Chapter 6 Wireless LANs 1
Chapter 6
Wireless LANs 1
Welcome to the first chapter on Wireless LAN communication. One of the main reasons I use the Panko & Panko text is because it dedicates almost 20% of the book to wireless communications while other CIS/MIS networking textbooks give relatively little emphasis on how wireless works, the pros & cons of wireless networking and, most importantly, how to deal with important aspects such as wireless interference and security. That’s just crazy because WiFi is an important part of almost every modern organization’s LAN and the flexibility it provides comes with a number of limitations that need to be understood. Well, lets get into it…
1
Why study 802.11/WiFi?
2
Be a WiFi Hero!!
As a reward for downloading the slides from the class website, you get to enjoy a great video. Is it really great? Why, yes it is when taken in the context that someone took the time to craft a somewhat humorous short video about troubleshooting WiFi. Watch it…
2
Basic Concepts
So let’s start out with some basic wireless networking concepts
3
6.1 802.11 / Wi-Fi Wireless LAN (W LAN) Technology
Wireless LANs
Require standards at the physical and data link layer
So 802.11 WiFi standards are O S I standards
Standards created by the I E E E 802.11 Working Group
Wi-Fi
Certification system managed by the Wi-Fi Alliance
Wi-Fi now synonymous for 802.11
First, take a look at the graphic on this slide. Awesome, isn’t it? The graphic is there as a matter of perspective. Up to now, we’ve been focusing on the wired Ethernet LANs that dominate every organization’s network. Well, with a few exceptions such as facilities that require extremely strong security, wireless LAN connectivity is also omnipresent in modern organizations. The proliferation of devices without physical Ethernet ports (such as smartphones, tablets, and most modern laptops) mandates some way for these devices to connect to the Internet and corporate information resources, both of which reside largely on the wired side of the network.
Just as 802.3 is synonymous with Ethernet, 802.11 is synonymous with WiFi. Wireless communications standards are created by the IETF IEEE 802.11 working group. Outside of the IETF, but often working with the 802.11 WG, is the WiFi alliance, a group of businesses that “want to drive the direction of the WiFi industry and develop “WiFi-certified” products. The WiFi alliance helps propagate new 802.11 standards, but they can also get ahead of the IETF and present their own “standards” which sometimes work out well, and other times not. For example…
4
6.1 802.11/Wi-Fi WLAN Technology
Wi-Fi Alliance sometimes develops new standards outside of the IETF
Two notable standards are
WPA
WPS
.
6-5
The WiFi alliance pushed for the adoption of the WPA (WiFi Protected Access) protocol in 2003 when it became evident that criminals were easily exploiting the dominant wireless security protocol known as WEP (Wired Equivalence Privacy) (which we will talk about later). WPA was a critical stopgap between WEP and the IETF-approved WPA2 and, in that sense, was an important development pushed by the WiFi Alliance to its members.
WPS, on the on the otherhand, is widely hated by those that are concerned about security. WPS stands for WiFi Protected Setup and was created in 2006 to make connecting a host device to a wireless network as easy as pushing a button. The problem is that WPS is easy to break using a brute-force password /PIN guessing attack. More on this later.
5
6.2 Access Point Operation (2 of 2)
The device that connects a wireless client to the wired Ethernet LAN is called a wireless access point (or WAP). Note that at home you normally call your wireless device a Wireless Router, and that is true because the home wireless router you have includes many functions which we have discussed earlier (such as basic firewalls, router, DHCP server, etc). The WAP in a business network has one main function – convert wireless frames to Ethernet frames, and vice versa.
It is important to note that in almost all corporate and home networks, the purpose of wireless communications is to extend the capabilities of the wired Ethernet LAN out to wireless clients. The wired LAN provides access to critical services, from Internet access to printing and file servers. The exception to this would be for an all-wireless LAN, something called wireless mesh networking. All-wireless mesh networking is rare in just about any organizational network, but there are some home wireless vendors that are capitalizing on new smart-home technologies and offering an all-mesh architecture. The cost is quite high right now, but you would expect the prices to decrease as more vendors engage in the market.
6
Radio Signal Propagation
As with the previous chapter, we are not going into the physics of radio waves and signal propagation, but we do need to discuss some basic terminology.
7
6.3 Electromagnetic Wave
If you note, the terminology for radio waves is similar to that of light pulses in fiber optics in terms of amplitude and wavelength. However, for RF (RF = radio frequency) communications, wavelength has units of Hertz (Hz) which measures cycles per second. How was wavelength for fiber optic light pulses? Not sure…. You should review that information.
8
6.4 Omnidirectional and Dish Antennas
As with all RF comms, antennas are involved. The most common antenna we use in modern computing devices engaged in wireless comms is the omnidirectional antenna, which radiates and receives signals in all directions. As you can imagine, it takes a lot of energy to transmit a signal in all directions. We’ll get into this and other wireless propagation limitations shortly.
Directional antennas have an advantage over omnidirectional in that they can send signals in a very narrow beam, allowing data to be sent to a comm partner at much lower energies. This also allows longer distances for communications, which is why you see dish-like antennas on the roofs of houses to receive satellite TV signals. However, the main issue is that sender and receiver have to be oriented towards each other, something very difficult for small and moving electronic devices. However, as more advanced algorithms and technology enters the market, the ability to create and benefit from virtual directional antennas offers great potential benefits to consumer wireless comms.
Regardless of which type of antenna you are using, you are still sending radio waves of energy that represenent 1’s and 0’s through the air between two comms partners, and a lot can go wrong in this process. Let’s discuss the big five issues…
9
6.5 Wireless Propagation Problems
Two forms of attenuation
Two grow worse at higher frequencies
This picture summarizes the five forms of wireless propagation problems. As called out, there are two forms of attenuation, inverse square and absorptive. And two of these problems are worse at higher frequencies, shadow zones and absorptive attenuation. Like the discussion of wired propagation problems in the previous chapter, you need to understand these problems, and how to deal with them, because you will encounter some or all of these problems at some point in your IT career.
To put this in proper context… imagine your company relies on wireless communications to connect multiple, important employee and network devices (such as the office coffee maker). What problems could interfere with these important comms?
10
6.5 Wireless Propagation Problems (1 of 4)
The first wireless propagation problem is EMI. As you can imagine, energy from other electronic devices can interfere at the frequency you are using to send 1’s and 0’s via your wireless access point, and this can be a huge problem in almost any environment, but especially in manufacturing and crowded spaces.
How do you deal with EMI… well, the difficulty depends on the source of the EMI. If you own or control the placement of the thing that is causing EMI with your wireless network, you can possibly move it or place some form of shielding near the source to limit the impact. However, if the source of the EMI is a nearby radio tower, arc welder, or some other device that you have no ability to manipulate or control, you are going to have to live with it and maybe move your wireless devices further away from the source of the EMI, if you can.
11
6.5 Wireless Propagation Problems (2 of 4)
There are two types of attenuation problems that you have to deal with. First is absorptive attenuation and simply means that wireless signal energy gets absorbed by the things it passes through. Things such as rain and particulate matter in the air increase absorptive attenuation and decrease the range that wireless signals can reach while providing accurate data transmissions. The main thing you can control with absorptive attenuation is to limit the things between the WAP and important wireless devices that might absorb the signal.
The second big problem is worse for omnidirectional antennas (which, as you recall, dominate in corporate and consumer wireless devices) and it is called inverse square attenuation. The Inverse square attenuation formula is simply stated as a the power at a certain distance is inversely proportional to the square of the distance moved from the antenna. Not so simple to understand? Let’s look a table, because that makes everything simpler… maybe not, but might in this case.
12
6.6 Inverse Square Law Attenuation
Distance Ratio Distance Ratio Squared Signal Strength Compared to Original Initial Power (P2)
(watts) Final Power (P2)
(watts)
1 1 100.0% 100 100
2 4 25.0% 100 25
3 9 11.1% 100 11
4 16 6.3% 100 6.3
5 25 4.0% 100 4.0
6 36 2.8% 100 2.8
7 49 2.0% 100 2.0
8 64 1.6% 100 1.6
What does this table tell you? The further you move away from a WAP, the drastically worse your received signal will be. The example the text provides illustrates the problem – let’s say the wireless energy received by a device at 10 meters from the WAP is 100 watts. A second device, which is only 10 meters further away (now double the distance from the WAP) receives only 25% of the power associated with same wireless signal. Eventually you will move far enough away from the WAP where the signal received will be so weak that it cannot be differentiated from background noise, so you won’t be able to see the 1’s and 0’s being transmitted.
How do you deal with this? Two ways (1) move closer to the WAP if you can or install more WAPs to decrease the distance, and / or (2) increase the power of the transmitter (although this will have limited benefits based upon the inverse square formula).
13
6.5 Wireless Propagation Problems (3 of 4)
As mentioned previously, two propagation problems get worse with higher frequency transmission, and this will become more important once we discuss the two frequencies used by WiFi – 2.4 and 5 GHz. For Dead Zones, think about things like typical walls in an office, made of sheetrock, wood, with some metal in there. As you can see just about every time you look at the WiFi signals near your computer, there are MANY more 2.4 GHz WiFi networks available than 5GHz networks, and part of the reason for that is that 5GHz WiFi networks don’t penetrate typically building materials as well AND anything in the air or environment that can absorb radio waves absorbs higher frequency WiFi comms more readily.
14
6.5 Wireless Propagation Problems (4 of 4)
One of the more confusing propagation effects is multipath interference, which is the near simultaneous receipt of a transmitted radio signal from multiple paths due to reflection of that signal off objects, such as walls, floors, ceilings, filing cabinets, etc. So what?
15
6.6 Multipath Interference
The worst case scenario is that two reflected signals can arrive at a receiver in opposite phases of the radio wave, effectively canceling out the received signal. Or perhaps the signal comes in twice as strong, and every combination in between. In the end, over time and with a lot of data transmitted over WiFi, you end up with less reliable communications. The impact can be drastic and it has been noted that in some cases of multipath interference, moving your computer even a few centimeters away from a bad-interference convergence point can result in markedly better performance. The text notes that multi-path interference is the MOST SERIOUS propagation problem for wireless networks, and this is true to a point. It was most definitely the most serious problem for earlier 802.11 protocols, but newer implementations of 802.11n, 802.11ac, and future wireless protocols and equipment can actually use planned or predicted multipath interference to help provide greater bandwidth at longer distances. You don’t need to know the details of how this is done for this class.
16
Service Bands and Bandwidth
Let’s talk about service bands and bandwidth
17
6.8 The Frequency Spectrum, Service Bands, and Channels (1 of 2)
Anyone that has used an FM radio would recognize that there is a certain range of frequencies associated with the channels you can receive. The entire spectrum of radio energy starts at 0Hz and technically goes to infinity, but there is a logical limit because, as discussed earlier, the higher the frequency, the greater the absorption and shorter the distance traveled (in general, and again, I’m not getting into the physics of RF energy here).
What we care about is the specific service band, or the range of frequencies, that we are allowed to use for our networking WiFi comms.
The two most important WiFi service bands are 2.4 – 2.4835 GHz and 5.25 – 5.725 GHz. Yes, these are numbers you should know.
18
6.8 The Frequency Spectrum, Service Bands, and Channels (2 of 2)
Within each service band, there are designated channels. Let’s bring this back to the FM radio example. The FM radio band in North America is 88 to 108 MHz, but note that this range is different depending on the region of the world and the governing laws associated with RF radio bands. All countries use FM channel center frequencies ending in 0.1, 0.3, 0.5, 0.7, and 0.9 MHz and, in the Americas, the channel bandwidth is 200 kHz. Thus, if your favorite radio station is 96.5 MHz, the 96.5 MHz represents the center of the channel that has a low frequency of 96.4 and a high of 96.6 MHz.
19
6.9 Signal Bandwidth
Now, as you can guess, even though your radio is tuned into, say 96.5 MHz, there is some energy from the radio signal spread throughout the entire 200 MHz channel width, with the strongest signal at the center of the channel. If you were to manually tune your car radio to 96.4 MHz you would probably be able to hear a garbled and weak version of the main signal for the radio station.
Also, faster signals spread over a wider bandwidth. And, wider bandwidth allows more data to be sent. Think about AM vs FM radio. The AM radio service band in N. America is 535-1605 kHz and channels are only 10 kHz wide. So what? Well, the lower bandwidth of AM channels is why the same music transmitted on FM sounds so much better than the same song on an AM channel – there is less bandwidth to transfer the complex sounds of a song in an AM channel.
So how does FM channel width compare to TV, for example? Well, you would expect TV video to use a lot more bandwidth to transmit both sound and image, and you are correct. A VHF TV channel is 6 MHz wide compared to only 200 kHz for FM radio.
20
6.10: Channel Bandwidth and Transmission Speed
Channel Bandwidth
Channel bandwidth is the highest frequency in a channel minus the lowest frequency.
An 88.0 MHz to 88.2 MHz channel has a bandwidth of 0.2 MHz (200 kHz).
Higher-speed signals need wider channel bandwidths.
6-21
The bottom line here is an important one – if you want higher speeds, you need wider channel bandwidths, This fact is exploited by numerous 802.11 protocols.
21
6.10 Channel Bandwidth and Transmission Speed
Transmission Speed and Channel Bandwidth
There is a direct relationship between required transmission speed and required channel bandwidth
Doubling bandwidth doubles the possible transmission speed
Multiplying bandwidth by N makes possible N times the transmission speed
6-22
The simple rule here is that if you want to multiply the amount of data transmitted on a radio signal, the easiest thing to do is increase the width of the transmission channel proportionately. This is a gross simplification of Shannon’s Law, which defines the maximum rate of information that can be transmitted over a channel. Let’s push the I-believe button and move on.
22
6.11 2.4 GHz and 5 GHz Service Band
The 2.4 GHz Service Band
2.4 GHz to 2.485 GHz
Propagation characteristics are good
For 20 MHz 802.11 channels, only three nonoverlapping channels are possible
Channels 1, 6, and 11
6-23
As previously mentioned, there are 2 important WiFI service bands. The most commonly used WiFi service band in use is 2.4 – 2.485 GHz, and this is associated with 802.11g and 802.11n wireless communications, as well as some earlier versions that should hopefully no longer be in use. This is a great frequency for communication and because it is, it is also used by other technologies like home wireless phones (not cell phones, the other kind), Bluetooth comms, and other non-regulated radio comms equipment. Because of the popularity of the 2.4 GHz service band, EMI quickly becomes a big problem.
Now, the default bandwidth used in 802.11 comms at this frequency is 20 MHz to allow the desired speeds for WiFI (remember, wider channels = faster). There is, however, a problem… the designated range of 802.11 channels are very close to each other, in 50 kHz intervals. Thus, you need 4 channels in this frequency range to transmit 802.11 data using 20 MHz wide transmission channels. This results in only 3 non-overlapping channels, 1, 6, and 11. This is important stuff. Why?
23
2.4 GHz 802.11 Channels
24
First, look at this graphic of the 2.4 GHz WiFi spectrum. Note that because of some funny math, there are actually 14 possible 50 kHz-wide channels. However, in the Americas, only channels 1-13 are available for use. If your WiFi allows you to use Channel 14, DO IT!!! This is the only guaranteed non-overlapping WiFi channel in a populated area. Why is this important, you ask, to have non-overlapping channels? Let’s look at a visual to help explain…
24
6.11 2.4 GHz and 5 GHz Service Band
The 2.4 GHz Service Band
Co-channel interference between nearby access points transmitting in the same channel degrades performance
Except in very small networks, difficult or impossible to put nearby access points on different channels (Figure 6-12)
As this fishing graphic shows, waves that are initiated too close to one another cause interference, which is horrible for wireless comms because it causes data errors which requires retransmission which is slow. Let’s go back to the previous graphic.
25
2.4 GHz 802.11 Channels
26
Soon I will show you how to use some tools to look at the WiFI channels in any location where you can physically get to. For example, at my house recently, I was able to see a multitude of WiFi signals. When I look at the characteristics of 4 of these WiFi networks, I see that there are 4 networks within “sight” of each other that are operating on the same frequency, 2.437 GHz, which is channel 6, one of the three non-overlapping channels.
Great, these WAPs are operating on non-overlapping channel (so there is no interference from WAPs on channel 4, for instance), but all 4 are transmitting their 1’s and 0’s on the same frequencies at the same time. Like 4 people shouting over each other to be hear, I can guess that none of these folks are enjoying good WiFi performance.
What can we do for the folks that own these networks? Probably not much. All WAP owners could coordinate with each other and choose non-conflicting channels, and that would be the best thing to do, but it’s probably not practical. What would I do? Constantly survey my WiFi environment for these types of conflicts and change my channel to a least busy one of channels 1, 6, or 11. OR, you could load custom firmware on your WAP and transmit on Channel 14 and not worry about anything. This is harder to do, of course.
26
The 5 GHz Service Band
More bandwidth, so between 11 and 24 non-overlapping 20 MHz channels
Makes it easy to have nearby access points operate on non-overlapping channels
Increasing channel bandwidth in newer standards reduces the number of possible channels
6.11 2.4 GHz and 5 GHz Service Band
At this point I am going to deviate from the text a little and take things to the next level (but not too next level…). The text rightfully points out the following: 5GHz WiFi has a much wider service band and this allows allows 11-24 non-overlapping channels (depending on whether you use 40MHz or 20Mhz channels), which is way more awesome than the 2.4 GHz band that has a total of 3 non-overlapping channels. Likewise, if newer 802.11 protocols, such as 802.11ac, decide to use wider channels to allow greater speeds (such as 80 or 160 MHz channels), you will be able to have faster communications and still have the same or more non-overlapping channels than the 2.4 GHz band. Also, the wireless propagation effect of higher-frequency comms being more vulnerable to dead zones and absorption can actually come in handy. For instance, [advance slide] the WiFi capture earlier had a total of 13 2.4 GHz WiFi networks visible in my house compared to just 2 5GHz networks visible. Thus, this potential negative has a positive effect of higher frequency WiFi can be expected to not travel as far and interfere with your network transmissions. [advance slide]
The text does not, however, talk about a limitation in the 5GHz range called DFS, which stands for Dynamic Frequency Selection. Wi-Fi networks operate in areas of RF spectrum that require no license to operate. This is great because wow, would it be painful, to have to get a license every time you move or add a new WAP to your home or office network. However, just because WiFi operates in an unlicensed band does mean there are not restrictions to ensure that there will be NO unintended frequency conflicts with important services related to public safety. One particular service that shares spectrum with wireless LANs is radar. Some types of radar installations operate in the 5GHz band. This means that they may use some of the same frequencies that are used for Wi-Fi networks. Due to the potential coexistence of both radar and Wi-Fi networks in the same area of spectrum, the official 802.11 Wi-Fi standard was designed to incorporate a spectrum sharing mechanism on 5GHz to ensure that Wi-Fi networks do not operate on frequencies (hence causing interference) that are used by nearby radar stations. This mechanism is known as Dynamic Frequency Selection (DFS) and is designed to mitigate interference to 5GHz radar by WLANs. So what does this REALLY mean to the previous discussion? In short, there really aren’t as many useful, non-overlapping channels in 5GHz WiFi as the math allows. Please check out the URL in the notes section of the slides for a nice, simple enough, article on what DFS is, how it works, and the impact on WiFI operations. The real point of this last discussion is that, contrary to the popular belief that network administrators don’t have to worry about overlapping channels in 5GHz, there are still limitations in 5GHz and the same level of RF management needed for both WiFi bands.
https://wifinigel.blogspot.com/2018/05/the-5ghz-problem-for-wi-fi-networks-dfs.html
27
6.11 2.4 GHz and 5 GHz Service Band
6-28
How would you answer these questions? For the first, 2.4GHz provides excellent speed and distance for wireless communications. For the second, 5GHz offers the potential of much greater speeds due to greater WiFi channel bandwidth.
28
What is the main advantage of 2.4 GHz operation?
What is the main advantage of 5 GHz operation?
6.12 Channels and Co-Channel Interference in Wi-Fi (2 of 2)
So I just mentioned WiFi RF management… want an example? In this slide, you see 6 WAPs operating in the 2.4GHz band. Notice how WAP E has interference with WAP D, as does WAP A & B. How would you fix the above, assuming there were no other WiFi networks within range? The simplest answer is to switch the channels for WAPs A & D resulting in no WAP being able to see any other WAP on the same non-overlapping channel (1, 6, 11). Yes, you should know that channels 1, 6, 11 are the only non-overlapping WiFi channels in the 2.4GHz band using 20 MHz channels.
29
6.13 Licensed & Unlicensed Service Bands
Licensed Radio Bands
If two nearby radio hosts transmit in the same channel, their signals will interfere
Most radio bands are licensed bands, in which hosts need a license to transmit: FM radio, etc.
The government limits licenses to control interference
In cellular telephone bands, only the central antennas are licensed, not the mobile phones
Copyright © 2015 Pearson Education, Inc.
6-30
My discussion of DFS earlier got ahead of the text and the slides. Sorry about that, but the material is pretty straight forward. Some frequencies are licensed and some are not. Licensed radio bands, like AM and FM radio, mobile phone frequencies, etc. require a government controlled license in order to operate. No one, even if they had the spare radio tower, is allowed to just start transmitting their own radio station on FM just because they want. And, yes, it is easy to find a rouge radio tower – they consume and emit a lot of energy that is easy to find with simple direction-finding equipment.
30
6.13 Licensed & Unlicensed Service Bands
Unlicensed Radio Bands
A few service bands are set aside as unlicensed bands
Hosts in these service bands do not need to be licensed to be turned on or moved
Anybody can operate anywhere
802.11 Wi-Fi operates in unlicensed radio bands
This allows access points and hosts to be moved freely
Copyright © 2015 Pearson Education, Inc.
6-31
As you can imagine, it would be horrible if we had to get a license for every WAP we wanted to install, which is why we use unlicensed radio bands for our WiFi? Which bands? Come on, now, we just went over that… 2.4 & 5 GHz.
31
6.13 Licensed & Unlicensed Service Bands
Unlicensed Radio Bands
However, there is no legal recourse against interference from other nearby users
Your only recourse is to negotiate
At the same time, you may not cause unreasonable interference by transmitting at illegally high power
6-32
Let’s go back to my earlier WiFi survey. [advance]. What if my network is conflicting with any of the 4 networks shown here? I have no legal ability to tell anyone else how to operate in an unlicensed RF band. I either have to change my WiFi channel or architect a different WiFi network, perhaps one that uses more 5GHz bands.
32
Spread Spectrum Transmission
Now, I know you want to talk about spread spectrum transmission in great detail. Unfortunately, we have so many other things to discuss in this chapter that I really can only spend a minute or so on it, at a very high level.
33
6.15 Orthogonal Frequency Division Multiplexing
OFDM is the most widely used type of spread spectrum transmission
It is easier to control signals in smaller channels than in one big channel
Also, information is sent redundantly across the channels
The textbook gives a good and simple discussion of spread spectrum transmission. What you have to know is 1) which related protocol is used in modern WiFi networks, and that is OFDM. And 2) how OFDM works at a very high level. In OFDM, information to be transmitted is split into multiple smaller chunks and transmitted independently. A common analogy is the shipment of goods using multiple small boxes compared to using only one huge box. The smaller boxes are easier to manipulate, and if some boxes are lost or damaged in transit, the receiver still gets the remaining boxes.
34
802.11 W LAN Operation
Let’s put some of these concepts together and talk about how WLANs work at a high level.
35
6.16 Packet and Frame Transmission (1 of 2)
As stressed earlier, 802.11 exists, for the most part, to extend the wired Ethernet LAN out to wireless clients. Because of this, it would seem to make sense for 802.11 to use a similar frame format as 802.3 to make communication easier between hosts on a LAN. However, as you will soon see, that is very much NOT the case.
36
6.16 Packet and Frame Transmission (2 of 2)
802.11 Frames have a different format than 802.3 frames, so cannot travel over and Ethernet network
In reality, a wireless client will use an 802.11 frame to communicate with a WAP. The WAP, which is also connected to the wired LAN, will translate the 802.11 frame into and 802.3 frame for use on the Ethernet network. Just like any other message that transits two different LANs, the data packet stays the same, but the Layer-2 frame wrapper changes to support the wired or wireless media.
37
2-38
This is an Ethernet frame, which we have already discussed. Simple enough when you know what the field functions are. Compare that to an 802.11 frame…
38
802.11 Frame Format
39
And you see that an 802.11 frame is much more complicated, which is a function of the inherent difficulties of communicating in a wireless environment compared to a closed wired LAN. Don’t worry, you will not be responsible for knowing all the field functions of an 802.11 frame. This is just here to show how more complicated the frame is in 802.11 to ensure reliable communications. Reliable? Yes, and we will get to that shortly.
39
6.17 Wi-Fi Wireless LAN with Multiple Access Points (1 of 3)
Time for some definitions. A BSS, basic service set, consists of a WAP and it’s connected wireless hosts. The SSID is the name identifier of the wireless network associated with a WAP.
40
6.17 Wi-Fi Wireless LAN with Multiple Access Points (2 of 3)
An ESS is a group of BSSs with the same SSID. Yep, that is a lot of acronyms. All of you know of at least 3 different ESS’s on the university campus – TUWpa, TUWireless, and TUGuest.
Why is the SSID the same for the different access points? Simple…
41
6.17 Wi-Fi Wireless LAN with Multiple Access Points (3 of 3)
802.11r
It’s all for you! Because we care! Honestly, how much of a pain would it be to constantly log into the same TU network using different SSID’s across campus. The protocol that allows us to seamlessly transit across SSIDs in the same ESS is 802.11r, which governs how a host is transferred between WAPs while roaming between two BSS’s.
42
6.18 Hosts and Access Points Transmit on the Same Channel (1 of 2)
Now let’s talk about a reality of WiFi that is sometimes confusing to understand. Think about sitting in a large classroom at TU with 50-100 students all connected to the same WAP. Of course none of you are paying attention to the riveting lecture on 802.11 Media Access Control that the professor worked hard to make sure was informative and just the right level of complexity. Nope, all of you are checking your Twitter feed for retweets and likes, checking Pinterest for the latest quinoa avocado breakfast recipe, or liking memes on the IG. All at the same time. All at the same time, not taking into consideration just how hard that lowest-price, 7 year-old WAP in the room is working so you don’t get salty and claim that the wireless on campus sucks. We need to care about wireless Media Access Control, however.
43
6.18 Hosts and Access Points Transmit on the Same Channel (2 of 2)
If we didn’t have wireless MAC, all the participants on a WiFi network could attempt to transmit at the same time, resulting in a confused mess of interfering radio signals, and NO ONE would be happy with the network performance because everyone would suffer. We need MAC to govern when a WiFI host and WAP will transmit so that collisions are minimized. The MAC protocol used by modern WiFi is called CSMA/CA + ACK and you need to know what it stands for and how it very basically works.
44
Box: Media Access Control
CMSA/CA + ACK sounds exciting, doesn’t it! It is!! So let’s get into it.
45
6.19: CSMA/CA+ACK
6-46
Carrier Sense Multiple Access with Collision Avoidance and Acknowledgement
Mandatory for 802.11 Wi-Fi Operation
Carrier Sensing with Multiple Access
Sender listens for traffic (senses the carrier)
If another device is transmitting, it waits
This controls access by multiple devices that must not transmit simultaneously
CSMA/CA + Ack stands for Carrier Sense Multiple Access with Collision Avoidance and Acknowledgement and it is mandatory on all 802.11 WAPs. The protocol works like its name says it does…
First, a host on a wireless network “listens” before it transmits anything. This is the Carrier Sense part. If it hears an active transmission, it backs off a set amount of time and listens again. It keeps doing this until it does not hear someone transmitting,
46
6.19: CSMA/CA+ACK
6-47
Collision Avoidance
When the current sender stops, two or more waiting devices may immediately want to transmit
This will cause a collision
Instead, the devices must wait a randomized amount of time before sending
This usually avoids collision, but it is inefficient
The host then waits a random short period of time and listens one more time (kind of like checking both ways before crossing the street) and then, if still no other device is detected, it will transmit. This is the Multiple Access part – assuming that there are other hosts that want to transmit, the extra random wait gives another host the chance to start comms. This USUALLY avoids a collision (and thus the CA – collision avoidance part of the name), but not always, and as you can tell, all this waiting is inefficient. It’s also very fast compared to how humans track time.
47
6.19: CSMA/CA+ACK
6-48
ACK (Acknowledgement) and Reliability
Receiver immediately sends back an acknowledgement
CA random delay for other devices guarantees there will be enough time for an immediate ACK
If sender does not receive the acknowledgement, it retransmits using CSMA
All communication in a wireless network boils down to one host talking to a WAP at a time, no matter how many clients are waiting or seemingly connected to the WiFi network at that time. In order to release the communication channel, the WAP (or host if receiving traffic from the WAP) sends an immediate ACK to let the transmitter know it received the data. That ACK provides reliability just like TCP; if all of the data is not ACK’d, retransmission will occur, which will happen while all other WiFi hosts are patiently waiting their turn to communicate with the WAP. And, just like TCP, if the sender does not get an ACK for any reason, it will resend the un-acknowledged data and keep doing so until it receives an ACK. Thus, even though modern wireless networks are very fast, there is a lot of inefficiency as there is lots of waiting to transmit.
It is important to note at this point that we didn’t talk about Media Access Control for Ethernet. There is a MAC protocol for Ethernet called CSMA/CD, which is the same except there is no collision avoidance, just detection. Why no avoidance an ACK for wired Ethernet? Mainly because wired communications don’t have the same myriad of wireless propagation errors and we can rely upon the Transport later to deal with any Ethernet frame errors that cause lost data. Because wireless comms is so fraught with peril, it must be made reliable between the two communicating hosts.
48
6.20: RTS-CTS
6-49
CSMA/CA + ACK is a very robust and reliable protocol and it works great. It works great IFF (if and only if) two hosts are within range of each other to determine if there is another host trying to communicate with the WAP. However, if one host is at the outer limits of a WAP’s range to the north, and another is at the outer range of the WAP to the south, they can’t hear each other and thus can not do the Carrier-sense part of CSMA/CA + ACK. What to do, what to do???? This is where RTS-CTS mode comes into play, and it take wireless efficiency to a much lower low than CSMA/CA + ACK. RTS-CTS stands for request to send, clear to send and is basically forced turn taking by wireless hosts, enforced by the WAP letting hosts know explicitly when they can transmit.
If a WAP senses that two of its hosts are probably unable to detect each others transmissions, it can fall back to RTS-CTS mode where it sends out a broadcast message to all hosts to wait until cleared before they transmit. A waiting host that wants to transmit will send an RTS message to the WAP and, based upon a number of factors in the protocol
49
6.20: RTS-CTS
6-50
… it will be granted permission to transmit with a CTS message.
50
6.20: RTS-CTS
6-51
That same CTS message tells all other hosts to shut up and wait to transmit until they send an RTS that is approved. As you can see, this is a much more arduous and time-inefficient protocol that will ensure reliable WiFi but drastically reduce speeds for all hosts on the wireless LAN.
51
Comparison
6-52
CSMA/CA is Mandatory
It is the default MAC method.
It is more efficient than RTS/CTS.
RTS/CTS
Is usually optional.
Is good if two or more client stations cannot hear each other.
Stop them from transmitting at the same time.
Copyright © 2015 Pearson Education, Inc.
In summary, 802.11 requires CSMA/CA + ACK and RTS-CTS is optional, but often configured by default. You should be able to describe both Wireless MAC protocols.
52
802.11 Transmission Standards
We spent some time on wired Lan transmission standards. I’m afraid its even more confusing for Wireless. Its OK, we’ll get through it together.
53
Today’s Main 802.11 Wi-Fi Transmission Standards
802.11n
Operates in both the 2.4 G H z and 5 G H z bands
40 M H z channels but drops back to 20 M H z if it senses interference
802.11a c
Channel bandwidths of 80 M H z or 160 M H z
Therefore much higher speeds than 802.11n
Can only operate in the 5 G H z band because will not fit in the 2.4 G H z band
However, because channels are wider, there are fewer channels in the 5 G H z service band
Although you will still some legacy 802.11g still in operation, the dominant 802.11 installations as of 2020 are 802.11n and 802.11ac. 802.11n basically pushed 802.11g out of existence because it has all the same frequencies as 802.11g but adds 5GHz and MIMO (something we will discuss shortly). However, 802.11ac does not have 2.4 GHz comms and that is part of the reason why it has not yet pushed 802.11n out of favor completely because, if you remember 2.4 GHz has pretty good propagation and speed characteristics and higher frequency WiFi comes with faster speeds but some increased propagation effects.
54
6.21 Number of Channels in the 5 G H z Unlicensed Radio Band
While I don’t expect you to memorize the data on this table, know that a big part of the increased speeds from 802.11ac comes from wider channel bandwidth, which also means less non-overlapping channels. Add in the effects of DFS (discussed earlier), and 802.11ac definitely requires you to manage WAP placement and configuration. It is not a simple set it and forget it operation.
55
6.22 Characteristics of Major 802.11 Wi-Fi Standards (1 of 3)
Characteristic 802.11n Dual Band 802.11ac
Rated Speed 100 M b p s to 600 M b p s 433 M b p s to 6.9 G b p s
Common Throughput 2018 300 M b p s 1.5 G b p s
Status Widely used Widely used and dominates sales
It is somewhat important to know the common throughput for both protocols using 2020 numbers, however.
56
6.22 Characteristics of Major 802.11 Wi-Fi Standards (2 of 3)
Characteristic 802.11n Dual Band 802.11ac
Unlicensed Band(s) 2.4 G H z and 5 G H z 5 G H z
Channel bandwidth 40 M H z, but will drop back to 20 M H z if there is interference with older 20 M H z devices 80 M H z or 160 M H z
And it is also very important to know the frequency bands and channel bandwidth options for both protocols.
57
6.24 Multiple Input/Multiple Output (MIMO)
You were told earlier that if two transmissions in the same channel will interfere. Uh, that was a lie. We now have a way to permit this, M I M O.
So earlier the text, and I mentioned that two things are bad for WiFi networks – multipath interference and data being transmitted in the same channel at the same time. And I meant it. It is bad… except when you are doing it on purpose. And for that I mean using MIMO, which is used in both 802.11n and 802.11ac.
MIMO (multiple input, multiple output) is an antenna technology for wireless communications in which multiple antennas are used at both the source (transmitter) and the destination (receiver). The antennas at each end of the communications circuit are combined to minimize errors and optimize data speed. That is an oversimplified way of explaining a lot of complex communications math and algorithms for signal processing, and it is all we need to know for this class.
The whole point of MIMO is that you get multiple independent channels due to the multipath propagation. In this case, one channel could be the direct line of sight, and one could bounce off of an adjacent wall. If the channels are sufficiently independent, different data can be sent down each channel. This means you can have greater throughput than if you had just 1 channel. And the more antennas you have at both ends, the more independent channels can be accessed, giving a higher potential throughput. As a frame of reference, an iPhone X has 2 WiFi MIMO antennas for 802.11ac (and 4 LTE MIMO antennas for cellular data). Press the I believe button and move on or let me know and I will give you more detail on how MIMO works.
58
6.25 Beamforming and Multiuser MIMO (1 of 2)
Also receives with greater sensitivity in that direction, again bringing longer range
Earlier I discussed omnidirectional vs directional antennas and how most consumer wireless devices use omnidirectional antennas. Beamforming is a method that allows a device with an omnidirectional antenna to simulate a dish-like antennae, focusing signal transmit and receiving in the direction of a particular comms partner.
59
6.25 Beamforming and Multiuser MIMO (2 of 2)
Different groups of antennas are aimed at different devices
Device only receives the collective speed of those antennas
Spatial separation permits different conversations with each
Recall that the more devices that connect to a WAP, the slower it moves. That’s because most routers can only communicate with one device at a time. With these single-user (SU-MIMO) WAPs, each device waits its turn to send and receive data, so when a new device connects, the line—and the wait—becomes a little longer. MU-MIMO allows a Wi-Fi router to communicate with multiple devices simultaneously. This decreases the time each device has to wait for a signal and dramatically speeds up your network. How does this happen? The WAP / wireless router uses beamforming using different antennas to support multiple simultaneous users. Note, that MU-MIMO is only available on 802.11ac wave 2. Huh? What’s wave 2? You’ll find out soon enough.
60
6.26 Backward Compatibility
But before we get there, understand that the reason why WiFi device manufacturers are able to push new 802.11 standards (and other protocols that might not become standards) is because they ensure that a new WAP will support, for the most part, legacy equipment of a reasonable age. This wasn’t always the case, and it led to some massive security problems with WiFi in the past, which we will discuss more in the next chapter.
61
Box: Emerging Wi-Fi Standards
Now, on to that Wave 2 thing
62
6.27 Profile Waves for Wi-Fi Devices (1 of 4)
Standards have many options
Some may be impossible or too expensive to implement initially
So Wi-Fi standards are implemented in products in successively more capable profile “waves”
Capability Wave 1 Wave 2
A A A
B B B
C Blank C
D Blank Blank
Many new protocols have a defined end state of capabilities that are not yet technologically or fiscally feasible. However, to start the process of getting to full implementation of the protocol, manufacturers roll out new devices in waves that are tied to capabilities defined in the protocol. This will make more sense if we look at 802.11ac.
63
802.11ac waves
Maximum 802.11ac standard speed is 6.9 Gbps
Wave 1 profile gives a data stream of up to 1.3 Gbps
Wave 2 profile gives a data stream of 2.5 Gbps, plus M U-M I M O
This is a great graphic that shows a comparison of 802.11ac waves 1 & 2 in comparison to the full IEEE specification for 802.11ac. Note that for each wave, the capabilities get closer to the actual specification. For example, Wave 2 improves the overall wireless throughput to up to 3.47 Gbps, allows MU-MIMO, and allows 160 MHz channels, none of which are available in 802.11ac wave 1.
64
The appetite for more bandwidth and different wireless device use-cases are driving WiFi standards ahead at a rapid pace. This graphic describes some of the new 802.11 protocol being drafted, some of which (perhaps all) will eventually become official IEEE 802.11 standards. Note that some of these emerging protocols use 60 GHz bands. Why would they do this? Well, larger bandwidth means greater speeds, up to 100 Gbps in some cases. Also, this high of a frequency would probably stay within the confines of normal office space, not leaking signals out of a work area, enhancing security as well as negating the problem of competing for frequencies with adjacent WAPs. You don’t need to know the capabilities of the draft standards provided above, but you should be aware of them as you enter into the workforce – the 802.11 protocols are changing and that will drive new opportunities and challenges.
65
Now… you are gonna hate me for this, but you shouldn’t. Who should you hate? The WiFi alliance, that is who. Remember, the WiFi Alliance is funded by a group of technology vendors, and those vendors have marketing people, and marketing people don’t care that IT people have to learn more terminology, they care that its simple enough for regular people to know WiFi “quality” easily. Thus, they came up with new nomenclature to make it easier to identify which WiFi tech is newer / faster. Now, 802.11n becomes known as WiFi-4, 802.11ac becomes WiFi-5, and 802.11ax becomes WiFi-6. Here’s the great news! You have to know these new name pairs because while your customers might use the simple naming convention, you are highly likely to have to use the IEEE nomenclature. Yay! Thanks Obama. I mean… Thanks WiFi alliance.
66
Power-over-Ethernet (PoE)
What does this have to do with wireless communications?!?
2003: 802.3af (15.4 watts)
2009: 802.3at (25.5 watts) a.k.a. PoE+
So… what is the MAIN purpose of WiFi again? Well, its really to extend the capabilities of the wired LAN to wireless clients. Well, we know we need WAPs to transmit and receive our 802.11 signals, and depending on the geometry of our spaces that need WAPs, we may have a lot of them. There is a great chance that where you need a WAP, there may not be a power supply. Running electrical power is A LOT more expensive than running CAT-6 cable. You will need an electrician to run electrical power and you, yourself, can run CAT-6 cable if you had to. We needed a way to deply WAPs without having an army of electricians at hand, and that is where PoE comes in.
As you already know, IEEE 802.3 is the standard that defines the physical layer for Ethernet cabling. It includes the type of wire, the signal levels, and the bandwidth. In 2003 the standard was updated to include PoE. IEEE 802.3af defined the voltage and power available power for devices on the network. This specification provided PoE that provided up to 15.4 watts. The specification was updated again in 2009. IEEE 802.3at provided higher powered (25.5 watts) which is referred to as PoE+. It is important to note here that not any old Ethernet switch can provide PoE – you need a switch that has additional circuitry needed to push additional energy through a CAT-6 cable. PoE switches do cost more than plain old dumb switches, but not as much as paying for an electrician, that’s for sure. Do you have to remember the 802.3 nomenclature for PoE. Nah, no one actually uses 802.3at or af in normal networking conversation. You will hear PoE all the time and most times people are really talking about PoE+. So… just make sure you know what PoE is, that it is an 802.3 protocol, and how much energy (25.5 watts) you get from PoE+. Now, if you go to work at an organization that uses a lot of PoE, you’ll want to take the time to learn all about which PoE protocol is being used and more about the
So what type of devices use PoE? Well… WAPs, of course. Also VoIP phones, too. And, IP cameras. And security card readers and other security system components. Even some thin-client computers can use PoE. But, not a server or a PC.
67
Quadrature Amplitude Modulation (QAM)
OK… last topic, but it comes with an awesome video (well… I think its awesome). We talked a very little about UTP signaling where digital and binary signaling is based upon voltages measured during a clock cycle. And in fiber optics, 1’s and 0’s are based upon the presence or absence of light. Wireless signaling is waaaaaaaay more complicate, but also way more cool. Wireless signaling encoding uses something called QAM, quadrature amplitude modulation. QAM is a form of modulation that is a combination of phase modulation and amplitude modulation. The QAM scheme represents bits as points in a quadrant grid know as a constellation map like you see on the slide. When a wireless signal is captured during a specific clock cycle, the phase and amplitude of the received signal is translated to a point on the constellation map, which represents some digital value. 802.11ac uses 256 QAM, which means a single received signal can represent any of 256 digital values (2 to the 8th), and 802.11ax uses 1024 QAM. You do not need to know the science behind how QAM works, although there are some great articles online if you are interested. You do need to know that QAM is the primary wireless signaling method used in modern WiFi networks.
68
And that ends Chapter 6, wireless LANs part 1. And there’s lots more to discuss in Chapter 7. Until the next video…. ALOHA!
69