1. TCP/IP Protocol Suite 1 Chapter 3 Objectives Upon completion you will be able to: Underlying Technology Understand the different versions of wired Ethernet Understand wireless Ethernet Understand the types of point-to-point WANs Understand the types of switched WANs, especially ATM Differentiate between repeaters, bridges, routers, and hubs

2. TCP/IP Protocol Suite 2 Figure 3.1 Internet model

3. TCP/IP Protocol Suite 3 3.1 Local Area Networks A local area network (LAN) is a data communication system that allows a number of independent devices to communicate directly with each other in a limited geographic area such as a single department, a single building, or a campus. A large organization may need several connected LANs.The most popular LANs are Ethernet and wireless LANs. We briefly review these technologies in this section. The topics discussed in this section include: Wired LANs: Ethernet Wireless LANs: IEEE 802.11

4. TCP/IP Protocol Suite 4 Ethernet Xerox (Bob Metcalfe) performed initial development of Ethernet in 1976 (2.94 Mbps over 100 personal workstations, 1-km long cable) and was later joined by the Digital Equipment Corporation (DEC) and Intel to define the Ethernet 1 specification in 1980. The same group subsequently released the Ethernet 2 specification in 1984. The Ethernet specification describes a packet switching CSMA/CD LAN.

5. TCP/IP Protocol Suite 5 Data link layer Physical layer Original Ethernet architecture 0.5 “ Coax tap BNC connector transceiver AUI cable station interface data encapsulation link management encoding and decoding transmission and receipt Network Interface Card ( NIC)

6. TCP/IP Protocol Suite Carrier sense: station listens to media before transmitting Multiple access: multiple stations may access at same time Carrier Sense Multiple Access Media idle? YES Transmit immediately NO -- wait Listen until the media is idle, then transmit

7. TCP/IP Protocol Suite Collision More than one station may send a frame during overlapping times. How does a station know that a collision occurred? What does the station do after a collision?

8. TCP/IP Protocol Suite 8

9. Collision Detection Rules 1.Stations must listen to the cable while transmitting in order to detect a collision. 2.A frame must be at least 64 bytes (512 bits, 51.2 msec) long to ensure sender “hears” a collision before he finishes. (The transmission time must be more than the RTT.) 3. If a collision is detected, send a brief jamming signal and then wait before retransmitting. Jamming signal

10. TCP/IP Protocol Suite 10 Worst Case Collision Timing Assume delay for repeater is 1 msec; for transceiver 0.5 msec. ComponentPropagation Time Microsecs Five 500 meter segments2500m/0.77c10.8 Four repeaters4 x 1 msec 4.0 Nine 50 meter AUI cables 450m/0.65c 2.3 Nine transceivers9 x 0.5 msec 4.5 Total one-way time21.6 2,500 meters 500 meters

11. TCP/IP Protocol Suite 11 Worst Case Collision Timing (cont.) Under these assumptions, the round trip time would be 43.2 microseconds. Allowing some tolerance for equipment, IEEE chose 51.2 microseconds, equal to 512 bit-times, as the collision detection interval. This is why all frames must be at least 512 bits (64 bytes) long. 43.2 microsecond RTT 500 meters

12. TCP/IP Protocol Suite How does a node detect a collision? Transceiver: A node monitors the media while transmitting. If the observed power is more than transmitted power + attenuated reflection of its own signal, it indicates a collision. Transmitted signal Observed signal Collision! Simultaneous input on two ports Output “ collision presence ” on all ports Hub: if input occurs simultaneously on two ports, it indicates a collision. Hub sends a collision presence signal on all ports.

13. TCP/IP Protocol Suite Late Collisions A late collision occurs when sender finishes transmission before detecting collision presence. Usual causes: cable too long too many repeaters between stations Solution: higher layer protocol must detect packet loss and retransmit.

14. TCP/IP Protocol Suite 14 Figure 3.3 Ethernet layers

15. TCP/IP Protocol Suite 15 Figure 3.4 Ethernet frame

16. TCP/IP Protocol Suite 16 Figure 3.5 Ethernet implementations

17. TCP/IP Protocol Suite 17 10Base5 tap : cable does not need to be cut tap : cable does not need to be cut transceiver : send/receive, collision transceiver : send/receive, collision detection, electronics isolation detection, electronics isolation AUI : Attachment Unit Interface AUI : Attachment Unit Interface Use for backbone networks Use for backbone networks 0.5“ Coax  0.5“ Coax vampire tap BNC connector transceiver AUI cable maximum segment length=500m maximum number of stations per segment=100 minimum distance between two stations = 2.5 m maximum network distance between two stations = 2.8km NIC

18. TCP/IP Protocol Suite 18 10Base 2 0.25 “ Coax BNC T-connector NIC BNC connector No drop cable use for office LAN What is its benefit since length < 500m? maximum segment length=185m maximum number of stations per segment=30 minimum distance between two stations = 0.5 m maximum network distance between two stations = 925 m

19. TCP/IP Protocol Suite 19 A hub functions as a repeater UTP category 5 uses 2 pairs of wires terminated by an eight-bin (RJ-45 style) connector. This means that 4 pins of the 8-pin are used. The transmit and receive data signal on each pair of the segment are polarised, with one wire of the signal pair carrying the positive (+) signal and the other carrying the negative (-). 10BaseT NIC hub maximum segment length = 100m Medium Dependent Interface (MDI), RJ45

20. TCP/IP Protocol Suite 20 Half Duplex / Full Duplex Ethernet they were initially implemented in a half-duplex manner with the transceiver detecting a collision if an attempt was made to transmit and receive simultaneously and looping back data to the host so it could hear itself transmit (as it would on a shared medium). However if both ends of the link are not hubs, and the hardware supports it, the two channels can be split and used to make a full- duplex link. Unfortunately if autonegotiation is enabled on one end and forced full-duplex on the other, the end with autonegotiation will detect the link as half-duplex causing large numbers of errors due to duplex mismatch.full- duplexautonegotiationduplex mismatch Ethernet Charecteristics

21. TCP/IP Protocol Suite 21 Collision Domain - Segment Broadcast Domain – Subnet Hub Switch Ethernet Charecteristics

22. TCP/IP Protocol Suite 22 Figure 3.6 Fast Ethernet implementations

23. TCP/IP Protocol Suite 23 Figure 3.7 Gigabit Ethernet implementations

24. TCP/IP Protocol Suite 24 Wireless LAN (WLAN) A WLAN is a shared network. An access point is a shared device and functions like a shared Ethernet hub. Data is transmitted over radio waves. Two-way radio communications (half-duplex) are used. The same radio frequency is used for sending and receiving (transceiver).

25. TCP/IP Protocol Suite What Are WLANs? They are: Local In building or campus for mobile users Radio or infrared Not required to have RF licenses in most countries Using equipment owned by customers They are not: WAN or MAN networks Cellular phones networks Packet data transmission via celluar phone networks Cellular digital packet data (CDPD) General packet radio service (GPRS) 2.5G to 3G services

26. TCP/IP Protocol Suite 26 Similarities Between WLAN and LAN A WLAN is an 802 LAN. Transmits data over the air vs. data over the wire Looks like a wired network to the user Defines physical and data link layer Uses MAC addresses The same protocols/applications run over both WLANs and LANs. IP (network layer) IPSec VPNs (IP-based) Web, FTP, SNMP (applications)

27. TCP/IP Protocol Suite 27 Differences Between WLAN and LAN WLANs use radio waves as the physical layer. WLANs use CSMA/CA instead of CSMA/CD to access the network. Radio waves have problems that are not found on wires. Connectivity issues. Coverage problems Multipath issues Interference, noise Privacy issues. WLANs use mobile clients. No physical connection. Battery-powered. WLANs must meet country-specific RF regulations.

28. TCP/IP Protocol Suite 28 WLAN and LAN

29. TCP/IP Protocol Suite 29 Service Set Identifier (SSID) SSID is used to logically separate WLANs. The SSID must match on client and access point. Access point broadcasts one SSID in beacon. Client can be configured without SSID. Client association steps: 1. Client sends probe request. 2. A point sends probe response. 3. Client initiates association. 4. A point accepts association. 5. A point adds client MAC address to association table.

30. TCP/IP Protocol Suite WLAN Access Topology

31. TCP/IP Protocol Suite Wireless Repeater Topology

32. TCP/IP Protocol Suite 32 Workgroup Bridge Topology

33. TCP/IP Protocol Suite Alternative Peer-to-Peer Topology

34. TCP/IP Protocol Suite 34 Service Sets and Modes Ad hoc mode Independent Basic Service Set (IBSS) Mobile clients connect directly without an intermediate access point. Infrastructure mode Basic Service Set Mobile clients use a single access point for connecting to each other or to wired network resources. Extended Services Set Two or more Basic Service Sets are connected by a common distribution system.

35. TCP/IP Protocol Suite 35 Figure 3.9 ESS

36. TCP/IP Protocol Suite 36 Roaming Through Wireless Cells

37. TCP/IP Protocol Suite 37 Client Roaming Roaming without interruption requires the same SSID on all access points. Maximum data retry count exceeded Too many beacons missed Data rate shifted Periodic intervals

38. TCP/IP Protocol Suite 38 802.11b Standard Standard was ratified in September 1999 Operates in the 2.4-GHz band Specifies direct sequence spread spectrum (DSSS) Specifies four data rates up to 11 Mbps 1, 2, 5.5, 11 Mbps Provides specifications for vendor interoperability (over the air) Defines basic security, encryption, and authentication for the wireless link Is the most commonly deployed WLAN standard

39. TCP/IP Protocol Suite Channel Identifier Channel Center Frequency Channel Frequency Range [MHz] Regulatory Domain Americas Europe, Middle East, and Asia Japan 12412 MHz2401 – 2423XXX 22417 MHz2406 – 2428XXX 32422 MHz2411 – 2433XXX 42427 MHz2416 – 2438XXX 52432 MHz2421 – 2443XXX 62437 MHz2426 – 2448XXX 72442 MHz2431 – 2453XXX 82447 MHz2436 – 2458XXX 92452 MHz2441 – 2463XXX 102457 MHz2446 – 2468XXX 112462 MHz2451 – 2473XXX 122467 MHz2466 – 2478XX 132472 MHz2471 – 2483XX 142484 MHz2473 – 2495 X 2.4-GHz Channels

40. TCP/IP Protocol Suite 40 2.4-GHz Channel Use Each channel is 22 MHz wide. North America: 11 channels. Europe: 13 channels. There are three nonoverlapping channels: 1, 6, 11. Using any other channels will cause interference. Three access points can occupy the same area.

41. TCP/IP Protocol Suite 41 802.11b/g (2.4 GHz) Channel Reuse

42. TCP/IP Protocol Suite 42 802.11b Access Point Coverage

43. TCP/IP Protocol Suite 43 802.11g Standard Standard was ratified June 2003 Operates in the 2.4-GHz band as 802.11b Same three nonoverlapping channels: 1, 6, 11 DSSS (CCK) and OFDM transmission 12 data rates of up to 54 Mbps 1, 2, 5.5, 11 Mbps (DSSS / 802.11b) 6, 9, 12, 18, 24, 36, 48, 54 Mbps (OFDM) Full backward compatiblity to 802.11b standard

44. TCP/IP Protocol Suite 44 Range Comparisons

45. TCP/IP Protocol Suite 45 Figure 3.10 Physical layer

46. TCP/IP Protocol Suite 46 Figure 3.11 FHSS Frequency Hopping Sequential use of multiple frequencies Hop sequence and rate will vary

47. TCP/IP Protocol Suite 47 Figure 3.12 DSSS Direct Sequence Each symbol is transmitted over multiple frequencies at the same time Very efficient (no overhead) Higher speed than FH at comparable distances System capacity (multiple channels) higher than FH

48. TCP/IP Protocol Suite 48 Figure 3.13 MAC layers in IEEE 802.11 standard

49. TCP/IP Protocol Suite 49 802.11 Collisions Avoidance Similar to CSMA/CD (Ethernet) Transmit when medium is idle, back off on collision Problem: medium is not fully shared ABCD

50. TCP/IP Protocol Suite 50 802.11 Collisions Avoidance contd. Hidden node problem: A and C cannot hear each other If they both transmit to B at the same time, there will be a collision which won’t be detected by A or C

51. TCP/IP Protocol Suite 51 802.11 Collisions Avoidance contd. Exposed node problem: C could send to D while B is sending to A C is blocked when B is sending

52. TCP/IP Protocol Suite “Hidden stations” the solution IEEE 802.11 defines: MAC level RTS/CTS protocol (Request to Send / Clear to Send) Can be switched off to reduce overhead (when no hidden nodes exist) More robustness, and increased reliability No interruptions when large files are transmitted AB RTS: I want to send to B 500 bytes CTS: OK A, go ahead, so everybody quiet Data: the 500 bytes of data from A to B ACK: B received the data OK, so an ACK C

53. TCP/IP Protocol Suite 53 Figure 3.14 CSMA/CA

54. TCP/IP Protocol Suite 54 Figure 3.15 Frame

55. TCP/IP Protocol Suite 55 Table 3.1 Addresses in IEEE 802.11

56. TCP/IP Protocol Suite 56 3.2 Point-to-Point WANs A second type of network we encounter in the Internet is the point-to- point wide area network. A point-to-point WAN connects two remote devices using a line available from a public network such as a telephone network. We discuss the physical and data link layers of these technologies here.. The topics discussed in this section include: Physical Layer Data Link Layer

57. TCP/IP Protocol Suite 57 Figure 3.16 56K modem

58. TCP/IP Protocol Suite 58 ADSL is an asymmetric communication technology designed for residential users; it is not suitable for businesses. Note:

59. TCP/IP Protocol Suite 59 Figure 3.17 Bandwidth division

60. TCP/IP Protocol Suite 60 Figure 3.18 ADSL and DSLAM

61. TCP/IP Protocol Suite 61 Figure 3.19 Cable bandwidth

62. TCP/IP Protocol Suite 62 Figure 3.20 Cable modem configurations

63. TCP/IP Protocol Suite 63 Table 3.2 T line rates

64. TCP/IP Protocol Suite 64 Table 3.3 SONET rates

65. TCP/IP Protocol Suite 65 Figure 3.21 PPP frame

66. TCP/IP Protocol Suite 66 3.3 Switched WANs The backbone networks in the Internet are usually switched WANs. A switched WAN is a wide area network that covers a large area (a state or a country) and provides access at several points to the users. Inside the network, there is a mesh of point-to-point networks that connects switches. The switches, multiple port connectors, allow the connection of several inputs and outputs. The topics discussed in this section include: X.25 Frame Relay ATM

67. TCP/IP Protocol Suite 67 Figure 3.22 Frame Relay network

68. TCP/IP Protocol Suite 68 A cell network uses the cell as the basic unit of data exchange. A cell is defined as a small, fixed-size block of information. Note:

69. TCP/IP Protocol Suite 69 Figure 3.23 ATM multiplexing

70. TCP/IP Protocol Suite 70 Figure 3.24 Architecture of an ATM network

71. TCP/IP Protocol Suite 71 Figure 3.25 Virtual circuits

72. TCP/IP Protocol Suite 72 Note that a virtual connection is defined by a pair of numbers: the VPI and the VCI. Note:

73. TCP/IP Protocol Suite 73 Figure 3.26 An ATM cell

74. TCP/IP Protocol Suite 74 Figure 3.27 ATM layers

75. TCP/IP Protocol Suite 75 The IP protocol uses the AAL5 sublayer. Note:

76. TCP/IP Protocol Suite 76 We will discuss IP over ATM in Chapter 23. Note:

77. TCP/IP Protocol Suite 77 3.4 Connecting Devices LANs or WANs do not normally operate in isolation. They are connected to one another or to the Internet. To connect LANs or WANs, we use connecting devices. Connecting devices can operate in different layers of the Internet model. We discuss three kinds of connecting devices: repeaters (or hubs), bridges (or two-layer switches), and routers (or three-layer switches). Repeaters and hubs operate in the first layer of the Internet model. Bridges and two-layer switches operate in the first two layers. Routers and three-layer switches operate in the first three layers The topics discussed in this section include: RepeatersHubsBridgesRouter

78. TCP/IP Protocol Suite 78 Figure 3.28 Connecting devices

79. TCP/IP Protocol Suite 79 Figure 3.29 Repeater

80. TCP/IP Protocol Suite 80 A repeater connects segments of a LAN. Note:

81. TCP/IP Protocol Suite 81 A repeater forwards every bit; it has no filtering capability. Note:

82. TCP/IP Protocol Suite 82 A repeater is a regenerator, not an amplifier. Note:

83. TCP/IP Protocol Suite 83 Figure 3.30 Function of a repeater

84. TCP/IP Protocol Suite 84 A bridge has a table used in filtering decisions. Note:

85. TCP/IP Protocol Suite 85 Figure 3.31 Bridge

86. TCP/IP Protocol Suite 86 A bridge does not change the physical (MAC) addresses in a frame. Note:

87. TCP/IP Protocol Suite 87 Figure 3.32 Learning bridge

88. TCP/IP Protocol Suite 88 A router is a three-layer (physical, data link, and network) device. Note:

89. TCP/IP Protocol Suite 89 A repeater or a bridge connects segments of a LAN. A router connects independent LANs or WANs to create an internetwork (internet). Note:

90. TCP/IP Protocol Suite 90 Figure 3.33 Routing example

91. TCP/IP Protocol Suite 91 A router changes the physical addresses in a packet. Note: