Chapter 7 Ethernet

IEEE 802.3 MAC: CSMA/CD

10-Mbps Ethernet
100-Mbps Ethernet
Gigabit Ethernet
 CSMA/CD
• Random Access (or Contention) techniques
• Precursors:
– Pure ALOHA, initially developed for packet radio
networks.
• A station may transmit a frame at any time.
– The station then listens for an amount of time equals to
the maximum possible round-trip propagation delay on the
network plus a small fixed time increment.
» If the station hears an acknowledgement during that
time, OK.
» Otherwise, it resends the frame.
– If the station fails to receive an acknowledgement after
repeated transmissions, it gives up.
 (Cont.)
• A receiving station determines the correctness of an
incoming frame by examining a frame-check-
sequence field:
– If the frame is valid and if the destination address matches
the receiver’s address, the station immediately sends an
acknowledgement.
– If the frame is invalid, due to noise or
collision(interference with other’s transmission), the
receiving station ignores the frame.
• The maximum channel utilization by pure(unslotted)
ALOHA is only about 18% .
 (Cont.)
– Slotted ALOHA:
• Time on the channel is organized into slots,
whose size equals the frame transmission
time.
• It is necessary to synchronize all stations.
• Transmission is permitted to begin only at a
slot boundary.
• The maximum channel utilization by slotted
ALOHA is about 37% .
 (Cont.)
• CSMA (Carrier Sense Multiple Access):
– Take Take advantage of the fact that propagation delay
between stations is usually very small, compared to
frame transmission time.
– A station wishing to transmit first listens to the medium
to determine if another transmission is in progress
(Carrie r Sense):
• If the medium is in use, the station must wait;
• If the medium is idle, the station may transmit:
– It may happen that two or more stations attempt to
transmit at about the same time time. If this happens,
therefore will be a collision.
 (Cont.)
– The station waits (maximum round-trip
propagation delay plus the time for contending
for the channel for response) for
acknowledgement after transmission:
– If there is no acknowledgement, the station
assumes a collision of transmission and
retransmits.
– If there is no collision within the duration of
propagating the leading edge of a packet from a
transmitting station to the farthest station, there will be
no collision for this frame.
– The longer the frames or the shorter the propagation
time, the higher the utilization.
 Deference in CSMA (Fig. 7-1)
• Regarding the algorithm specifies what should
do if the medium is found busy.
• Nonpersistent CSMA:
1. If the medium is idle, transmit; otherwise, go to step 2.
2. If the medium is busy, wait an amount of time drawn
from a probability distribution of “attempt”
(retransmission) delay and repeat step 1.
– A problem with nonpersistent CSMA is that capacity
is wasted because the medium will remain idle
following the end of a transmission even if there are
one or more stations waiting to transmit.
 Nonpersistent:
•Transmit if idle
Constant or variable delay •If busy, wait random time
and repeat process
•If collision, back off

Channel Busy

Ready
P-Persistent:
•Transmit as soon as channel
1-Persistent: goes idle with probability P
•Transmit as soon as •Otherwise, delay one time slot
channel goes idle and repeat process
•If collision, back off •If collision, back off

Figure 7.1 CSMA Persistence and Backoff

 (Cont.)
• 1-persistent CSMA
1. If the medium is idle, transmit; otherwise, go
to step 2.
2. If the medium is busy, continue to listen until
the channel is sensed idle; then transmit
immediately.
– Nonpersistent CSMA can reduce collisions; 1
persistent CSMA can reduce idle time.
 (Cont.)
• p-persistent CSMA:
1. If the medium is idle, transmit with probability p and
delay one time unit with probability (1-p). The time
unit is typically equal to the maximum propagation
delay.
2. If the medium is busy, continue to listen until the
channel is idle and repeat step 1.
3. If the transmission is delayed one time unit, repeat
step 1.
– With p-persistent, instability can possibly happen under heavy
load and long delay may happen when p and load are small.
– CSMA is more efficient than slotted or pure ALOHA, but the
medium remains unusable for the duration of transmission when
two frames collide.
 Description of CSMA/CD
• The waste of channel time due to collided frame
transmission can be reduced if a station continues
to listen to the medium while transmitting.
• CSMA/CD(CSMA with collision detection):
1. If the medium is idle, transmit; otherwise, go to step 2.
2. If the medium is busy, continue to listen until the
channel is idle; then transmit immediately.
3. If a collision is detected during transmission, transmit a
brief jamming signal to assure that all stations know
that there has been a collision and then cease
transmission.
4. After transmitting the jamming signal, wait a random
amount of time, then attempt to transmit again(repeat
from step 1).
 (Cont., Fig. 7-2)
• With CSMA/CD, the amount of wasted capacity is
reduced to the time it takes to detect a collision.
– For a baseband bus, the amount of time to detect a
collision is no greater than twice the end-to-end
propagation delay.
– For a broadband bus (dual cable), the maximum time to
detect a collision is four times the propagation delay
from an end of the cable to the headend.
• Frames should be long enough to allow collision
detection prior to the end of transmission.
 A B C D

TIME t0

A's transmission

C's transmission

Signal on bus

TIME t1

A's transmission

C's transmission

Signal on bus

TIME t2

A's transmission

C's transmission

Signal on bus

TIME t3

A's transmission

C's transmission

Signal on bus

Figure 7.2 CSMA/CD Operation

 (Cont.)
• If shorter frame is used, collision detection
does not occur and the CSMA/CD performs
similarly as the CSMA protocol does.
• 1 persistent is usually used in Ethernet and
IEEE 802.3 standard.
• Random back-off (Deference process) is
used when collision is detected.
 Truncated Binary Exponential
Backoff
• The Truncated Binary exponential backoff:
– (The range of numbers(r) is 0<r<2k, where for
repeated collisions, k is expanded from 1 until
10, after which it stays at 10.)
• A station will attempt to transmit repeatedly in the
face of repeated collision,but after each collision,
the mean value of the random delay is doubled.
• After 16 unsuccessful attempts, the station gives up
and reports an excessive collisions error.
 (Cont.)
• 1-persistent with binary backoff :
– At low load, a station can seize the channel as
soon as it goes idle(1 persistent).
– At high load, the system(with binary backoff)
is at least as stable as the other technique.
– Last-in and first-out regarding to stations)
effect possibly happens by the backoff
algorithm
 Carrier Sense and Collision
Detection
• Carrier sense:
– Baseband transmission
• By detecting signal transition (of Manchester coding)
– Broadband transmission
• By sensing real carrier
 (Cont.)
• Collision detection:
– Baseband transmission:
• A transmitting transceiver will detect a collision if the signal
on the cable at the transceiver exceeds the maximum that could
be produced by the transceiver alone.
• Collision detection at a repeater:
– If a repeater is repeating a frame onto a segment and a collision
occurs on this segment, then the repeater detects a collision in
the same fashion as any transmitting transceiver; that is, if the
signal exceeds the maximum that could be produced by the
transceiver alone.
– If a collision is caused by two stations transmitting on one
segment, a repeater attached to that segment will hear a
transmission and should transmit a jamming signal on the other
segment to which it attaches. In this case, a collision is detected
if the signal strength exceeds that which could be produced by
two transceiver outputs in the worst case.
 (Cont.)
• Collision detection with twisted pair star-
wiring approach (Fig 7-3):
– Collision detection is based on logic rather than
sensing voltage magnitudes.
– For any hub, if there is activity(signal) on more
than one input, a collision is assumed. A
collision presence signal is generated and sent
out as long as activity is sensed on any of the
input lines.
 HHUB
Fa
Fa
Fa
Fa
IHUB IHUB E
Fa Fa Fa
Fa Fa
A B C D

(a) A transmitting

HHUB
CP
CP
CP

IHUB IHUB E
Fa CP CP
CP CP
Fb
A B C D

(b) A and B transmitting

HHUB
CP
CP
CP

IHUB IHUB E
Fa CP CP
Fc
CP Fb CP
A B C D

(c) A, B, and C transmitting

Fx = Frame from station x
CP = collision presence signal

Figure 7.3 Operation of a Two-Level Star-Topology CSMA/CD Configuration

 MAC Frame
• Fig.7.4, the 802.3 MAC frame and its fields:
– Preamble
• A 7-octet pattern of alternating zero and ones used by the
receiver to establish bit synchronization
– Start frame delimiter (SFD)
• The sequence 10101011 that indicates the beginning of the
MAC frame and enables the receiver to locate the first bit of
the rest of the frame.
– Destination address (DA)
• It specifies the stations for which the frame is intended. It may
be a unique physical address, a group address, or a global
address.
– The group MAC DA of all 1 bits refers to all stations on the
LAN data link and is called the broadcast address.
 46 to 1500 octets

7 octets 1 6 6 2 ≥ 0 ≥ 0 4
S P
Preamble F DA SA Length LLC Data a FCS
D d

SFD = Start of frame delimiter

DA = Destination address
SA = Source address
FCS = Frame check sequence

Figure 7.4 IEEE 802.3 Frame Format

 (Cont.)
– Source address (SA):
• It specifies the station that sent the frame.
– Length/type:
• Length of LLC data field in octets, or Ethernet Type field:
– If it is greater than 1500 (05DC-HEX), it conforms to the earlier
Ethernet specifications. The number corresponds to a network
layer PDU encapsulated in the frame body, for example 0800
(HEX) for IP.
– If it is less or equal 1500, it conforms to the IEEE 802.3 standard.
It gives the length of LLC data field.
– In either case, the maximum frame size, excluding the Preamble
and SFD, is 1518 octets.
– LLC data
• Data unit supplied by LLC.
 (Cont.)
– Pad
• Octets added to ensure that the frame is long enough
for proper collision detection operation.
– Frame check sequence (FCS)
• A 32 bit cyclic redundancy check, based on all fields
except preamble, SFD and FCS.
 MAC Compatibility
• In 802.3 MAC spec, some parameters are to some
extent medium dependent---part of physical layer
spec, as in Table 7.1
– slotTime
• The roundtrip propagation delay of the network. This is the
time it takes for a node at one end of the network that starts
transmitting to recognize a collision caused by another node
located at the furthest point away on the network.
– interFrameGap
• The minimum delay or time gap between frames, intended to
allow recovery time for other CSMA/CD sublayers and the
physical medium.
 Table 7.1 IEEE 802.3 Parameterized Values

Parameter 10-Mbps Value 100-Mbps Value 1000-Mbps Value

slotTime 512 bit times 512 bit times 4096 bit times
interFrameGap 9.6 µs 0.96 µs 0.096 µs

attemptLimit 16 16 16
backoffLimit 10 10 10
jamSize 32 bits 32 bits 32 bits
maxFrameSize 1518 octets 1518 octets 1518 octets
minFrameSize 64 octets 64 octets 64 octets
burstLimit not applicable not applicable 8192 octets
 (Cont.)
– attemptLimit
• The maximum number of retries that a station will make when
collisions occur.
– backoffLimit
• When multiple retries after collision occur, the number of
times that the backoff time distribution is doubled.
– jamSize
• The number of bit times of the jamming signal transmitting
upon collision detection.
– maxFrameSize
• The maximum size of a frame, from the destination address
field through the frame check sequence field.
 (Cont.)
– minFrameSize
• The minimum size of a frame, from the destination
address field through the frame check sequence field.
– burstLimit
• The maximum number of octets, consisting of a
sequence of frames, that a station may transmit in a
burst.
 10-Mbps Ethernet
• A notation represents different implementation of
802.3 standard:
<data rate in Mbps><signaling method><maximum segment length in
hundreds of meters>
• The alternative at 10 Mbps:
– 10BASE5
– 10BASE2
– 10BASE-T
– 10BROAD36
– 10BASE-F
• Table 7.2
 Table 7.2 IEEE 802.3 10-Mbps Physical Layer Medium Alternatives

10BASE5 10BASE2 10BASE-T 10BASE-FP

Transmission Coaxial Cable (50 Coaxial Cable (50 Unshielded twisted 850-nm optical fiber
medium ohm) ohm) pair pair
Signaling technique Baseband Baseband Baseband Manchester/On-off
(Manchester) (Manchester) (Manchester)
Topology Bus Bus Star Star
Maximum segment 500 185 100 500
length (m)
Nodes per segment 100 30 — 33
Cable diameter 10 mm 5 mm 0.4 to 0.6 mm 62.5/125 µm
 Medium Access Unit (MAU)
• The MAU is supposed to perform the functions:
– Transmit signals on the medium.
– Receive signals from the medium.
– Recognize the presence of a signal on the medium.
– Recognize a collision.

• When the MAU is not physically integrated with the remainder

of the 802.3 logic, the two are joined by a set of twisted pair
cables referred to as the attachment unit interface (AUI).
• BNC
• RJ45
 (Cont.)
• The maximum transmission path permitted
between any two stations is five segments
and four repeater sets.
– A segment is either a point-to-point link
segment or a coaxial cable 10BASE5 or
10BASE2 segment.
• The maximum number of coaxial cable
segments in a path is three.
Multiport Multiport
repeater repeater
set set
Repeater Repeater
unit unit

MAU MAU MAU MAU MAU MAU MAU MAU

Station Station Station Station Station Station

Figure 7.5 Simple 10BASE-T Configuration

 10BASE5

Figure 7.6 Mixed 10BASE-T and 10BASE5 Configuration

 10BASE-F medium spec.
• 10BASE-FP (passive)
– A passive-star topology for interconnecting stations and
repeaters up to 1 km per segment.
• 10BASE-FL (link)
– Define a point-to-point link that can be used to connect
stations or repeaters at up to 2 km.
• 10BASE-FB (backbone)
– Define a point-to-point link that can be used to connect
repeaters at up to 2 km.
All use a pair of optical fibers in a link for
transmission in each dirction.
 100-MBPS Ethernet
• Fast Ethernet, with data transmission rate 100M
bps
• Fig 7.7:
– 100BASE-T options
• Use the IEEE 802.3 MAC protocol&frame format.
– 100BASE-X
• Refer to a set of options that use the physical medium
specifications originally defined for FDDI.
• Use two physical links between nodes, one for transmission
and one for reception.
– 100BASE-TX uses STP or UTP-Category 5.
– 100BASE-FX uses optical fibers.
 100BASE-T

100BASE-X

100BASE-TX 100BASE-FX 100BASE-T4

2 Category 5 UTP 2 STP 2 Optical Fiber 4 Category 3 or Category 5 UTP

Figure 7.7 IEEE 802.3 100BASE-T Options

 (Cont.)
– 100BASE-T4
• UTP-Category 3, 4 ,or 5
• 4 twisted pair lines between nodes, with data
transmission over three pairs in one direction at a
time.
• Table 7.3: 100BASE-T options
 Table 7.3 IEEE 802.3 100BASE-T Physical Layer Medium Alternatives

100BASE-TX 100BASE-FX 100BASE-T4

Transmission 2 pair, STP 2 pair, Category 2 optical fibers 4 pair, Category
medium 5 UTP 3, 4, or 5 UTP
Signaling MLT-3 MLT-3 4B5B, NRZI 8B6T, NRZ
technique
Data rate 100 Mbps 100 Mbps 100 Mbps 100 Mbps
Maximum 100 m 100 m 100 m 100 m
segment length
Network span 200 m 200 m 400 m 200 m
 100BASE-X
• A unidirectional data rate of 100 Mbps over
a single link(single twisted pair, single
optical fiber).
• Signal encoding: 4B/5B-NRZI, roiginally
defined fro FDDI. It is further modified for
each option.
• Appendix 7A
 4B/5B-NRZI
– 4B5B, each 4 bits of data are encoded into a symbol of
5 code bits, with efficiency of 100Mbps achieved
at 125Mbaud
– NRZI, (differential encoding), a binary 1 is represented
with a transition at the beginning of the bit interval, and
a binary 0 with no transition at the beginning of the bit
interval---improve detection reliability.
– Synchronization. The 5 bit patterns for encoding
sixteen 4 bit data are chosen to guarantee no more than
three zeros in a row and at least twice transitions for
each 5-code group.
 Table 7.4 Maximum Collision Domain (meters)

Repeater Type Copper Copper and Fibera Fiber

Station-to-station 100 N/A 400

1 Class I Repeater 200 230 240
1 Class II Repeater 200 285 318
2 Class II Repeaters 205 (200 Category 3) 212 226

a: Mixed copper and fiber assumes a 100-meter copper link

 Table 7.5 4B/5B Code Groups (page 1 of 2)

Data Input Code Group NRZI pattern Interpretation

(4 bits) (5 bits)

0000 11110 Data 0

0001 01001 Data 1

0010 10100 Data 2

0011 10101 Data 3

0100 01010 Data 4

0101 01011 Data 5

0110 01110 Data 6

0111 01111 Data 7

1000 10010 Data 8

1001 10011 Data 9

1010 10110 Data A

1011 10111 Data B

1100 11010 Data C

 Table 7.5 4B/5B Code Groups (page 2 of 2)

1101 11011 Data D

1110 11100 Data E

1111 11101 Data F

11111 Idle

11000 Start of stream

delimiter, part 1

10001 Start of stream

delimiter, part 2

01101 End of stream

delimiter, part 1

00111 End of stream

delimiter, part 2

00100 Transmit error

other invalid codes

 Table 7.6 Portion of 8B6T Code Table

Data 6T code Data 6T code Data 6T code Data 6T code

octet group octet group octet group octet group
00 +-00+- 10 +0+--0 20 00-++- 30 +-00-+

01 0+-+-0 11 ++0-0- 21 --+00+ 31 0+--+0

02 +-0+-0 12 +0+-0- 22 ++-0+- 32 +-0-+0

03 -0++-0 13 0++-0- 23 ++-0-+ 33 -0+-+0

04 -0+0+- 14 0++--0 24 00+0-+ 34 -0+0-+

05 0+--0+ 15 ++00-- 25 00+0+- 35 0+-+0-

06 +-0-0+ 16 +0+0-- 26 00-00+ 36 +-0+0-

07 -0+-0+ 17 0++0-- 27 --+++- 37 -0++0-

08 -+00+- 18 0+-0+- 28 -0-++0 38 -+00-+

09 0-++-0 19 0+-0-+ 29 --0+0+ 39 0-+-+0

0A -+0+-0 1A 0+-++- 2A -0-+0+ 3A -+0-+0

0B +0-+-0 1B 0+-00+ 2B 0--+0+ 3B +0--+0

0C +0-0+- 1C 0-+00+ 2C 0--++0 3C +0-0-+

0D 0-+-0+ 1D 0-+++- 2D --00++ 3D 0-++0-

0E -+0-0+ 1E 0-+0-+ 2E -0-0++ 3E -+0+0-

0F +0--0+ 1F 0-+0+- 2F 0--0++ 3F +0-+0-

 (Cont.)
• 4B/5B signal of 100BASE-TX: (Appendix 7A)
1. NRZI-to-NRZ conversion
2. Scrambling, producing a more uniform
spectrum.
3. Encoder, in MLT-3.
4. Driver
• 4B/5B-NRZI of 100BASE-FX:
Intensity modulation.
 100BASE-T4
• 100BASE-T4 over Category 3 cable.
– A data stream to be transmitted is split up into 3
separate substreams, each with an effective data rate of
33 1/3 Mbps.
– Fig 7.8: Four twisted pairs are used.
• Data are transmitted using pairs labeled D1, D3 and D4.
• Data are received on pairs D2, D3 and D4.
• Pair D2 is also used for collision detection.
– Ternary signal(positive voltage, negative
voltage, zero voltage) is used.
 (Cont.)
• The block-coding scheme 8B6T is used. In
Appendix 7A.
– Each block of 8 bits is mapped into a codegroup of 6
ternary symbols.
– The stream of code groups is transmitted in a round-
robin fashion across the three output channels.
– The ternary transmission rate on each output channel is
6/8×33 1/3=25 Mbaud
– Mapping of 8B6T code is chosen in the consideration:
• Synchronization: Maximize the average number of
transitions per code group.
• Dc balance:
 (Cont.)
– Code groups selected have either an equal
number of positive and negative symbols or an
excess of one positive symbol.
– dc balancing algorithm:
• Monitors the cumulative weight of code groups
transmitted on a single pair.
• Possibly negates a transmitted code group so that the
cumulative weight at the conclusion of each code
group is always 0 or 1. (Fig 7.17)
 6T (25 Mbaud)
Stream of 8-bit 8B (100 Mbps) 8B6T Splitter
bytes Coder

6T (25 MBaud)

6T (25 MBaud)

Figure 7.17 8B6T Transmission Scheme

 Configuration and Operation
• Star-wired topology via multiport repeaters which
function
– A valid signal appearing on any single input is repeated
on all output links.
– If two inputs occur at the same time, a jam signal is
transmitted on all links.
• Stations connected to a multiport repeater are in
the same collision domain, whereas stations are
separated by a bridge are in different collision
domains.
 (Cont.)
• Two types of repeater:
– Class I:
• Supports unlike physical medium segments.
• The repeater needs to handle the conversion from one signaling
scheme to another.
• Only one class I repeater can be used in a collision domain.
– Class II: two class II repeaters can appear in a single
collision domain. (Fig. 7-10)
(Fig 7.9 and Table 7.4)
Collision Domain Bridge Collision Domain

Repeater Repeater

Figure 7.9 100BASE-T Collision Domains

 Class I Repeater

Class II Repeater Class II Repeater

Figure 7.10 100BASE-T Repeater Types

 Full-Duplex Operation
• A traditional Ethernet is half duplex.
• With full-duplex, a station can transmit and
receive simultaneously.
– If a 100-Mbps Ethernet runs in full-duplex mode, the
theoretical transfer rate becomes 200 Mbps.
– The central point in the star wire cannot be a simple
multiport repeater but rather must be a LAN switch.
• Each station constitutes a separate collision domain.
• There are no collisions and the CSMA/CD algorithm is no
longer needed.
• The same 802.3 MAC frame format is still used and the
attached stations can continue to execute the CSMA/CD
algorithm, even though no collisions will ever detected.
 (Cont.)
• For full-duplex operation, IEEE 802.3 introduces
MAC-level flow control, which keeps switches
from losing frames due to buffer overflow.
– A receiver can restrain a transmitter by issuing a pause
frame, which is a short frame containing a timer value.
– The timer value is expressed as a multiple of the
transmission time for 512 bits.
– The pause protocol operates between the two ends of a
point-to-point link.
 Mixed Configuration
• Fig. 7-11:
– 10BASE-T
– 100BASE-T
– Hub
– LAN switches (A building backbone)
– A router, providing connection to an outside
WAN.
 10 Mbps link

100 Mbps link WAN

Router

Server 100 Mbps hubs

10/100
Switches

10 Mbps hubs

Figure 7.11 Example 100-Mbps Ethernet Backbone Strategy

 Autonegotiation
• It allows a LAN switch to support a mixture of
devices that conform to various 100BASE-T and
the 10BASE-T medium options.
• It is performed by transmitting a pulse burst
during idle times on the link, which does not
interfere with normal traffic.
• The pulse burst consists of 33 pulse positions:
– The 17 odd-numbered pulse position contains a link
pulse.
– Each of the 16 even-numbered pulse positions contains
either a pulse to represent binary 1 or no pulse to
represent binary 0.
 (Cont.)
• The 16-bit codeword embedded in the pulse burst:
– Selector field(5 bits): message type (for example, IEEE
803.3)
– Technology ability field (8 bits)
• 100BASE-TX full duplex
• 100BASE-T4
• 100BASE-TX
• 10BASE-T full duplex
• 10BASE-T
– Remote fault bit
– Acknowledge bit
– Next page bit
 GIGABIT Ethernet
• It defines a new medium and transmission
specification but retains CSMA/CD
protocol and frame format of its 10-Mbps
and 100-Mbps predecessors.
• Fig. 7-12, an example.
• Fig 7-13, protocol architecture
– GMII, gigabit medium-independent interface
 Central
100 Mbps link 1 Gbps Servers
switching
1 Gbps link
hub

100/1000-Mbps Hubs

Workgroup Workgroup

Figure 7.12 Example Gigabit Ethernet Configuration

 Medium Access Control
Full Duplex and/or Half Duplex

Gigabit Media-Dependent
Interface (GMII)

4D-PAM5
8B/10B
Encoder/
Encoder/Decoder
Decoder

1300-nm 850-nm Copper UTP

Transceiver Transceiver Transceiver Transceiver
1000BASE-LX 1000BASE-SX 1000BASE-CX 1000BASE-T

Figure 7.13 Gigabit Ethernet Layers

 (Cont.)
• Medium Access Layer
– The same CDMA/CD frame format and MAC protocol.
– For hub operation(half duplex), there are two
enhancements:
• Carrier extension: A set of special symbols is appended to the
end of short MAC frame, so that the resulting block is at least
4096 bit times in duration, up from the minimum 512 bit (64
octet) times imposed at 10 and 100 Mbps.
• Frame bursting: Multiple frames are allowed to be transmitted
consecutively, up to a limit, without relinquishing control for
CSMA/CD between frames. Frame bursting avoids the over
head of carrier extension when a single station has a number of
small frames ready to send.
• Fig 7.14
 10-micron single-mode fiber
1000BASE-LX 50-micron multimode fiber
62.5-micron multimode fiber

50-micron multimode fiber

1000BASE-SX
62.5-micron multimode fiber

1000BASE-T Category5 UTP

1000BASE-CX Shielded cable

25 m 50 m 250 m 500 m 2500 m 5000 m

Maximum distance

Figure 7.14 Gigabit Ethernet Media Options (log scale)

 (Cont.)
• With a LAN switch which provides
dedicated access to the medium, the carrier
extension and frame bursting features are
not needed.
• With a switching technique and full duplex
operation, the CSMA/CD protocol is not
needed.
• Physical Layer (Fig 7-14)
 Input = 0

0,
from +V Input = 1
Input = 1

Input = 0 +V -V Input = 0

Input = 1 0, Input = 1
from -V

Input = 0

Figure 7.15 MLT-3 Encoder State Diagram

 1 0 0 1 1 1 1 0 0 0 0 1 0 1 1 1 0 1 1 1 0

+V
0
-V

Figure 7.16 Example of MLT-3 Encoding

 Parallel Data Byte
Control

Adapter Interface
Clock
A B C D E F G H K

5B/6B 3B/4B
Functions Functions

Disparity
Control

Encoding Switch
Clock
a b c d e i j g h j

10 Binary Lines to Serializer

Figure 7.18 8B/10B Encoding

 5 5
4 4

⊕ ⊕
3 3
2 2
1 1

⊕ ⊕
A B B A

(a) Scrambler (b) Descrambler

Figure 7.19 Scrambler and Descrambler

 1 0 1 1 1 0 0 0 1 1 0 1 0 0 1 B
P 1 0 0 1 0 1 1 0 1 0 1 0 1 0 0 0 0 0 1 1 1 - - - - - A
1 0 0 1 0 1
1 1 1 1 1 0
1 0 0 1 0 1
1 1 0 1 1 0
1 0 0 1 0 1
1 0 0 1 1 0
1 0 0 1 0 1
1 1 0 0 1 1
1 0 0 1 0 1
1 0 1 1 0 1
1 0 0 1 0 1
1 0 0 0 0 0
1 0 0 1 0 1
1 0 1 0 0 0

(a) Scrambling

1 0 1 1 1 0 0 0 1 1 0 1 0 0 1 B
1 0 0 1 0 1 P
1 0 1 1 1 0 0 0 1 1 0 1 0 0 1
1 0 1 1 1 0 0 0 1 1 0 1 0 0 1
1 0 1 1 1 0 0 0 1 1 0 1 0 0 1
C=A 1 0 1 0 1 0 1 0 0 0 0 0 1 1 1

(b) Descrambling

Figure 7.20 Example of Scrambling with P(X) = 1 + X–3 + X–5