NET 211 - Wireless Networking

Chapter 5, Physical Layer Standards

Objectives:

This lesson presents background on standards used in several wireless methods as they relate to the Physical Layer of the ISO-OSI model. Objectives important to this lesson:

1. Wireless modulation technologies
   
2. 802.11b physical layer standards
   
3. Technologies in the 801.22a Physical layer standards
   
4. 802.11g Physical layer standards
5. 802.11n Physical layer standards

Concepts:

Chapter 5

The chapter begins with some background on the creation of the ISO-OSI networking model. The way a network works can be understood in terms of this model of a network that was created by the International Organization for Standardization, called the ISO for short. (No it isn't an acronym, it is from the Greek word isos, meaning equal.) Their model is called the Open Systems Interconnect (OSI) Reference Model, hence the ISO-OSI model.

Once you understand this model, you will have a general, powerful reference for examining and comparing networks. 

The seven layers of the model are usually written in a list, numbering the top as layer seven and the bottom as layer one.

Several mnemonic sentences exist to help us remember the proper order. I recommend "Please Do Not Throw Sausage Pizza Away", because this is in the correct numeric order (bottom to top, 1 to 7). If you want one that goes from top to bottom, try "All People Studying This Need Drastic Psychotherapy".  On any certification test that covers this model, you MUST remember the correct order, the correct numbers, and the correct details for each layer.

The processes that happen in each layer communicate with the next layer. Which way is next, up or down? It depends whether data is being passed out of the stack (down) or into it (up). Typically, a computer generates a request starting at the top layer, and working down. The request is passed across the network (probably to a server) and the received request is passed up the layers. When a response is generated, the process reverses.

That's more information than your text gives you. The author introduces this networking model to tell you that we will learn about 802.11 functions that occur at Layer 1, the Physical Layer. There are other functions that take place on other layers, but this is a good place to start.

One of the topics of the Physical layer concerns how we transmit bits across the network. The text tells us there are two major types of RF modulation, but it breaks the second type into several subtypes.

• Narrowband Transmission - This method uses one frequency to transmit data. It can be completely shut down if there is strong enough interference on the same frequency. The text tells us that this method is not allowed in 802.11 standards, so it is odd that it is even mentioned.
• Spread-Spectrum Transmission - So, if 802.11 does not use narrowband, what does it use? Well, it's not exactly broadband, but the text explains that spread-spectrum uses several frequencies. There are several versions of doing this, which the text discusses, but it first gives us some reasons to prefer this general technique:
  • Resistance to narrowband interference - since narrowband uses one frequency, that is the only frequency that would be affected out of the ones a spread-spectrum system is using
  • Resistance to spread-spectrum interference - a spread-spectrum system can switch to another frequency if the one currently being used is not working well; most systems will be switching automatically anyway, as explained below
  • Lower power - a spread spectrum system is designed to use lower power than a narrowband system
  • Less interference with other systems and increased security - the author's concept is that a system of this sort is powered at a level that other systems will consider to be background noise, causing them to ignore it by design, and making harder for an eavesdropper to read; this security concept is flawed because an eavesdropper will be likely to use a spread-spectrum radio as well, which leads to the three scenarios for actual use
• Frequency-hopping spread spectrum (FHSS) - The text gives us a snapshot of history that explains how technology often evolves. The US Navy had a problem during World War II that was not a secret. German warships would jam the radio frequencies that were used to control US torpedoes.
  
  This fact was discussed by a composer, George Antheil, who knew something about controlling a number player pianos remotely, and a remarkable woman named Hedy Lamarr. She was an actress, but she was also an inventor. As the Google Doodle honoring her shows, she was not just a person having a conversation with a friend, she was a person who knew enough science to develop the concept of coordinated switching from one radio frequency to another by the torpedo and the radio controller.
  
  
  
  This is the essence of many technological innovations. Someone knows of a problem, a human pressure (as James Burke would call it) requires an answer to the problem, and that problem can be resolved by the application of an existing technology in a new way, or by adding a new concept to a technology that can work better or do something new.
  
  
  • Back to FHSS, the technology was not used in the 1940s, but became practical in the 1960s and later with electronic components. The transmitter on a system that uses FHSS may send data for about 100 milliseconds. The length of time for a transmission is called the dwell time, the time we dwell on a given frequency.
  • Every time the dwell time is reached, the system changes to the next scheduled frequency.
  • In the system invented by Ms. Lamarr and Mr. Antheil, a paper roll with control codes would signal each change. In practice in modern systems, a mathematical function known to devices on both ends of the signal will tell them when to switch, and which frequency to switch to next.
  • If there is a problem with a given frequency, the system will go to the next one. The one with a problem may not have that problem the next time it is tried.
  • FCC rules say systems in the 900 MHz band must use a selection of 50 specific frequencies, and may not dwell on any one of them longer than 400 milliseconds in every 20 seconds.
  • FCC rules say systems in the 2.4 GHz band may use 75 specific frequencies (full-channel FHSS) or may use a selection of those channels (reduced-channel FHSS). For full-channel, systems may not exceed 1 W, must hop at least 25 KHz in each change, must select the next channel randomly, must dwell the same time on average, and may not dwell on any channel longer than 400 milliseconds in every 30 seconds.
  • FHSS is less prone to interference than DSSS (see below), but its transmission rate is limited because its bandwidth is typically only 1 MHz. Its data rate is usually about 3 Mbps.
• Bluetooth - the text tells us that Bluetooth uses a modified FHSS system called Adaptive Frequency Hopping (AFH) that uses 79 channels in the 2.4 GHz range, and it hops 1600 times per second
• Direct Sequence Spread Spectrum (DSSS) - This method seeks more error control by changing each transmitted bit to a sequence of shorter transmissions. The original bits are converted to a codifying sequence called the chipping code, which is shown in a simplified version in the text.
  • The original sequence is translated into a new sequence according to the chipping code. The original bits are called chips in this system.
  • The original sequence and the chipping code translation are combined in a Boolean XOR operation.
  • The combined sequence is transmitted and decoded by the receiver, using the chipping code.
  • The text does not discuss the use of multiple frequencies. Multiple frequencies are used, but the same transmission is made on all of them.
  • DSSS has a potentially better transmission rate than FHSS, because its typical bandwidth covers about 22 MHz, and it can reach 11 Mbps.
• Orthogonal Frequency Division Multiplexing (OFDM) - The text admits that this technology does not use a spread spectrum, but it has similar qualities. It sends several parts of a message sort of at the same time, but over separate channels. That's the frequency division multiplexing part. It also incorporates a guard interval, a delay between successive transmissions on each channel.
  • In this system, transmissions on a channel are called symbols.
  • A guard interval is also used when using FHSS and DSSS, as shown on page 167. The nature of those technologies allows successive symbols to interfere with each other, which the text refers to as Intersymbol Interference (ISI).
  • OFDM, having a slower data rate, uses its guard interval to reduce ISI. The receiver is more likely to receive multiple copies of the same transmission at once, which will reinforce each other instead of interfering with each other.
  • OFDM can have up to 600 Mbps throughput on 802.11n networks, and 54 Mbps throughput on 802.11a and 802.11g networks.

   

The next section of the chapter covers eight CWNA objectives. It becomes a bit obscure on page 169 if you are not familiar with some terms. Let's hit some background first:

• On every layer in the ISO-OSI model, there is a preferred term to use for clusters of data that are sent to the next layer.
• This chapter is mostly about the Physical Layer (Layer 1), which has to do with transmitting bits.
• The next section mentions frames. Frame is a word used for a cluster of data on the Data Link Layer (Layer 2) of the network model. On page 171, the author explains that he is using the word frame to mean a cluster of data passed at either the Data Link or Physical layers.
• Notice that I made both of those layers shades of green in the table above. The reason I did that is to show that these two layers work closely together. (By the way, that's also the reason Layer 3 and 4 are shades of blue.)

The text tells us that the Data Link layer is divided into two sublayers.

• The Logical Link Control (LLC) sublayer deals with reliability and flow control.
• The Media Access Control (MAC) sublayer deals with hardware addresses and methods of access to the network.

No other networking text I have seen has mentioned that the IEEE divided the Physical layer into two sublayers for wireless LANs.

• The Physical Layer Convergence Procedure (PLCP) sublayer receives data frames from the MAC sublayer, and turns them into wireless frames that can be understood and transmitted by the PMD sublayer.
• The Physical Medium Dependent (PMD) sublayer defines the characteristics of the wireless medium and the method used to transmit and receive data.

The text explains that adding these services to the Physical layer for wireless LANs removes the need to modify services on any other layers of the ISO-OSI model. Standard wired networking services interface with wireless services here as they do with wired services on more standard networks. From this point, the chapter discusses the Physical layer standards as they apply to 802.11 b/a/g/n. (802.11b was implemented before 802.11a. Different committees, different work rates.)

802.11b

We are told that this standard uses a PLCP sublayer method based on DSSS, which we learned earlier was workable, but not the best choice at this time. The text spends several lines discussing the structure of a frame for 802.11b. The important things to know are that the frame has three main parts: the preamble, the header, and the data section.

• The preamble allows the receiver to synchronize its timing with the timing of the sender. 
• The header contains addresses and useful information about the length of the frame and error correction data.
• The data portion of the frame contains the actual data being sent.

The discussion of processes on the PMD sublayer includes a list of the 14 channels this method has available, and a warning about them. 802.11b uses the Industrial, Scientific, and Medical (ISM) portion of the 2.4 GHz band. This is an example of using an existing technology that incurs no further cost to the developer or the user.

The 14 channel frequencies are shown in a table on page 172. 802.11b transmissions are supposed to stay within 11 MHz of the channel you choose, but in reality a transmission on any channel may bleed over into two channels above it and two channels below it. This means that if you want to use two channels, they had better have four channels between them to avoid interference. And that means that out of 14 channels, you can really only use three at once: 1, 6, and 11. (In the US, operation on channel 14 is not allowed.)

802.11a

We are told that this standard uses a PLCP sublayer method based on OFDM, which is one of the factors that increases the throughput greatly over 802.11b. 802.11a uses a portion of the 5 GHz band, the Unlicensed National Information Infrastructure (UNII) part. The FCC intends this band to be used for this purpose, and has divided it into four sub-bands with confusing names. The table below from Wikipedia gives common names and formal names for each of them:

The text tells us that more devices using these RF bands are being marketed, which is causing more congestion and interference, a good reason to choose a different technology.

An improvement over 802.11b that was added to the 802.11a standard was an enhanced error correction protocol: Forward Error Correction (FEC). It transmits a second copy of each signal that can be used to match against the primary copy.

The PLCP frame for 802.11a is illustrated on page 175. Note that it has more housekeeping information in its header section, and some has been added to the data section as well.

Another difference, showing further development, is the use of four different modulation methods, depending on the data rate being used:

• 6 Mbps - Phase Shift Keying (PSK) is used. It allows two kinds of signals, which can represent 0 and 1.
• 12 Mbps - An amplitude component to PSK is added, which makes it Quadrature Phase Shift Keying (QPSK). This method uses two amplitudes and two phases, which allows four kinds of signals, which can represent 00, 01, 10, and 11.
• 24 Mbps - 16-level Quadrature Amplitude Modulation (16-QAM) adds two more phases and two more amplitudes, making a four by four matrix of possible signals. Throughput is increased because each signal can stand for two bits instead of one, as shown in the image below from Wikipedia. The signals can each represent half a byte, or four bits.
  
  
  
  
  
• 54 Mbps - 64-level Quadrature Amplitude Modulation (64-QAM) adds another two phases and another two amplitudes, making an eight by eight matrix of possible signals. Throughput is increased again because each signal can stand for six bits. You might think it would be eight, but remember that we only doubled each dimension, which means we multiplied by 2, which adds two places to the binary code each symbol can represent.

This takes us to channels used in 802.11a, another area in which this method was an improvement. 802.11a provides 23 non-overlapping channels, compared to the 3 that are in 802.11b.

The biggest problem with this method was its limited range, which was corrected in 802.11g.

802.11g

The author seems intent on describing each protocol in different terms, making it harder to understand his points, which is one reason I have provided the table above this line.

Examine the table on page 181, and the paragraph just above it, and you may understand the key points about 802.11g:

• 802.11g provides several transmission rates, making it backward compatible with 802.11b, and it retains the operating range of that technology as well.
• 802.11g provides a maximum transmission rate equal to that of 802.11a: 54 Mbps.
• 802.11g has only three non-overlapping channels, like 802.11b.
• At different transmission rates, 802.11g uses different modulation methods, like 802.11a.

802.11n

The author ends the chapter with a discussion of IEEE 802.11n. It adds new terminology, as often happens in these standards. It also adds two key features:

• The text defines spectral efficiency as the number of bits transmitted per Hz in a channel. This will be mentioned in a moment.
• Channels can be 20 or 40 MHz wide.
• It uses a variable guard interval.

The text explains that 802.11n can use two adjacent 20 MHz wide channels to make an effective 40 MHz wide channel. This method allows us to use the space that would otherwise be a buffer between those two channels, which reduces the overhead and increases the spectral efficiency. However, the protocol may have to switch to 20 MHz wide channels if there is interference with other devices, using Dynamic Frequency Selection (DFS). Some equipment can use Phased Coexistence Operation (PCO), which alternates between 20 and 40 MHz wide channels.

The image below, from securityuncorked.com, shows the situation with frequencies that may be used in 802.11ac, in several bandwidths. Note that we should never use frequencies 116 or 132, but we may be able to use the ones in between them. This makes 9 in the green bands, 11 in the taupe bands, and a possible 3 in the weather bands, making 23 total. They are all in the 5 MHz range.

The guard interval for 802.11g can be reduced from the standard 800 nanoseconds to 400 ns. This increases the rate of transmission when it occurs.

The last discussion in the chapter is about coding scheme combinations that are available in 802.11n. The text tells us that there are 77 different combinations of modulation type, convolutional coding rate (error correction), guard interval, channel width, and number of spatial streams. These combinations are called the Modulation and Coding Scheme (MCS), and each one is given an MCS index number from 0 through 76.

This is a bit optimistic as far as actual performance is concerned. The text tells us that access points are only required to support MCS 0 through 15, while mobile devices are only required to support MCS 0 through 7.