Data Communications

Chapter 16
High Speed LANs
Introduction

 Range of technologies
 Fast and Gigabit Ethernet
 Fibre Channel
 High Speed Wireless LANs
Why High Speed LANs?
 Office LANs used to provide basic connectivity
 Connecting PCs and terminals to mainframes and midrange systems
that ran corporate applications
 Providing workgroup connectivity at departmental level
 Traffic patterns light
 Emphasis on file transfer and electronic mail
 Speed and power of PCs has risen
 Graphics-intensive applications and GUIs
 MIS organizations recognize LANs as essential
 Began with client/server computing
 Now dominant architecture in business environment
 Intranetworks
 Frequent transfer of large volumes of data 
Applications Requiring High Speed
LANs
 Centralized server farms
 User needs to draw huge amounts of data from multiple centralized servers
 E.g. Color publishing
 Servers contain tens of gigabytes of image data
 Downloaded to imaging workstations
 Power workgroups
 Small number of cooperating users
 Draw massive data files across network
 E.g. Software development group testing new software version or
computer-aided design (CAD) running simulations
 High-speed local backbone
 Processing demand grows
 LANs proliferate at site
 High-speed interconnection is necessary
Ethernet (CSMA/CD)

 Carriers Sense Multiple Access with Collision

Detection
 Xerox - Ethernet
 IEEE 802.3
IEEE802.3 Medium Access Control

 Random Access
 Stations access medium randomly
 Contention
 Stations content for time on medium
ALOHA
 Packet Radio
 When station has frame, it sends
 Station listens (for max round trip time)plus small increment
 If ACK, fine. If not, retransmit
 If no ACK after repeated transmissions, give up
 Frame check sequence (as in HDLC)
 If frame OK and address matches receiver, send ACK
 Frame may be damaged by noise or by another station
transmitting at the same time (collision)
 Any overlap of frames causes collision
 Max utilization 18%
Slotted ALOHA

 Time in uniform slots equal to frame transmission

time
 Need central clock (or other sync mechanism)
 Transmission begins at slot boundary
 Frames either miss or overlap totally
 Max utilization 37%
CSMA
 Propagation time is much less than transmission time
 All stations know that a transmission has started almost
immediately
 First listen for clear medium (carrier sense)
 If medium idle, transmit
 If two stations start at the same instant, collision
 Wait reasonable time (round trip plus ACK contention)
 No ACK then retransmit
 Max utilization depends on propagation time (medium length)
and frame length
 Longer frame and shorter propagation gives better utilization
Nonpersistent CSMA
 If medium is idle, transmit; otherwise, go to 2  
 If medium is busy, wait amount of time drawn from
probability distribution (retransmission delay) and
repeat 1
 Random delays reduces probability of collisions
 Consider two stations become ready to transmit at same time
 While another transmission is in progress
 If both stations delay same time before retrying, both will
attempt to transmit at same time
 Capacity is wasted because medium will remain idle
following end of transmission
 Even if one or more stations waiting
 Nonpersistent stations deferential
1-persistent CSMA
 To avoid idle channel time, 1-persistent protocol used
 Station wishing to transmit listens and obeys following: 
1. If medium idle, transmit; otherwise, go to step 2
2. If medium busy, listen until idle; then transmit
immediately
 1-persistent stations selfish
 If two or more stations waiting, collision guaranteed
 Gets sorted out after collision
P-persistent CSMA
 Compromise that attempts to reduce collisions
 Like nonpersistent
 And reduce idle time
 Like1-persistent
 Rules:
1. If medium idle, transmit with probability p, and delay one time
unit with probability (1 – p)
 Time unit typically maximum propagation delay
2. If medium busy, listen until idle and repeat step 1
3. If transmission is delayed one time unit, repeat step 1
 What is an effective value of p?
Value of p?
 Avoid instability under heavy load
 n stations waiting to send
 End of transmission, expected number of stations attempting to transmit is
number of stations ready times probability of transmitting
 np
 If np > 1on average there will be a collision
 Repeated attempts to transmit almost guaranteeing more collisions
 Retries compete with new transmissions
 Eventually, all stations trying to send
 Continuous collisions; zero throughput
 So np < 1 for expected peaks of n
 If heavy load expected, p small
 However, as p made smaller, stations wait longer
 At low loads, this gives very long delays
CSMA Picture HERE

 NEEDS EDITING fig 16.1

CSMA/CD

 With CSMA, collision occupies medium for

duration of transmission
 Stations listen whilst transmitting

1. If medium idle, transmit, otherwise, step 2

2. If busy, listen for idle, then transmit
3. If collision detected, jam then cease transmission
4. After jam, wait random time then start from step
1
CSMA/CD
Operation
Which Persistence Algorithm?
 IEEE 802.3 uses 1-persistent
 Both nonpersistent and p-persistent have
performance problems
 1-persistent (p = 1) seems more unstable than p-
persistent
 Greed of the stations
 But wasted time due to collisions is short (if frames long
relative to propagation delay
 With random backoff, unlikely to collide on next tries
 To ensure backoff maintains stability, IEEE 802.3 and
Ethernet use binary exponential backoff
Binary Exponential Backoff
 Attempt to transmit repeatedly if repeated collisions
 First 10 attempts, mean value of random delay doubled
 Value then remains same for 6 further attempts
 After 16 unsuccessful attempts, station gives up and reports error
 As congestion increases, stations back off by larger amounts to
reduce the probability of collision.
 1-persistent algorithm with binary exponential backoff efficient
over wide range of loads
 Low loads, 1-persistence guarantees station can seize channel once idle
 High loads, at least as stable as other techniques
 Backoff algorithm gives last-in, first-out effect
 Stations with few collisions transmit first
Collision Detection
 On baseband bus, collision produces much higher
signal voltage than signal
 Collision detected if cable signal greater than single
station signal
 Signal attenuated over distance
 Limit distance to 500m (10Base5) or 200m
(10Base2)
 For twisted pair (star-topology) activity on more
than one port is collision
 Special collision presence signal
IEEE 802.3 Frame Format
10Mbps Specification (Ethernet)
 <data rate><Signaling method><Max segment length>

 10Base5 10Base2 10Base-T 10Base-F

 Medium Coaxial Coaxial UTP 850nm fiber

 Signaling Baseband Baseband Baseband Manchester
 Manchester Manchester Manchester On/Off
 Topology Bus Bus Star Star
 Nodes 100 30 - 33
100Mbps Fast Ethernet
 Use IEEE 802.3 MAC protocol and frame format
 100BASE-X use physical medium specifications from FDDI
 Two physical links between nodes
 Transmission and reception
 100BASE-TX uses STP or Cat. 5 UTP
 May require new cable
 100BASE-FX uses optical fiber
 100BASE-T4 can use Cat. 3, voice-grade UTP
 Uses four twisted-pair lines between nodes
 Data transmission uses three pairs in one direction at a time
 Star-wire topology
 Similar to 10BASE-T
100Mbps (Fast Ethernet)
 100Base-TX 100Base-FX 100Base-T4

 2 pair, STP 2 pair, Cat 5 UTP 2 optical fiber 4 pair, cat 3,4,5
 MLT-3 MLT-3 4B5B,NRZI 8B6T,NRZ
100BASE-X Data Rate and Encoding

 Unidirectional data rate 100 Mbps over single link

 Single twisted pair, single optical fiber
 Encoding scheme same as FDDI
 4B/5B-NRZI
 Modified for each option
100BASE-X Media
 Two physical medium specifications
 100BASE-TX
 Two pairs of twisted-pair cable
 One pair for transmission and one for reception
 STP and Category 5 UTP allowed
 The MTL-3 signaling scheme is used
 100BASE-FX
 Two optical fiber cables
 One for transmission and one for reception
 Intensity modulation used to convert 4B/5B-NRZI code group stream
into optical signals
 1 represented by pulse of light
 0 by either absence of pulse or very low intensity pulse 
100BASE-T4
 100-Mbps over lower-quality Cat 3 UTP
 Taking advantage of large installed base
 Cat 5 optional
 Does not transmit continuous signal between packets
 Useful in battery-powered applications
 Can not get 100 Mbps on single twisted pair
 Data stream split into three separate streams
 Each with an effective data rate of 33.33 Mbps
 Four twisted pairs used
 Data transmitted and received using three pairs
 Two pairs configured for bidirectional transmission
 NRZ encoding not used
 Would require signaling rate of 33 Mbps on each pair
 Does not provide synchronization
 Ternary signaling scheme (8B6T)
100BASE-T Options
Full Duplex Operation
 Traditional Ethernet half duplex
 Either transmit or receive but not both simultaneously
 With full-duplex, station can transmit and receive
simultaneously
 100-Mbps Ethernet in full-duplex mode, theoretical transfer rate
200 Mbps
 Attached stations must have full-duplex adapter cards
 Must use switching hub
 Each station constitutes separate collision domain
 In fact, no collisions
 CSMA/CD algorithm no longer needed
 802.3 MAC frame format used
 Attached stations can continue CSMA/CD
Mixed Configurations

 Fast Ethernet supports mixture of existing 10-Mbps

LANs and newer 100-Mbps LANs
 E.g. 100-Mbps backbone LAN to support 10-Mbps hubs
 Stations attach to 10-Mbps hubs using 10BASE-T
 Hubs connected to switching hubs using 100BASE-T
 Support 10-Mbps and 100-Mbps
 High-capacity workstations and servers attach directly to 10/100
switches
 Switches connected to 100-Mbps hubs using 100-Mbps links
 100-Mbps hubs provide building backbone
 Connected to router providing connection to WAN
Gigabit Ethernet Configuration
Gigabit Ethernet - Differences

 Carrier extension
 At least 4096 bit-times long (512 for 10/100)
 Frame bursting
Gigabit Ethernet – Physical
 1000Base-SX
 Short wavelength, multimode fiber
 1000Base-LX
 Long wavelength, Multi or single mode fiber
 1000Base-CX
 Copper jumpers <25m, shielded twisted pair
 1000Base-T
 4 pairs, cat 5 UTP

 Signaling - 8B/10B
Gbit Ethernet Medium Options
(log scale)
10Gbps Ethernet - Uses
 High-speed, local backbone interconnection between large-capacity
switches
 Server farm
 Campus wide connectivity
 Enables Internet service providers (ISPs) and network service providers
(NSPs) to create very high-speed links at very low cost
 Allows construction of (MANs) and WANs
 Connect geographically dispersed LANs between campuses or points of

presence (PoPs)
 Ethernet competes with ATM and other WAN technologies
 10-Gbps Ethernet provides substantial value over ATM
10Gbps Ethernet - Advantages
 No expensive, bandwidth-consuming conversion between
Ethernet packets and ATM cells
 Network is Ethernet, end to end
 IP and Ethernet together offers QoS and traffic policing
approach ATM
 Advanced traffic engineering technologies available to
users and providers
 Variety of standard optical interfaces (wavelengths and
link distances) specified for 10 Gb Ethernet
 Optimizing operation and cost for LAN, MAN, or WAN 
10Gbps Ethernet - Advantages
 Maximum link distances cover 300 m to 40 km
 Full-duplex mode only
 10GBASE-S (short):
 850 nm on multimode fiber
 Up to 300 m
 10GBASE-L (long)
 1310 nm on single-mode fiber
 Up to 10 km
 10GBASE-E (extended)
 1550 nm on single-mode fiber
 Up to 40 km
 10GBASE-LX4:
 1310 nm on single-mode or multimode fiber
 Up to 10 km
 Wavelength-division multiplexing (WDM) bit stream across four light waves
10Gbps Ethernet Distance Options
(log scale)
Token Ring (802.5)

 Developed from IBM's commercial token ring

 Because of IBM's presence, token ring has gained
broad acceptance
 Never achieved popularity of Ethernet
 Currently, large installed base of token ring
products
 Market share likely to decline
Ring Operation
 Each repeater connects to two others via unidirectional
transmission links
 Single closed path
 Data transferred bit by bit from one repeater to the next
 Repeater regenerates and retransmits each bit
 Repeater performs data insertion, data reception, data
removal
 Repeater acts as attachment point
 Packet removed by transmitter after one trip round ring
Listen State Functions

 Scan passing bit stream for patterns

 Address of attached station
 Token permission to transmit
 Copy incoming bit and send to attached station
 Whilst forwarding each bit
 Modify bit as it passes
 e.g. to indicate a packet has been copied (ACK)
Transmit State Functions

 Station has data

 Repeater has permission
 May receive incoming bits
 If ring bit length shorter than packet
 Pass back to station for checking (ACK)
 May be more than one packet on ring
 Buffer for retransmission later
Bypass State

 Signals propagate past repeater with no delay

(other than propagation delay)
 Partial solution to reliability problem (see later)
 Improved performance
Ring Repeater States
802.5 MAC Protocol
 Small frame (token) circulates when idle
 Station waits for token
 Changes one bit in token to make it SOF for data frame
 Append rest of data frame
 Frame makes round trip and is absorbed by transmitting
station
 Station then inserts new token when transmission has
finished and leading edge of returning frame arrives
 Under light loads, some inefficiency
 Under heavy loads, round robin
Token Ring
Operation
Dedicated Token Ring

 Central hub
 Acts as switch
 Full duplex point to point link
 Concentrator acts as frame level repeater
 No token passing
802.5 Physical Layer
 Data Rate 4 16 100
 Medium UTP,STP,Fiber
 Signaling Differential Manchester
 Max Frame 4550 18200 18200
 Access Control TP or DTR TP or DTR DTR

 Note: 1Gbit specified in 2001

 Uses 802.3 physical layer specification
Fibre Channel - Background
 I/O channel
 Direct point to point or multipoint comms link
 Hardware based
 High Speed
 Very short distance
 User data moved from source buffer to destiation buffer
 Network connection
 Interconnected access points
 Software based protocol
 Flow control, error detection &recovery
 End systems connections
Fibre Channel
 Best of both technologies
 Channel oriented
 Data type qualifiers for routing frame payload
 Link level constructs associated with I/O ops
 Protocol interface specifications to support existing I/O
architectures
 e.g. SCSI
 Network oriented
 Full multiplexing between multiple destinations
 Peer to peer connectivity
 Internetworking to other connection technologies
Fibre Channel Requirements
 Full duplex links with two fibers per link
 100 Mbps to 800 Mbps on single line
 Full duplex 200 Mbps to 1600 Mbps per link

 Up to 10 km
 Small connectors
 High-capacity utilization, distance insensitivity
 Greater connectivity than existing multidrop channels
 Broad availability
 i.e. standard components

 Multiple cost/performance levels

 Small systems to supercomputers

 Carry multiple existing interface command sets for existing channel and network
protocols 
 Uses generic transport mechanism based on point-to-point links and a switching
network
 Supports simple encoding and framing scheme
 In turn supports a variety of channel and network protocols
Fibre Channel Elements

 End systems - Nodes

 Switched elements - the network or fabric
 Communication across point to point links
Fibre Channel Network
Fibre Channel Protocol Architecture
(1)
 FC-0 Physical Media
 Optical fiber for long distance
 coaxial cable for high speed short distance
 STP for lower speed short distance
 FC-1 Transmission Protocol
 8B/10B signal encoding
 FC-2 Framing Protocol
 Topologies
 Framing formats
 Flow and error control
 Sequences and exchanges (logical grouping of frames)
Fibre Channel Protocol Architecture
(2)
 FC-3 Common Services
 Including multicasting
 FC-4 Mapping
 Mapping of channel and network services onto fibre
channel
 e.g. IEEE 802, ATM, IP, SCSI
Fibre Channel Physical Media

 Provides range of options for physical medium,

the data rate on medium, and topology of network
 Shielded twisted pair, video coaxial cable, and
optical fiber
 Data rates 100 Mbps to 3.2 Gbps
 Point-to-point from 33 m to 10 km
Fibre Channel Fabric
 General topology called fabric or switched topology
 Arbitrary topology includes at least one switch to
interconnect number of end systems
 May also consist of switched network
 Some of these switches supporting end nodes
 Routing transparent to nodes
 Each port has unique address
 When data transmitted into fabric, edge switch to which node attached
uses destination port address to determine location
 Either deliver frame to node attached to same switch or transfers frame
to adjacent switch to begin routing to remote destination
Fabric Advantages
 Scalability of capacity
 As additional ports added, aggregate capacity of network increases
 Minimizes congestion and contention
 Increases throughput
 Protocol independent
 Distance insensitive
 Switch and transmission link technologies may change
without affecting overall configuration
 Burden on nodes minimized
 Fibre Channel node responsible for managing point-to-point
connection between itself and fabric
 Fabric responsible for routing and error detection
Alternative Topologies
 Point-to-point topology
 Only two ports
 Directly connected, with no intervening switches
 No routing
 Arbitrated loop topology
 Simple, low-cost topology
 Up to 126 nodes in loop
 Operates roughly equivalent to token ring
 Topologies, transmission media, and data rates may
be combined
Five Applications of Fibre Channel
Fibre Channel Prospects
 Backed by Fibre Channel Association
 Interface cards for different applications available
 Most widely accepted as peripheral device interconnect
 To replace such schemes as SCSI
 Technically attractive to general high-speed LAN
requirements
 Must compete with Ethernet and ATM LANs
 Cost and performance issues should dominate the
consideration of these competing technologies
Data Communications

Chapter 17
Wireless LANs
Overview

 A wireless LAN uses wireless transmission

medium
 Used to have high prices, low data rates,
occupational safety concerns, and licensing
requirements
 Problems have been addressed
 Popularity of wireless LANs has grown rapidly
Applications - LAN Extension
 Saves installation of LAN cabling
 Eases relocation and other modifications to network structure
 However, increasing reliance on twisted pair cabling for LANs
 Most older buildings already wired with Cat 3 cable
 Newer buildings are prewired with Cat 5
 Wireless LAN to replace wired LANs has not happened
 In some environments, role for the wireless LAN
 Buildings with large open areas
 Manufacturing plants, stock exchange trading floors, warehouses
 Historical buildings
 Small offices where wired LANs not economical
 May also have wired LAN
 Servers and stationary workstations
Single Cell Wireless LAN Configuration
Multi-Cell Wireless LAN Configuration
Applications –
Cross-Building Interconnect
 Connect LANs in nearby buildings
 Point-to-point wireless link
 Connect bridges or routers
 Not a LAN per se
 Usual to include this application under heading of
wireless LAN
Applications - Nomadic Access

 Link between LAN hub and mobile data terminal

 Laptop or notepad computer
 Enable employee returning from trip to transfer data
from portable computer to server
 Also useful in extended environment such as
campus or cluster of buildings
 Users move around with portable computers
 May wish access to servers on wired LAN
Infrastructure Wireless LAN
Applications –
Ad Hoc Networking
 Peer-to-peer network
 Set up temporarily to meet some immediate need
 E.g. group of employees, each with laptop or
palmtop, in business or classroom meeting
 Network for duration of meeting
Add Hoc LAN
Wireless LAN Requirements
 Same as any LAN
 High capacity, short distances, full connectivity, broadcast capability
 Throughput: efficient use wireless medium
 Number of nodes:Hundreds of nodes across multiple cells
 Connection to backbone LAN: Use control modules to connect to both types of
LANs
 Service area: 100 to 300 m
 Low power consumption:Need long battery life on mobile stations
 Mustn't require nodes to monitor access points or frequent handshakes
 Transmission robustness and security:Interference prone and easily eavesdropped
 Collocated network operation:Two or more wireless LANs in same area
 License-free operation
 Handoff/roaming: Move from one cell to another
 Dynamic configuration: Addition, deletion, and relocation of end systems without
disruption to users
Technology
 Infrared (IR) LANs: Individual cell of IR LAN limited
to single room
 IR light does not penetrate opaque walls
 Spread spectrum LANs: Mostly operate in ISM
(industrial, scientific, and medical) bands
 No Federal Communications Commission (FCC) licensing
is required in USA
 Narrowband microwave: Microwave frequencies but
not use spread spectrum
 Some require FCC licensing
Infrared LANs Strengths and Weaknesses
 Spectrum virtually unlimited
 Infrared spectrum is unregulated worldwide
 Extremely high data rates
 Infrared shares some properties of visible light
 Diffusely reflected by light-colored objects
 Use ceiling reflection to cover entire room
 Does not penetrate walls or other opaque objects
 More easily secured against eavesdropping than microwave
 Separate installation in every room without interference
 Inexpensive and simple
 Uses intensity modulation, so receivers need to detect only amplitude
 Background radiation
 Sunlight, indoor lighting
 Noise, requiring higher power and limiting range
 Power limited by concerns of eye safety and power consumption
Infrared LANs Transmission Techniques
 Directed-beam IR
 Point-to-point links
 Range depends on power and focusing
 Can be kilometers
 Used for building interconnect within line of sight
 Indoor use to set up token ring LAN
 IR transceivers positioned so that data circulate in ring
 Omnidirectional
 Single base station within line of sight of all other stations
 Typically, mounted on ceiling
 Acts as a multiport repeater
 Other transceivers use directional beam aimed at ceiling unit
 Diffused configuration
 Transmitters are focused and aimed at diffusely reflecting ceiling
Spread Spectrum LANs Hub Configuration
 Usually use multiple-cell arrangement
 Adjacent cells use different center frequencies
 Hub is typically mounted on ceiling
 Connected to wired LAN
 Connect to stations attached to wired LAN and in other cells
 May also control access
 IEEE 802.11 point coordination function
 May also act as multiport repeater
 Stations transmit to hub and receive from hub
 Stations may broadcast using an omnidirectional antenna
 Logical bus configuration
 Hub may do automatic handoff
 Weakening signal, hand off
Spread Spectrum LANs
Peer-to-Peer Configuration
 No hub
 MAC algorithm such as CSMA used to control
access
 Ad hoc LANs
Spread Spectrum LANs Transmission Issues
 Licensing regulations differ from one country to another
 USA FCC authorized two unlicensed applications within the ISM
band:
 Spread spectrum - up to 1 watt
 Very low power systems- up to 0.5 watts
 902 - 928 MHz (915-MHz band)
 2.4 - 2.4835 GHz (2.4-GHz band)
 5.725 - 5.825 GHz (5.8-GHz band)
 2.4 GHz also in Europe and Japan
 Higher frequency means higher potential bandwidth
 Interference
 Devices at around 900 MHz, including cordless telephones, wireless microphones, and
amateur radio
 Fewer devices at 2.4 GHz; microwave oven
 Little competition at 5.8 GHz
 Higher frequency band, more expensive equipment
Narrow Band Microwave LANs

 Just wide enough to accommodate signal

 Until recently, all products used licensed band
 At least one vendor has produced LAN product in
ISM band
Licensed Narrowband RF
 Microwave frequencies usable for voice, data, and video licensed within
specific geographic areas to avoid interference
 Radium 28 km
 Can contain five licenses
 Each covering two frequencies
 Motorola holds 600 licenses (1200 frequencies) in the 18-GHz range
 Cover all metropolitan areas with populations of 30,000 or more in USA
 Use of cell configuration
 Adjacent cells use nonoverlapping frequency bands
 Motorola controls frequency band
 Can assure nearby independent LANs do not interfere
 All transmissions are encrypted
 Licensed narrowband LAN guarantees interference-free communication
 License holder has legal right tointerference-free data channel
Unlicensed Narrowband RF
 1995, RadioLAN introduced narrowband wireless LAN using
unlicensed ISM spectrum
 Used for narrowband transmission at low power
 0.5 watts or less
 Operates at 10 Mbps
 5.8-GHz band
 50 m in semiopen office and 100 m in open office
 Peer-to-peer configuration
 Elects one node as dynamic master
 Based on location, interference, and signal strength
 Master can change automatically as conditions change
 Includes dynamic relay function
 Stations can act as repeater to move data between stations that are out of
range of each other
IEEE 802.11 - BSS
 MAC protocol and physical medium specification for wireless
LANs
 Smallest building block is basic service set (BSS)
 Number of stations
 Same MAC protocol
 Competing for access to same shared wireless medium
 May be isolated or connect to backbone distribution system (DS)
through access point (AP)
 AP functions as bridge
 MAC protocol may be distributed or controlled by central
coordination function in AP
 BSS generally corresponds to cell
 DS can be switch, wired network, or wireless network
BSS Configuration

 Simplest: each station belongs to single BSS

 Within range only of other stations within BSS
 Can have two BSSs overlap
 Station could participate in more than one BSS
 Association between station and BSS dynamic
 Stations may turn off, come within range, and go out of
range
Extended Service Set (ESS)

 Two or more BSS interconnected by DS

 Typically, DS is wired backbone but can be any
network
 Appears as single logical LAN to LLC
Access Point (AP)

 Logic within station that provides access to DS

 Provides DS services in addition to acting as station
 To integrate IEEE 802.11 architecture with wired
LAN, portal used
 Portal logic implemented in device that is part of
wired LAN and attached to DS
 E.g. Bridge or router
IEEE 802.11 Architecture
Services
Service Provider Category
Association Distribution system MSDU delivery
Authentication Station LAN access and security

Deauthentication Station LAN access and security

Dissassociation Distribution system MSDU delivery

Distribution Distribution system MSDU delivery
Integration Distribution system MSDU delivery
MSDU delivery Station MSDU delivery
Privacy Station LAN access and security

Reassocation Distribution system MSDU delivery

Categorizing Services
 Station services implemented in every 802.11 station
 Including AP stations
 Distribution services provided between BSSs
 May be implemented in AP or special-purpose device
 Three services used to control access and confidentiality
 Six services used to support delivery of MAC service data
units (MSDUs) between stations
 Block of data passed down from MAC user to MAC layer
 Typically LLC PDU
 If MSDU too large for MAC frame, fragment and transmit in series of
frames (see later)
Distribution of Messages
Within a DS
 Distribution is primary service used by stations to
exchange MAC frames when frame must traverse DS
 From station in one BSS to station in another BSS
 Transport of message through DS is beyond scope of 802.11
 If stations within same BSS, distribution service logically goes through
single AP of that BSS
 Integration service enables transfer of data between
station on 802.11 LAN and one on an integrated 802.x
LAN
 Integrated refers to wired LAN physically connected to DS
 Stations may be logically connected to 802.11 LAN via integration service
 Integration service takes care of address translation and media
conversion
Association Related Services
 Purpose of MAC layer transfer MSDUs between MAC
entities
 Fulfilled by distribution service (DS)
 DS requires information about stations within ESS
 Provided by association-related services
 Station must be associated before communicating
 Three transition types of based on mobility
 No transition: Stationary or moves within range of single BSS
 BSS transition: From one BSS to another within same ESS
 Requires addressing capability be able to recognize new location
 ESS transition: From BSS in one ESS to BSS in another ESS
 Only supported in sense that the station can move
 Maintenance of upper-layer connections not guaranteed
 Disruption of service likely
Station Location
 DS needs to know where destination station is
 Identity of AP to which message should be delivered
 Station must maintain association with AP within current BSS
 Three services relate to this requirement: 
 Association: Establishes initial association between station and AP
 To make identity and address known
 Station must establish association with AP within particular BSS
 AP then communicates information to other APs within ESS
 Reassociation: Transfer established association to another AP
 Allows station to move from one BSS to another
 Disassociation: From either station or AP that association is terminated
 Given before station leaves ESS or shuts
 MAC management facility protects itself against stations that disappear without
notification
Access and Privacy Services - Authentication
 On wireless LAN, any station within radio range other devices can transmit
 Any station within radio range can receive
 Authentication: Used to establish identity of stations to each other
 Wired LANs assume access to physical connection conveys authority to

connect to LAN
 Not valid assumption for wireless LANs
 Connectivity achieved by having properly tuned antenna
 Authentication service used to establish station identity
 802.11 supports several authentication schemes
 Allows expansion of these schemes
 Does not mandate any particular scheme
 Range from relatively insecure handshaking to public-key encryption
schemes
 802.11 requires mutually acceptable, successful authentication before
association
Access and Privacy Services -
Deauthentication and Privacy
 Deauthentication: Invoked whenever an existing
authentication is to be terminated
 Privacy: Used to prevent messages being read by others
 802.11 provides for optional use of encryption
Medium Access Control

 MAC layer covers three functional areas

 Reliable data delivery
 Access control
 Security
 Beyond our scope
Reliable Data Delivery
 802.11 physical and MAC layers subject to unreliability
 Noise, interference, and other propagation effects result in
loss of frames
 Even with error-correction codes, frames may not
successfully be received
 Can be dealt with at a higher layer, such as TCP
 However, retransmission timers at higher layers typically order of seconds
 More efficient to deal with errors at the MAC level
 802.11 includes frame exchange protocol
 Station receiving frame returns acknowledgment (ACK) frame
 Exchange treated as atomic unit
 Not interrupted by any other station
 If noACK within short period of time, retransmit
Four Frame Exchange
 Basic data transfer involves exchange of two frames
 To further enhance reliability, four-frame exchange may be used
 Source issues a Request to Send (RTS) frame to destination
 Destination responds with Clear to Send (CTS)
 After receiving CTS, source transmits data
 Destination responds with ACK
 RTS alerts all stations within range of source that exchange is
under way
 CTS alerts all stations within range of destination
 Stations refrain from transmission to avoid collision
 RTS/CTS exchange is required function of MAC but may be
disabled
Media Access Control
 Distributed wireless foundation MAC (DWFMAC)
 Distributed access control mechanism
 Optional centralized control on top
 Lower sublayer is distributed coordination function
(DCF)
 Contention algorithm to provide access to all traffic
 Asynchronous traffic
 Point coordination function (PCF)
 Centralized MAC algorithm
 Contention free
 Built on top of DCF
IEEE 802.11 Protocol Architecture
Distributed Coordination Function
 DCF sublayer uses CSMA
 If station has frame to transmit, it listens to medium
 If medium idle, station may transmit
 Otherwise must wait until current transmission complete
 No collision detection
 Not practical on wireless network
 Dynamic range of signals very large
 Transmitting station cannot distinguish incoming weak signals from
noise and effects of own transmission
 DCF includes delays
 Amounts to priority scheme
 Interframe space
Interframe Space
 Single delay known as interframe space (IFS)
 Using IFS, rules for CSMA:
1. Station with frame senses medium
• If idle, wait to see if remains idle for one IFS. If so, may transmit immediately
2. If busy (either initially or becomes busy during IFS) station defers
transmission
• Continue to monitor until current transmission is over
3. Once current transmission over, delay another IFS
• If remains idle, back off random time and again sense
• If medium still idle, station may transmit
• During backoff time, if becomes busy, backoff timer is halted and resumes
when medium becomes idle
 To ensure stability, binary exponential backoff used
IEEE 802.11
Medium
Access
Control
Logic
Priority
 Use three values for IFS
 SIFS (short IFS):
 Shortest IFS
 For all immediate response actions (see later)
 PIFS (point coordination function IFS):
 Midlength IFS
 Used by the centralized controller in PCF scheme when issuing polls
 DIFS (distributed coordination function IFS):
 Longest IFS
 Used as minimum delay for asynchronous frames contending for
access
SIFS Use - ACK
 Station using SIFS to determine transmission opportunity has
highest priority
 In preference to station waiting PIFS or DIFS time
 SIFS used in following circumstances:
 Acknowledgment (ACK): Station responds with ACK after
waiting SIFS gap
 No collision detection so likelihood of collisions greater than CSMA/CD
 MAC-level ACK gives efficient collision recovery
 SIFS provide efficient delivery of multiple frame LLC PDU
 Station with multiframe LLC PDU to transmit sends out MAC frames one at a
time
 Each frame acknowledged after SIFS by recipient
 When source receives ACK, immediately (after SIFS) sends next frame in
sequence
 Once station has contended for channel, it maintains control of all fragments sent
SIFS Use – CTS

 Clear to Send (CTS): Station can ensure data

frame will get through by issuing RTS
 Destination station should immediately respond with
CTS if ready to receive
 All other stations hear RTS and defer
 Poll response: See Point coordination Function
(PCF)
PIFS and DIFS

 PIFS used by centralized controller

 Issuing polls
 Takes precedence over normal contention traffic
 Frames using SIFS have precedence over PCF poll
 DIFS used for all ordinary asynchronous traffic
IEEE 802.11 MAC Timing
Basic Access Method
Point Coordination Function (PCF)
 Alternative access method implemented on top of DCF
 Polling by centralized polling master (point coordinator)
 Uses PIFS when issuing polls
 PIFS smaller than DIFS

 Can seize medium and lock out all asynchronous traffic while it issues

polls and receives responses

 E.g. wireless network configured so number of stations with time-
sensitive traffic controlled by point coordinator
 Remaining traffic contends for access using CSMA

 Point coordinator polls in round-robin to stations configured for polling

 When poll issued, polled station may respond using SIFS
 If point coordinator receives response, it issues another poll using PIFS
 If no response during expected turnaround time, coordinator issues poll
Superframe
 Point coordinator would lock out asynchronous traffic by issuing polls
 Superframe interval defined
 During first part of superframe interval, point coordinator polls round-robin to all stations
configured for polling
 Point coordinator then idles for remainder of superframe
 Allowing contention period for asynchronous access
 At beginning of superframe, point coordinator may seize control and issue polls
for given period
 Time varies because of variable frame size issued by responding stations
 Rest of superframe available for contention-based access
 At end of superframe interval, point coordinator contends for access using PIFS
 If idle, point coordinator gains immediate access
 Full superframe period follows
 If busy, point coordinator must wait for idle to gain access
 Results in foreshortened superframe period for next cycle
IEEE 802.11 MAC Timing
PCF Superframe Construction
IEEE 802.11 MAC Frame Format
MAC Frame Fields (1)
 Frame Control:
 Type of frame
 Control, management, or data
 Provides control information
 Includes whether frame is to or from DS, fragmentation information, and privacy
information
 Duration/Connection ID:
 If used as duration field, indicates time (in s) channel will be allocated for
successful transmission of MAC frame
 In some control frames, contains association or connection identifier
 Addresses:
 Number and meaning of address fields depend on context
 Types include source, destination, transmitting station, and receiving station
MAC Frame Fields (2)
 Sequence Control:
 4-bit fragment number subfield
 For fragmentation and reassembly
 12-bit sequence number
 Number frames between given transmitter and receiver
 Frame Body:
 MSDU (or a fragment of)
 LLC PDU or MAC control information
 Frame Check Sequence:
 32-bit cyclic redundancy check
Control Frames
 Assist in reliable data delivery 
 Power Save-Poll (PS-Poll)
 Sent by any station to station that includes AP
 Request AP transmit frame buffered for this station while station in power-saving mode
 Request to Send (RTS)
 First frame in four-way frame exchange
 Clear to Send (CTS)
 Second frame in four-way exchange
 Acknowledgment (ACK)
 Contention-Free (CF)-end
 Announces end of contention-free period part of PCF
 CF-End + CF-Ack:
 Acknowledges CF-end
 Ends contention-free period and releases stations from associated restrictions
Data Frames – Data Carrying
 Eight data frame subtypes, in two groups
 First four carry upper-level data from source station to
destination station
 Data
 Simplest data frame
 May be used in contention or contention-free period
 Data + CF-Ack
 Only sent during contention-free period
 Carries data and acknowledges previously received data
 Data + CF-Poll
 Used by point coordinator to deliver data
 Also to request station send data frame it may have buffered
 Data + CF-Ack + CF-Poll
 Combines Data + CF-Ack and Data + CF-Poll
Data Frames –
Not Data Carrying
 Remaining four data frames do not carry user data
 Null Function
 Carries no data, polls, or acknowledgments
 Carries power management bit in frame control field to AP
 Indicates station is changing to low-power state
 Other three frames (CF-Ack, CF-Poll, CF-Ack +
CF-Poll) same as corresponding frame in
preceding list (Data + CF-Ack, Data + CF-Poll,
Data + CF-Ack + CF-Poll) but without data
Management Frames

 Used to manage communications between stations

and Aps
 E.g. management of associations
 Requests, response, reassociation, dissociation, and
authentication
802.11 Physical Layer
 Issued in four stages
 First part in 1997
 IEEE 802.11
 Includes MAC layer and three physical layer specifications
 Two in 2.4-GHz band and one infrared
 All operating at 1 and 2 Mbps
 Two additional parts in 1999
 IEEE 802.11a
 5-GHz band up to 54 Mbps
 IEEE 802.11b
 2.4-GHz band at 5.5 and 11 Mbps
 Most recent in 2002
 IEEE 802.g extends IEEE 802.11b to higher data rates
Original 802.11 Physical Layer -
DSSS
 Three physical media 
 Direct-sequence spread spectrum
 2.4 GHz ISM band at 1 Mbps and 2 Mbps

 Up to seven channels, each 1 Mbps or 2 Mbps, can be used

 Depends on bandwidth allocated by various national

regulations
 13 in most European countries

 One in Japan

 Each channel bandwidth 5 MHz

 Encoding scheme DBPSK for 1-Mbps and DQPSK for 2-

Mbps
Original 802.11 Physical Layer - FHSS
 Frequency-hopping spread spectrum
 2.4 GHz ISM band at 1 Mbps and 2 Mbps
 Uses multiple channels
 Signal hopping from one channel to another based on a pseudonoise sequence
 1-MHz channels are used
 23 channels in Japan
 70 in USA
 Hopping scheme adjustable
 E.g. Minimum hop rate forUSA is 2.5 hops per second
 Minimum hop distance 6 MHz in North America and most of Europe and 5 MHz in
Japan
 Two-level Gaussian FSK modulation for 1-Mbps
 Bits encoded as deviations from current carrier frequency
 For 2 Mbps, four-level GFSK used
 Four different deviations from center frequency define four 2-bit combinations
Original 802.11 Physical Layer –
Infrared
 Omnidirectional
 Range up to 20 m
 1 Mbps used 16-PPM (pulse position modulation)
 Each group of 4 data bits mapped into one of 16-PPM symbols
 Each symbol a string of 16 bits
 Each 16-bit string consists of fifteen 0s and one binary 1
 For 2-Mbps, each group of 2 data bits is mapped into one
of four 4-bit sequences
 Each sequence consists of three 0s and one binary 1
 Intensity modulation
 Presence of signal corresponds to 1 
802.11a
 5-GHz band
 Uses orthogonal frequency division multiplexing (OFDM)
 Not spread spectrum
 Also called multicarrier modulation
 Multiple carrier signals at different frequencies
 Some bits on each channel
 Similar to FDM but all subchannels dedicated to single source
 Data rates 6, 9, 12, 18, 24, 36, 48, and 54 Mbps
 Up to 52 subcarriers modulated using BPSK, QPSK, 16-QAM,
or 64-QAM
 Depending on rate
 Subcarrier frequency spacing 0.3125 MHz
 Convolutional code at rate of 1/2, 2/3, or 3/4 provides forward error correction
802.11b
 Extension of 802.11 DS-SS scheme
 5.5 and 11 Mbps
 Chipping rate 11 MHz
 Same as original DS-SS scheme
 Same occupied bandwidth
 Complementary code keying (CCK) modulation to achieve higher data
rate in same bandwidth at same chipping rate
 CCK modulation complex
 Overview on next slide
 Input data treated in blocks of 8 bits at 1.375 MHz
 8 bits/symbol  1.375 MHz = 11 Mbps
 Six of these bits mapped into one of 64 code sequences
 Output of mapping, plus two additional bits, forms input to QPSK modulator
11-Mbps CCK Modulation Scheme
802.11g

 Higher-speed extension to 802.11b

 Combines physical layer encoding techniques
used in 802.11a and 802.11b to provide service at
a variety of data rates