access Part 2

11. SIFS/DIFS Part2

So when a client transmits a unicast frame, the duration ID field is going to have a value from zero to 32 767. And that really represents the amount of time in microseconds that’s going to be required to transmit an active frame exchange process.

And doing that, the other radios are not going to try to interrupt. And again, remember, everybody’s hearing this because, you know it’s a WiFi medium and they all hear your radio signal.

12. Carrier Sense Part1

Now, Carrier Sense, a part of the Csmaca, is the ability to hear and see if it’s clear to be able to send your traffic. So Virtual Carrier Sense is going to use this timer mechanism that they call the NAV, the network Allocation vector. The NAV timer is basically a prediction of what future traffic will look like on the media based on everybody’s duration, value, information that they saw in previous frames. In other words, it’s kind of like if you live in a city that has traffic, you might say,

okay, I’m going to go onto the computer. I’m going to look to see what traffic is like on interstate whatever. And when you see that amount of traffic being measured, you’re going to say, okay, it’s going to take me probably 15 or 20 minutes more to get to work. The NAV timer is a prediction of what future traffic on that medium is going to look like.

And so that’s a part of what Carrier Sense is about. It’s like, okay, you know, I know it’s busy here. I have some traffic I want to send. But based on everything I also I’ve heard, let’s wait for that NAV time in case I might step on somebody else’s traffic. And everybody, hopefully all the other clients that are on there, are going to be adding that into their choice of when they want to send their traffic. And remember, we’re putting it all in microseconds. So even though this really sounds like things are going to go slow, they really don’t.

13. Carrier Sense Part2

So when we continue to talk about carrier sense, another option is the physical carrier sense, and there’s two purposes of that. One is to determine whether a frame transmission is inbound for a station to receive. The other, of course, would be going out, is it busy? Before I can transmit? So if the medium is busy on that inbound, the radio is going to attempt to synchronize itself with the transmission.

And the purpose of that is to determine, like I said, if the medium is busy before it transmits. Now, this is known as the Clear Channel Assessment, or the CCA. And the CCA involves listening basically for those transmissions at the physical layer, meaning you’re actually hearing data coming over the radio frequency. And the medium must be clear before anybody else can transmit.

14. Carrier Sense Part3

So an OFDM station will select usually a random number from what they call a contention window. And that’s something I didn’t mention. Ethernet, back in the days of half duplex, was called a contention based media, meaning that you took a turn when you thought it was your well, when you thought everything was clear, you might collide with somebody else. So that was a contention, right? And with this half duplex frequency, we’re talking about contentions here.

So let’s just say in this example, the station chooses the number four. So the station is going to multiply that random number of four by the number of slot times, which is usually in the nine millisecond range. And then the random back off timer is going to have a value of that time. In this case, four times 936 microseconds, which is four slots. And for every slot time during which there is no medium activity, the back off timer is going to be decremented by the slot time. Now, the station that’s decrementing that back off timer will do so until the timer reaches zero. And if at that point the medium is still clear, then we’re going to say, hey, this is now clear to say end.

15. Point Coordination Function (PCF) Part1

The PCF or the point coordination function is another access method and it’s kind of a form of polling. In this case, the access point is going to do the polling and basically find out if there’s anybody who wants to send traffic. And in this case, it only really works in a basic service set.

16. Point Coordination Function (PCF) Part2

In some cases, we don’t want to slow down traffic by having an acknowledgement for every frame we send. But that’s how it works by default. So there is an opportunity for your senders to be able to send what we call a block acknowledgement request. And what it does is it basically says to the access point, let me send you three frames. And then you just acknowledge all three of those rather than by sending it to you, getting the Acknowledgment, sending it to you, getting the Acknowledgment. So it makes it a little bit more efficiency by aggregating those Acknowledgments into one instead of one for each frame. Now, there are two types of these block Acknowledgments.

One is immediate, which is designed for low latency traffic. Low latency would be like voice over IP video, things that don’t do well when the amount of time from one host to the other takes a long time. The other is more suitable for that type of traffic, like a file transfer. That it’s okay if there’s latency. We call that a delayed block ACK mechanism.

17. Block Acknowledgment

So to give you an example of this block Acknowledgment, again, is that the originator station is going to send a block of QoS data frames to a recipient station. That just basically means that I’m going to send you stuff that is time sensitive. Maybe there’s somebody with a telephone over here, and they’re going through a bunch of other routers to get through the network, and then they have to hit this access point. And rather than slowing the traffic down by waiting for an acknowledgement and acknowledgement and acknowledgement, we’re just going to say, hey, I’m going to send you this traffic and ask you for a block Acknowledgment request. And the recipient is going to say, oh, okay, no problem. I won’t acknowledge each and every one of these frames. I’ll wait for a number of them, and then I’ll send you an Acknowledgment that I got them.

18. Airtime Fairness

Lastly, the thing called airtime fairness. Well, let’s be fair, or maybe let’s come up with a way of defining fair. And that’s tough. I mean, if you’ve been a parent, you know that you’ll have kids. If you have more than one kid asking to do something or each wanting to do something, whatever your decision is, one of them is going to say, that’s not fair. Well, rather than just looking at them saying it’s not fair to get used to it, maybe I shouldn’t say that on my kids when they’re that young, but what I do is I say, well, how do you define fair? So let’s think about it.

If you have some older technology that maybe can only transmit information at eleven megabits per second, and you got somebody else over there who can transmit at 54 megabits per second, should I let the slower older technology send enough data to be equal to the 54 megabits per second on the other client? Or would it be fair for me to say, look, whatever your speed is, you get this time slot and I’m going to give the other person the same amount of time, even though the one with a higher bandwidth is going to do a lot more in that time period than the other one. So which part of that is fair? Well, here’s the solution.

When it comes to WiFi, it’s about the time slots. If you’re too far away from the access point so you don’t have a lot of bandwidth. Or again, if you’re an older technology, and again, right, if it’s down to bandwidth, whose fault is it if you’re too far away to get a high speed? So the access point says, you know what, to make this fair, I’m just going to give you this many time slots. I’m going to give the other person the same number of time slots. You send me as much as you can in that time slot. So it’s a way of being able to let those slower data rates come into our networks.

Now, before Airtime Fairness and I mentioned this other places in the course, if you had like an 811 G, which could go up to 54 megs, and somebody came in with some old equipment that was doing eight or 211 B, that in those days the access point slowed everybody down to the slower speed. So now with Airtime Fairness, we can say, hey look, you got old technology, that’s fine, I’m sorry, you get the same amount of time as everybody else. And if we didn’t do that, it could be a huge hindrance to the overall performance of the network. If I had to give the slow senders more time to be equal to those who are fast.

19. Module 08 Review

So in this module, we talked about Csmaca and how that’s different than the CD. Collision avoidance versus collision detection. We talked about all of these types of methods we use to make sure we can avoid collisions.

The distributed coordination function, DCF, the PCF, the HCF, the block Acknowledgments, airtime fairness all these things that we incorporate into communications to make this half duplex medium work for all of the clients that are connecting to an access point.