Network+ 2005 In Depth (P11) doc

The IEEE 802.11 standard specifies communication between two wireless nodes, orstations, and between a station and an access point.. Chapter 6 273 WIRELESS NETWORKS FIGURE 6-17 A WLAN wi

known as a probe, on all available channels within its frequency range When an access point

finds the probe frame, it issues a probe response This response contains all the information astation needs to associate with the access point, including a status code and station ID num-ber for that station After receiving the probe response, a station can agree to associate withthat access point The two nodes begin communicating over the frequency channel specified

by the access point

In passive scanning, a wireless station listens on all channels within its frequency range for a special signal, known as a beacon frame, issued from an access point The beacon frame con-

tains information that a wireless node requires to associate itself with the access point For

example, the frame indicates the network’s transmission rate and the SSID (Service Set tifier), a unique character string used to identify an access point After detecting a beacon

Iden-frame, the station can choose to associate with that access point The two nodes agree on a quency channel and begin communicating When setting up a WLAN, most network admin-istrators use the access point’s configuration utility to assign a unique SSID (rather than thedefault SSID provided by the manufacturer) This can contribute to better security and easiernetwork management For example, the access point used by employees in the Customer Ser-vice Department of a company could be assigned the SSID “CustSvc”

fre-Some WLANs contain multiple access points If a station detects the presence of severalaccess points, it will choose the one with the strongest signal and the lowest error rate com-

pared to other access points Notice that a station does not necessarily choose the closest access

point For instance, in the previous example, if another user brought his own access point tothe Internet café and his access point had a signal twice as strong as the café’s access point,your laptop would associate with it instead Other users’ laptops would also associate with hisaccess point (that is, unless those stations were configured to connect to one specific accesspoint, identified by its SSID in the station’s wireless connection properties)

Later, a station might choose a different access point through a process called reassociation.

This can happen if a mobile user moves out of one access point’s range and into the range ofanother, or if the initial access point is experiencing a high rate of errors On a network withmultiple access points, network managers can take advantage of the stations’ scanning feature

to automatically balance transmission loads between those access points Figure 6-17 depicts

a WLAN with multiple points

The IEEE 802.11 standard specifies communication between two wireless nodes, orstations, and between a station and an access point However, it does not specifyhow two access points should communicate Therefore, when designing an 802.11network, it is best to use access points manufactured by the same company, toensure full compatibility

TIPNET+

1.7

You have learned about some types of overhead required to manage access to the 802.11 less networks—for example, ACKs, probes, and beacons For each function, the 802.11 stan-dard specifies a frame type at the MAC sublayer These multiple frame types are divided intothree groups: control, management, and data Management frames are those involved in asso-ciation and reassociation, such as the probe and beacon frames Control frames are thoserelated to medium access and data delivery, such as the ACK and RTS/CTS frames Dataframes are those that carry the data sent between stations An 802.11 data frame is illustrated

wire-in Figure 6-18

Chapter 6 273

WIRELESS NETWORKS

FIGURE 6-17 A WLAN with multiple access points

FIGURE 6-18 Basic 802.11 MAC frame format

Compare the 802.11 data frame with the Ethernet_II data frame pictured in Figure 6-13.Notice that the wireless data frame contains four address fields, rather than two These fouraddresses are the source address, transmitter address, receiver address, and destination address.The transmitter and receiver addresses refer to the access point or another intermediary device

NET+

1.7

(if used) on the wireless network The source and destination addresses have the same ing as they do in the Ethernet_II frame.

mean-Another unique characteristic of the 802.11 data frame is its Sequence Control field Thisfield is used to indicate how a large packet is fragmented, or subdivided into smaller packetsfor more reliable delivery Recall that on wire-bound TCP/IP networks, error checking occurs

at the Transport layer of the OSI Model and packet fragmentation, if necessary, occurs at theNetwork layer However, in 802.11 networks, error checking and packet fragmentation is han-dled at the MAC sublayer of the Data Link layer By handling fragmentation at a lower layer,802.11 makes its transmission—which is less efficient and more error-prone—transparent tohigher layers This means 802.11 nodes are more easily integrated with 802.3 networks andprevent the 802.11 segments of an integrated network from slowing down the 802.3 segments.The Frame Control field in an 802.11 data frame holds information about the protocol in use,the type of frame being transmitted, whether the frame is part of a larger, fragmented packet,whether the frame is one that was reissued after an unverified delivery attempt, what type ofsecurity the frame uses, and so on Security is a significant concern with WLANs, becauseaccess points are more vulnerable than devices on a wire-bound network Wireless security isdiscussed in detail along with other network security later in this book

Although 802.11b, 802.11a, and 802.11g share all of the MAC sublayer characteristicsdescribed in the previous sections, they differ in their coding methods, frequency usage, andranges In other words, each varies at the Physical layer The following sections summarizethose differences

802.11b

In 1999, the IEEE released 802.11b, also known as “Wi-Fi,” for Wireless Fidelity 802.11b

uses DSSS (direct sequence spread spectrum) signaling Recall that in DSSS, a signal is tributed over the entire bandwidth of the allocated spectrum 802.11b uses the 2.4–2.4835-GHz frequency range (also called the 2.4-GHz band) and separates it into 14 overlapping22-MHz channels 802.11b provides a theoretical maximum of 11-Mbps throughput; actualthroughput is typically around 5 Mbps To ensure this throughput, wireless nodes must staywithin 100 meters (or approximately 330 feet) of an access point or each other, in the case of

dis-an ad-hoc network Among all the 802.11 stdis-andards, 802.11b was the first to take hold dis-andremains the most popular It is also the least expensive of all the 802.11 WLAN technologies

802.11a

Although the 802.11a task group began its standards work before the 802.11b group, 802.11a

was released after 802.11b The 802.11a standard differs from 802.11b and 802.11g in that it

uses multiple frequency bands in the 5-GHz frequency range and provides a maximum retical throughput of 54 Mbps, though its effective throughput falls generally between 11 and

theo-18 Mbps 802.11a’s high throughput is attributable to its use of higher frequencies, its uniquemethod of encoding data, and more available bandwidth Perhaps most significant is that the

NET+

1.7

5-GHz band is not as congested as the 2.4-GHz band Thus, 802.11a signals are less likely tosuffer interference from microwave ovens, cordless phones, motors, and other (incompatible)wireless LAN signals However, higher frequency signals require more power to transmit andtravel shorter distances than lower frequency signals The average geographic range for an802.11a antenna is 20 meters, or approximately 66 feet As a result, 802.11a networks require

a greater density of access points between the wire-bound LAN and wireless clients to coverthe same distance that 802.11b networks cover The additional access points, as well as thenature of 802.11a equipment, make this standard more expensive than either 802.11b or802.11g

802.11g

IEEE’s 802.11g WLAN standard is designed to be just as affordable as 802.11b while

increas-ing its maximum capacity from 11 Mbps to a maximum theoretical throughput of 54 Mbpsthrough different encoding techniques The effective throughput of 802.11g ranges generallyfrom 20 to 25 Mbps An 802.11g antenna has a geographic range of 100 meters (or approxi-mately 330 feet)

802.11g, like 802.11b, uses the 2.4-GHz frequency band In addition to its high throughput,802.11g benefits from being compatible with 802.11b networks Thus, if a network adminis-trator installed 802.11b access points on her LAN last year, this year she could add 802.11gaccess points and laptops, and the laptops could roam between the ranges of the 802.11b and802.11g access points without an interruption in service 802.11g’s compatibility with themore established 802.11b has caused many network managers to choose it over 802.11a, despite802.11a’s comparative advantages

Bluetooth

In the early 1990s, Ericsson began developing a wireless networking technology for use betweenmultiple devices, including cordless telephones, PDAs, computers, printers, keyboards, tele-phone headsets, and pagers, in a home It was designed to carry voice, video, and data signalsover the same communications channels Besides being compatible with a variety of devices,this technology was also meant to be low-cost and short-range In 1998, Intel, Nokia, Toshiba,

and IBM joined Sony Ericsson to form the Bluetooth Special Interest Group (SIG) (its

mem-bers currently number over 2000 companies), whose aim was to refine and standardize this

technology The resulting standard was named Bluetooth Bluetooth is a mobile wireless

net-working standard that uses FHSS (frequency hopping spread spectrum) RF signaling in the2.4-GHz band Recall that in FHSS, a signal hops between multiple frequencies within aband in a synchronization pattern known only to the channel’s receiver and transmitter

Bluetooth was named after King Harald I of Denmark, who ruled in the tenth century Onelegend has it that he was so fond of eating blueberries that his teeth were discolored, earninghim the nickname “Bluetooth.” This king was also famous for unifying hostile tribes from Den-mark, Norway, and Sweden, just as Bluetooth can unify disparate network nodes

Chapter 6 275

WIRELESS NETWORKS

NET+

1.7

The original Bluetooth standard, version 1.1, was designed to achieve a maximum theoreticalthroughput of 1 Mbps However, its effective throughput is 723 Kbps, with error correctionand control data consuming the remaining bandwidth The latest version of the standard, ver-sion 2.0, was released in 2004 This version uses different encoding schemes that allow Blue-tooth to achieve up to 2.1-Mbps throughput (The newer version of Bluetooth is backwardcompatible, meaning that devices running version 2.0 can communicate with devices runningearlier versions of Bluetooth.) The Bluetooth 1.1 and 1.2 standards recommend that commu-nicating nodes be spaced no farther than 10 meters (or approximately 33 feet) apart Whenusing Bluetooth version 2.0, communicating nodes can be as far as 30 meters (or approxi-mately 100 feet) apart.

Bluetooth was designed to be used on small networks composed of personal communications

devices, also known as PANs (personal area networks) An example of a WPAN (wireless PAN) is shown in Figure 6-19 Bluetooth’s relatively low throughput and short range have

made it impractical for business LANs However, due to commercial support from several ential vendors in the Bluetooth SIG, it has become a popular wireless technology for commu-nicating between cellular telephones and PDAs Bluetooth has been codified by the IEEE in

influ-their 802.15.1 standard, which describes WPAN technology.

FIGURE 6-19 A Wireless personal area network (WPAN)

NET+

1.7

A Bluetooth PAN is also known as a piconet The simplest type of piconet is one that contains

one master and one slave, which communicate in a point-to-point fashion with each other Themaster determines the frequency hopping sequence and synchronizes the communication Apiconet consisting of only two devices requires no setup As soon as two devices that are run-ning Bluetooth version 1.x (the most common scenario) come within 10 meters of each other,they can communicate For example, you might use Bluetooth to send your address data fromyour PDA to another friend’s PDA However, a piconet can be larger With Bluetooth versions1.x a piconet can contain one master and up to seven slave stations With Bluetooth 2.0, thenumber of slaves is unlimited Figure 6-20 depicts a piconet with one master and three slaves

Chapter 6 277

WIRELESS NETWORKS

FIGURE 6-20 A Bluetooth piconet

Multiple Bluetooth piconets can be combined to form a scatternet In a scatternet, each piconet

still requires a single master, but a master from one piconet can act as a slave in another piconet,

as shown in Figure 6-21 Also, a slave can participate in more than one piconet

Bluetooth was designed as a better alternative to an older form of wireless communication alsoused on PANs, infrared signaling

Infrared (IR)

Even if you don’t run a wireless network in your home, you have probably used infrared (IR)signaling there—for example, to change channels on the TV from your TV remote You mayhave noticed that the TV remote works best if you point it directly at the TV and that it does-n’t work at all if you are behind a wall in a different room That’s because in general, infraredsignals depend on a line-of-sight transmission path between the sender and receiver Just aslight can’t pass through a wall, IR signals must follow an unobstructed path between sender

NET+

1.7

and receiver (However, some IR signals will bounce off of large, angular obstacles and findtheir way from sender to receiver in a multipath fashion.) Also, IR signals used for communi-cation between computer devices travel only approximately 1 meter (or 3.3 feet) (On theother hand, IR signals from very powerful transmitters could travel hundreds of feet.) Infraredtransmission occurs at very high frequencies, in the 300- to 300,000-GHz range, and just abovethe visible spectrum of light Like Bluetooth, IR technology is relatively inexpensive IRrequires less power than Bluetooth or the 802.11 transmission technologies The most recent

IR standard allows for a maximum throughput of up to 4 Mbps, significantly faster than tooth But IR’s inability to circumnavigate physical obstacles or travel long distances have lim-ited its uses on modern networks

Blue-Nevertheless, infrared signaling remains an appropriate option for wireless communication inwhich devices can be positioned close to each other IR ports are common on computers andperipherals, and IR signaling is used to exchange data between computers, printers, PDAs,cellular telephones, and other devices For example, you might purchase a wireless keyboardthat can communicate with your computer via infrared signaling In this case, the IR port onthe wireless keyboard must be pointed toward the receiving port In the case of the keyboardshown in Figure 6-22, the wireless keyboard communicates with a wireless keyboard receiverthat is attached to the computer’s keyboard port with a cable Specifications for using infrared

signaling between devices on a network have been established by the IrDA (Infrared Data Association), a nonprofit organization founded in 1994 to develop and promote standards for

wireless communication using infrared signals IrDA is also the term used to refer to the mostpopular IR networking specifications

FIGURE 6-21 A scatternet with two piconets

NET+

1.7

To summarize what you have learned about wireless network standards, Table 6-1 lists the nificant characteristics of each standard Table 6-1 offers a comparison of the common wire-less networking standards, their ranges, and throughputs.

sig-Table 6-1 Wireless standards

Theoretical Effective Average

802.11b 2.4 GHz 11 Mbps 5 Mbps 100 meters (or

802.11a 5 GHz 54 Mbps 11–18 Mbps 20 meters (or

approximately 66 feet) 802.11g 2.4 GHz 54 Mbps 20–25 Mbps 100 meters (or

approximately 330 feet) Bluetooth 2.4 GHz 1 Mbps 723 Kbps 10 meters (or

Bluetooth 2.4 GHz 2.1 Mbps 1.5 Mbps 30 meters (or

IrDA 300–300,000 GHz 4 Mbps 3.5 Mbps 1 meter (or

Chapter Summary

◆ A physical topology is the basic physical layout of a network; it does not specifydevices, connectivity methods, or addresses on the network Physical topologies arecategorized into three fundamental geometric shapes: bus, ring, and star

◆ A bus topology consists of a single cable connecting all nodes on a network withoutintervening connectivity devices At either end of a bus network, 50-ohm resistors(terminators) stop signals after they have reached their destination Without termi-nators, signals on a bus network experience signal bounce

◆ In a ring topology, each node is connected to the two nearest nodes so that theentire network forms a circle Data is transmitted in one direction around the ring.Each workstation accepts and responds to packets addressed to it, then forwards theother packets to the next workstation in the ring

◆ In a star topology, every node on the network is connected through a central device,such as a hub Any single cable on a star network connects only two devices, so acabling problem will affect only two nodes Nodes transmit data to the hub, whichthen retransmits the information to the rest of the network segment where the desti-nation node can pick it up

◆ Few LANs use the simple physical topologies in their pure form More often, LANsemploy a hybrid of more than one simple physical topology The star-wired ringtopology uses the physical layout of a star and the token-passing data transmissionmethod Data is sent around the star in a circular pattern Token Ring networks, asspecified in IEEE 802.5, use this hybrid topology

◆ In a star-wired bus topology, groups of workstations are connected to a hub in a starformation; all the hubs are networked via a single bus This design can cover longerdistances than a simple star topology and easily interconnect or isolate different net-work segments, although it is more expensive than using either the star or bus topol-ogy alone The star-wired bus topology commonly forms the basis for Ethernet andFast Ethernet networks

◆ Hubs that service star-wired bus or star-wired ring topologies can be daisy-chained

to form a more complex hybrid topology However, daisy-chaining can only extend anetwork so far before data errors are apt to occur In this case, maximum segmentand network length limits must be carefully maintained

The actual geographic range of any wireless technology depends on several factors,including the power of the antenna, physical barriers or obstacles between sendingand receiving nodes, and interference in the environment Therefore, although a tech-nology is rated for a certain average geographic range, it may actually transmit sig-nals in a shorter or longer range

NOTENET+

1.7

◆ Network backbones may follow serial, distributed, collapsed, or parallel topologies.

In a serial topology, two or more internetworking devices are connected to each

other by a single cable in a daisy-chain fashion This is the simplest type of bone Hubs or switches are often connected in this way to extend a network

back-◆ A distributed backbone consists of a number of connectivity devices connected to aseries of central devices in a hierarchy This topology allows for easy network man-agement and scalability

◆ The collapsed backbone topology uses a router or switch as the single central nection point for multiple subnetworks This is risky, because an entire network

con-could fail if the central device fails Also, if the central connectivity device becomesovertaxed, performance on the entire network suffers

◆ A parallel backbone is the most fault-tolerant backbone topology It is a variation ofthe collapsed backbone arrangement that consists of more than one connection fromthe central router or switch to each network segment and parallel connections

between routers and switches, if more than one is present Parallel backbones are themost expensive type of backbone to implement

◆ Network logical topologies describe how signals travel over a network The two maintypes of logical topologies are bus and ring Ethernet networks use a bus logical

topology, and Token Ring networks use a ring logical topology

◆ Switching manages the filtering and forwarding of packets between nodes on a work Every network relies on one of three types of switching: circuit switching,

net-message switching, or packet switching

◆ Ethernet employs a network access method called CSMA/CD (Carrier Sense tiple Access with Collision Detection) All Ethernet networks, independent of theirspeed or frame type, use CSMA/CD

Mul-◆ On heavily trafficked Ethernet networks, collisions are common The more nodesthat are transmitting data on a network, the more collisions will take place When anEthernet network grows to a particular number of nodes, performance may suffer as

a result of collisions

◆ Switching can separate a network segment into smaller logical segments, each pendent of the other and supporting its own traffic The use of switched Ethernetincreases the effective bandwidth of a network segment because at any given timefewer workstations vie for the access to a shared channel

inde-◆ Networks may use one (or a combination) of four kinds of Ethernet data frames

Each frame type differs slightly in the way it codes and decodes packets of data fromone device to another Most modern networks rely on Ethernet_II (“DIX”) frames

◆ Token Ring networks currently run at either 4, 16, or 100 Mbps, as specified by

IEEE 802.5 Token Ring networks use the token-passing routine and a star-ring

hybrid physical topology Workstations connect to the network through MAUs

(Multistation Access Units) Token Ring networks may use shielded or unshieldedtwisted-pair cabling

Chapter 6 281

CHAPTER SUMMARY

◆ Token Ring has traditionally been more expensive to implement than Ethernet, butbecause of its token-passing routine, does not suffer collisions and offers high relia-bility and fault tolerance Few Token Ring networks remain, as Ethernet can achievehigher throughput at lower costs.

◆ FDDI (Fiber Distributed Data Interface) is a networking standard originally fied by ANSI in the mid-1980s and later refined by ISO It uses a dual fiber-opticring to transmit data at speeds of 100 Mbps FDDI’s fiber-optic cable and dual fiberrings offer greater reliability and security than twisted-pair copper wire It is muchmore expensive than Fast Ethernet

speci-◆ ATM (Asynchronous Transfer Mode) is a Data Link layer standard that relies onfixed packets, called cells, consisting of 48 bytes of data plus a 5-byte header

◆ ATM is a connection-oriented technology Its switches establish virtual circuits, orlogical point-to-point connections between sender and receiver, and then transmitdata Having a reliable connection enables ATM to guarantee QoS (quality of ser-vice) levels for designated transmissions

◆ Wireless standards vary by frequency, methods of signal, and geographic range TheIEEE 802.11 committee has specified three notable wireless standards: 802.11b,802.11a, and 802.11g All three share characteristics at the MAC sublayer level,including the CSMA/CA access method, frame formats, and methods of associationbetween access points and stations

◆ Currently, 802.11b is the most popular standard used on wireless networks Its mum throughput is 11 Mbps (though actual throughput is typically half of that).Home networks might use Bluetooth or Infrared (IR) technology, whose ranges areshorter and throughputs are lower than those of 802.11 networks

maxi-Key Terms

802.11a—The IEEE standard for a wireless networking technique that uses multiple frequency

bands in the 5-GHz frequency range and provides a theoretical maximum throughput of 54Mbps 802.11a’s high throughput, compared with 802.11b, is attributable to its use of higherfrequencies, its unique method of encoding data, and more available bandwidth

802.11b—The IEEE standard for a wireless networking technique that uses DSSS (direct

sequence spread spectrum) signaling in the 2.4–2.4835-GHz frequency range (also called the2.4-GHz band) 802.11b separates the 2.4-GHz band into 14 overlapping 22-MHz channelsand provides a theoretical maximum of 11-Mbps throughput 802.11b is also known as Wi-Fi

802.11g—The IEEE standard for a wireless networking technique designed to be compatible

with 802.11b while using different encoding techniques that allow it to reach a theoretical imum capacity of 54 Mbps 802.11g, like 802.11b, uses the 2.4-GHz frequency band

max-802.15.1—The IEEE standard for wireless personal area network (WPAN) technology,

including Bluetooth

802.3af—The IEEE standard that specifies a way of supplying electrical power over Ethernet

(PoE) 802.3af requires CAT 5 or better UTP or STP cabling and uses power sourcing ment to supply current over a wire pair to powered devices PoE is compatible with existing10BASE-T, 100BASE-TX, and 1000BASE-T implementations

equip-access method—A network’s method of controlling how nodes equip-access the communications

channel CSMA/CD (Carrier Sense Multiple Access with Collision Detection) is the accessmethod specified in the IEEE 802.3 (Ethernet) standard CSMA/CA (Carrier Sense Multi-ple Access with Collision Avoidance) is the access method specified by IEEE 802.11 (wirelessLAN) standards

active monitor—On a Token Ring network, the workstation that maintains timing for token

passing, monitors token and frame transmission, detects lost tokens, and corrects problemswhen a timing error or other disruption occurs Only one workstation on the ring can act asthe active monitor at any given time

active scanning—A method used by wireless stations to detect the presence of an access

point In active scanning, the station issues a probe to each channel in its frequency range andwaits for the access point to respond

active topology—A topology in which each workstation participates in transmitting data over

the network

association—In the context of wireless networking, the communication that occurs between

a station and an access point to enable the station to connect to the network via that accesspoint

Asynchronous Transfer Mode—See ATM.

ATM (Asynchronous Transfer Mode)—A Data Link layer technology originally conceived

in 1983 at Bell Labs, and standardized by the ITU in the mid-1990s It relies on fixed ets, called cells, that each consist of 48 bytes of data plus a 5-byte header ATM relies on vir-tual circuits and establishes a connection before sending data Having a reliable connectiontherefore allows network managers to specify QoS levels for certain types of traffic

pack-beacon frame—In the context of wireless networking, a frame issued by an access point to

alert other nodes of its existence

Bluetooth—A wireless networking standard that uses FHSS (frequency hopping spread

spec-trum) signaling in the 2.4-GHz band to achieve a maximum throughput of either 723 Kbps

or 2.1 Mbps, depending on the version Bluetooth was designed for use primarily with smalloffice or home networks in which multiple devices (including cordless phones, computers, andpagers) are connected

Bluetooth Special Interest Group (SIG)—A consortium of companies, including Sony

Eric-sson, Intel, Nokia, Toshiba, and IBM, that formally banded together in 1998 to refine and dardize Bluetooth technology

stan-bus—The single cable connecting all devices in a bus topology.

Chapter 6 283

KEY TERMS

bus topology—A topology in which a single cable connects all nodes on a network without

intervening connectivity devices

Carrier Sense Multiple Access with Collision Avoidance—See CSMA/CA.

Carrier Sense Multiple Access with Collision Detection—See CSMA/CD.

cell—A packet of a fixed size In ATM technology, a cell consists of 48 bytes of data plus a

5-byte header

circuit switching—A type of switching in which a connection is established between two

net-work nodes before they begin transmitting data Bandwidth is dedicated to this connection andremains available until users terminate the communication between the two nodes

collapsed backbone—A type of backbone that uses a router or switch as the single central

con-nection point for multiple subnetworks

collision—In Ethernet networks, the interference of one network node’s data transmission with

another network node’s data transmission

collision domain—The portion of an Ethernet network in which collisions could occur if two

nodes transmit data at the same time

CSMA/CA (Carrier Sense Multiple Access with Collision Avoidance)—A network access

method used on 802.11 wireless networks In CSMA/CA, before a node begins to send data

it checks the medium If it detects no transmission activity, it waits a brief, random amount oftime, and then sends its transmission If the node does detect activity, it waits a brief period oftime before checking the channel again CSMA/CA does not eliminate, but minimizes, thepotential for collisions

CSMA/CD (Carrier Sense Multiple Access with Collision Detection)—A network access

method specified for use by IEEE 802.3 (Ethernet) networks In CSMA/CD, each node waitsits turn before transmitting data, to avoid interfering with other nodes’ transmissions If a node’sNIC determines that its data has been involved in a collision, it immediately stops transmit-ting Next, in a process called jamming, the NIC issues a special 32-bit sequence that indicates

to the rest of the network nodes that its previous transmission was faulty and that those dataframes are invalid After waiting, the NIC determines if the line is again available; if it is avail-able, the NIC retransmits its data

daisy chain—A group of connectivity devices linked together in a serial fashion.

data propagation delay—The length of time data takes to travel from one point on the

seg-ment to another point On Ethernet networks, CSMA/CD’s collision detection routine not operate accurately if the data propagation delay is too long

can-DB-9 connector—A connector containing nine pins that is used on STP-based Token Ring

networks

distributed backbone—A type of backbone in which a number of connectivity devices

(usu-ally hubs) are connected to a series of central connectivity devices, such as hubs, switches, orrouters, in a hierarchy

enterprise—An entire organization, including local and remote offices, a mixture of computer

systems, and a number of departments Enterprise-wide computing takes into account thebreadth and diversity of a large organization’s computer needs

Ethernet_II—The original Ethernet frame type developed by Digital, Intel, and Xerox, before

the IEEE began to standardize Ethernet Ethernet_II contains a 2-byte type field to identifythe upper-layer protocol contained in the frame It supports TCP/IP, AppleTalk, IPX/SPX,and other higher-layer protocols

FDDI (Fiber Distributed Data Interface)—A networking standard originally specified by

ANSI in the mid-1980s and later refined by ISO FDDI uses a dual fiber-optic ring to mit data at speeds of 100 Mbps It was commonly used as a backbone technology in the 1980sand early 1990s, but lost favor as Fast Ethernet technologies emerged in the mid-1990s.FDDI provides excellent reliability and security

trans-Fiber Distributed Data Interface—See FDDI.

High-Speed Token Ring—See HSTR.

HSTR (High-Speed Token Ring)—A standard for Token Ring networks that operate at 100

Mbps

hybrid topology—A physical topology that combines characteristics of more than one simple

physical topology

Infrared Data Association—See IrDA.

IrDA (Infrared Data Association)—A nonprofit organization founded in 1994 to develop and

promote standards for wireless communication using infrared signals IrDA is also used todenote the type of wireless technology this group has developed

jamming—A part of CSMA/CD in which, upon detecting a collision, a station issues a

spe-cial 32-bit sequence to indicate to all nodes on an Ethernet segment that its previously mitted frame has suffered a collision and should be considered faulty

trans-LAN Emulation—See trans-LANE.

LANE (LAN Emulation)—A method for transporting Token Ring or Ethernet frames over

ATM networks LANE encapsulates incoming Ethernet or Token Ring frames, then convertsthem into ATM cells for transmission over an ATM network

logical topology—A characteristic of network transmission that reflects the way in which

data is transmitted between nodes (which may differ from the physical layout of the paths thatdata takes) The most common logical topologies are bus and ring

message switching—A type of switching in which a connection is established between two

devices in the connection path; one device transfers data to the second device, then breaks theconnection The information is stored and forwarded from the second device after a connec-tion between that device and a third device on the path is established

network access method—See access method.

Chapter 6 285

KEY TERMS

packet switching—A type of switching in which data is broken into packets before it is

trans-ported In packet switching, packets can travel any path on the network to their destination,because each packet contains a destination address and sequencing information

padding—The bytes added to the data (or information) portion of an Ethernet frame to ensure

this field is at least 46 bytes in size Padding has no effect on the data carried by the frame

PAN (personal area network)—A small (usually home) network composed of personal

com-munications devices

parallel backbone—A type of backbone that consists of more than one connection from the

central router or switch to each network segment

passive scanning—In the context of wireless networking, the process in which a station

lis-tens to several channels within a frequency range for a beacon issued by an access point

PD (powered device)—On a network using Power over Ethernet, a node that receives power

from power sourcing equipment

personal area network—See PAN.

physical topology—The physical layout of a network A physical topology depicts a network

in broad scope; it does not specify devices, connectivity methods, or addresses on the network.Physical topologies are categorized into three fundamental geometric shapes: bus, ring, and star.These shapes can be mixed to create hybrid topologies

piconet—A PAN (personal area network) that relies on Bluetooth transmission technology PoE (Power over Ethernet)—A method of delivering current to devices using Ethernet con-

nection cables

Power over Ethernet—See PoE.

power sourcing equipment—See PSE.

powered device—See PD.

preamble—The field in an Ethernet frame that signals to the receiving node that data is

incoming and indicates when the data flow is about to begin

probe—In 802.11 wireless networking, a type of frame issued by a station during active

scan-ning to find nearby access points

PSE (power sourcing equipment)—On a network using Power over Ethernet, the device that

supplies power to end nodes

quality of service (QoS)—The result of standards for delivering data within a certain period

of time after their transmission For example, ATM networks can supply four QoS levels, from

a “best effort” attempt for noncritical data to a guaranteed, real-time transmission for sensitive data

time-reassociation—In the context of wireless networking, the process of a station establishing a

connection (or associating) with a different access point