Local Area Networks by R.S. Chang, Dept. CSIE, NDHU 1 Chapter 4 The Medium Access Control Sublayer 4.3 Etherent IEEE 802 Group.

Local Area Networks by R.S. Chang, Dept. CSIE, NDHU 1

Chapter 4 The Medium Access Control Sublayer

4.3 Etherent

IEEE 802 Group



4.3 Etherent



IEEE 802.3: 1-persistent CSMA/CD

4.3 Etherent

Classical Ethernet

Switched Ethernet

Fast Ethernet (100Mbps), Gigabit Ethernet, 10G Ethernet



4.3.1 Classical Ethernet Physical Layer

MIT->Harvard->Hawaii->Xerox PARC(Palo Alto Research Center)->Ethernet->3COM

http://www.parc.com/about/




The Xerox Ethernet was so successful that DEC, Intel, and Xerox drew up a standard in 1978 for a 10-Mbps Ethernet, called DIX standard. DIX became IEEE 802.3 in 1983.






Cable topologies. (a) Linear, (b) Spine, (c) Tree, (d) Segmented




To allow larger networks, multiple cables can be connected by repeaters.

A repeater is a physical layer device. It receives, amplifies, and retransmits signals in both directions. As far as the software is concerned, a series of cable segments connected by repeaters is no different than a single cable.




10BASE5 10BASE2 1BASE5 10BROAD36 10BASE-TEthernet Cheaper net StarLAN Broadband Twisted-pair

medium coaxial cable50ohm-10mm

coaxial cable50ohms-5mm

twisted-pairunshielded

coaxial cable75ohms

2 simplex TPunshielded

signals

maximumsegment

nodes persegment

collisiondetection

Notes

10MbpsManch

10MbpsManch

1MbpsManch

10MbpsDPSK

10MbpsManch

500m 185m 500m 1800m 100m

maximumdistance 2.5km 0.925km 2.5km 3.6km 1km

100 30 2

excess current2 active hubinputs

transmission=reception

activity onreceiver andtransmitter

slot time=512 bits; gap time=96 bits; jam=32 to 48 bits



Manchester Encoding




4.3.2 Classical Ethernet MAC Sublayer Protocol

Frame formats. (a) Ethernet (DIX). (b) IEEE 802.3.

>1500 is type, otherwise interprets as length



802.3 frame format

single address

group address

local address

global address

0

1

0

1

multicast (all 1's for broadcast)

No significance outside

one of 246 unique address

4.3.2 Classical Ethernet MAC Sublayer ProtocolThe first 3 bytes are OUI (Organizationally Unique Identifier) (Manufacturer)



802.3 frame format

Minimum frame length: 64 bytes (6+6+2+46+4)





For a 10 Mbps LAN with a maximum length of 2500 meters (with 4 repeaters), the round-trip time is 50 msec in the worst case.

(10M)x(50 msec) =500 bits~512 bits=64 bytes

Checksum= 32-bit CRC=x32+x26+x23+x22+x16+x12+x11+x10+x8+x7+x5+x4+x2+x+1



802.3 frame format

As the network speed goes up, the minimum frame length must go up or the maximum cable length must come down proportionally.

For a 2500-meter LAN operating at 1 Gbps, the minimum frame size would have to be 6400 bytes.

Alternatively, the minimum frame size could be 64 bytes and the maximum distance between any two stations 250 meters.




Ethernet Frame Structure v2 (or DIX Ethernet, for DEC, Intel, Xerox)

preamble SFD DA SA type CRC

synchronizethe receiver

7 1 6 6 2 4

60 to 1514 bytes

start framedelimiter

Cyclic Redundancy Check

Type>0x0600=1536

Data

0800: IPv4 datagram0806: ARP request/reply8035: RARP request/reply86DD: IPv6




The Binary Exponential Backoff Algorithm

If a frame has collided n successive times, where n<16, then thenode chooses a random number K with equal probability from theset {0,1,2,3,...,2m-1} where m=min{10,n}. The node then waits for bit times. (slot time=512 bit time)K 512

after first collision

after second collision

after thirdcollision

select one to start transmission




Acknowledgements

As far as CSMA/CD is concerned, an acknowledgement would be just another frame and would have to fight for channel time just like a data frame.

(What is the problem?)

A simple modification would allow speedy confirmation of frame receipt. All that would be needed is to reserve the first contention slot following successful transmission for the destination station.




Performance

Assume k stations are always ready to transmit and a constant retransmission probability in each slot. (A rigorous analysis of the binary exponential backoff algorithm is complicated.)

If each station transmits during a contention slot with probability p, the probability A that some station acquires the channel in that slot is

. as 1

with,1

whenmaximized is

)1( 1

k/eAk

pA

pkpA k

4.3.3 Ethernet Performance



Performance

The probability that the contention interval has exactly j slots in it is A(1-A)j-1, so the mean number of slots per contention is given by

AAjA

j

j 1)1(

0

1

Since each slot has a duration 2t, the mean contention interval, w, is 2t/A. Assuming optimal p, the mean number of contention slots is never more than e, so w is at most 2te5.4t.




Performance

If the mean frame takes P sec to transmit, when many stations have frames to send, channel efficiency=

AP

P

/2Here we see where the maximum cable distance between any two stations enters into the performance figures. The longer the cable, the longer the contention interval. By allowing no more than 2.5km of cable and four repeaters between any two transceivers, the round-trip time can be bounded to 51.2 msec, which at 10Mbps corresponds to 512 bits or 64 bytes, the minimum frame size.




Performance

Let P=F/B (frame_length/bandwidth) and t=L/C (cable_length/signal_propagation_speed). For the optimal case of e contention slots per frame, channel efficiency=

cFBLe /21

1

Increasing network bandwidth or distance (the BL product) reduces efficiency for a given frame size. Unfortunately, much research on network hardware is aimed precisely at increasing this product. People want high bandwidth over long distances, which suggests that 802.3 may not be the best system for these applications.







Many theoretical analysis assume the input traffic is Poisson. It now appears that network traffic is rarely Poisson, but self-similar. What this means is that averaging over long periods of time does not smooth out the traffic.

The average number of packets in each minute of an hour has as much variance as the average number of packets in each second of s minute.

The consequence of this discovery is that most models of network traffic do not apply to the real world and should be taken with a grain of salt.




4.3.4 Switched Ethernet

(a) Hub. (b) Switch.

Not necessarily this kind of wiring

Must know which station is in which port

Just like a single cable Ethernet



4.3.4 Switched Ethernet

An Ethernet switch.



The three primary reasons that the 803 committee decided togo with a souped-up 802.3 LAN (instead of a totally new one) were:1. The need to be backward compatible with thousands of

existing LANs.2. The fear that a new protocol might have unforeseen

problems.3. The desire to get the job done before the technology

changed.

Fast Ethernet

4.3.5 Fast Ethernet



The basic idea behind fast Ethernet was simple: keep all the old packet formats, interfaces, and procedural rules, but just reduce the bit time form 100 nsec to 10 nsec.

Technically, it would have been possible to copy 10Base5 or 10Base2 and still detect collisions on time by just reducing the maximum cable length by a factor of ten.

However, the advantages of 10BaseT wiring were so overwhelming that fast Ethernet is based entirely on this design. Thus all fast Ethernet systems use hubs and switches.

Fast Ethernet

4.3.5 Fast Ethernet



The category 3 UTP scheme, called 100Base-T4, uses a signaling speed of 25 MHz, only 25 percent faster than standard 802.3’s 20 MHz. To achieve the necessary bandwidth, 100BaseT4 requires four twisted pairs.

Fast Ethernet

4.3.5 Fast Ethernet



Of the four twisted pairs, one is always to the hub, one is always from the hub, and the other two are switchable to the current transmission direction.

To get the necessary bandwidth, Manchester encoding is not used, but with modern clocks and such short distances, it is no longer needed.

Fast Ethernet

4.3.5 Fast Ethernet



Ternary signals are sent, so that during a single clock period the wire can contain a 0, a 1, or a 2. With three twisted pairs going in the forward direction and ternary signaling, any one of the 27 possible symbols can be transmitted, making it possible to send 4 bits with some redundancy. Transmitting 4 bits in each of the 25 million clock cycles per second gives the necessary 100 Mbps.

In addition, there is always a 33.3 Mbps (100/3) reverse channel using the remaining twisted pair.

Fast Ethernet

4.3.5 Fast Ethernet



For category 5 wiring, the design, 100Base-TX, is simpler because the wires can handle clock rates up to 125 MHz and beyond. Only two twisted pairs per station are used, one to the hub and one from it.

Rather than just use straight binary coding, a scheme called 4B5B is used at 125 MHz. Every group of 5 clock periods is used to send 4 bits in order to give some redundancy, provide enough transitions to allow easy clock synchronization, create unique patterns for frame delimiting, and be compatible with FDDI in the physical layer.

Fast Ethernet

4.3.5 Fast Ethernet



Consequently, 100Base-TX is a full-duplex system; stations can transmit at 100 Mbps and receive at 100 Mbps at the same time. Often 100Base-TX and 100Base-T4 are collectively referred as 100Base-T.

The last option, 100Base-FX, uses two strands of multimode fiber, one for each direction, so it, too, is full duplex with 100 Mbps in each direction. In addition, the distance between a station and the hub can be up to 2 km.

Fast Ethernet

4.3.5 Fast Ethernet



Gigabit Ethernet

The ink was barely dry on the fast Ethernet standard when the 802 committee bagan working on a yet faster Ethernet. It was quickly dubbed gigabit Ethernet and was ratified by IEEE in 1999 under the name 802.3ab.

An important design goal: remain backward compatibility

4.3.6 Gigabit Ethernet



Gigabit Ethernet

All configurations of gigabit Ethernet are point-to-point.

Each individual Ethernet cable has exactly two devices on it, no more and no fewer.



Chapter 4 The Medium Access Control SublayerChapter 4 The Medium Access Sublayer

Gigabit Ethernet

Two different modes of operation: full duplex and half duplex

The normal mode is full-duplex used when computers are connected to a switch.

The sender does not have to sense the channel to see if anybody else is using it because contention is impossible. So CSMA/CD protocol is not used.

So the maximum length of the cable is determined by signal strength issues rather than by the collision detection issue.




Gigabit Ethernet

Half-duplex is used when the computers are connected to a hub. A hub does not buffer incoming frames. So collisions are possible and CSMA/CD is required.

But now the transmission time for a 64-byte frame is 100 times faster. So the distance is 100 times less than Ethernet. That is, only 25 meters.

The 802.3ab committee considered a radius of 25 meters to be unacceptable and added two features to the standard to increase the radius.




Gigabit Ethernet

The first feature, called carrier extension, essentially tells the hardware to add its own padding to extend the frame to 512 bytes. Of course, using 512 bytes to transmit 64 bytes of data has a line efficiency of 9%.

The second feature, called frame bursting, allows a sender to transmit a concatenated sequence of multiple frames in a single transmission. If the total length is less than 512 bytes, the hardware pads it again.

Just for backward compatibility. Most will use switches.




Gigabit Ethernet

Cabling

Gigabit Ethernet uses new encoding rules on the fiber. Manchester encoding at 1Gbps would require 2G baud signal, too difficult and too wasteful.




Gigabit Ethernet

8B/10B is used. Each 8-bit byte is encoded as 10 bits.256 out of 1024. Two rules are used:1. No codeword may have more than four identical bits in a row.2. No codeword may have more than six 0s or six 1s.

In addition, many input bytes have two possible codewords assigned to them. When there is a choice, the encoder always chooses the one that tries to equalize the number of 0s and 1s transmitted so far.




Gigabit Ethernet

1000Base-T uses a different encoding scheme since clocking data onto copper wire in 1 nsec is too difficult.

The solution uses four category 5 twisted pairs to allow four symbols to be transmitted in parallel.

Each symbol is encoded using one of five voltage levels. This scheme allows a single symbol to encode 00, 01, 10, 11, or a special value for control purposes.

The clock runs at 125MHz, allowing 1-Gbps operation.




Gigabit Ethernet

Gigabit Ethernet supports flow control which consists of one end sending a special control frame to the other end telling it to pause for some period of time.

For gigabit Ethernet, the time unit for pause is 512 nsec. The maximum is 33.6 msec.




4.3.7 10-Gigabit Ethernet

10 Gigabit Ethernet cabling



4.3.8 Retrospective on Ethernet

Ethernet has been around for over 30 years and has no serious competitions.

Few CPU architectures, operating systems, or programming languages have been king of the mountain for three decades going on strong.

Clearly, Ethernet did something right. What?




Simple and Flexible

Simple translates into reliable, cheap, and easy to maintain.

Ethernet interworks easily with TCP/IP, which has become dominant. (Both are connectionless)

Lastly and perhaps the most importantly, Ethernet has been able to evolve in certain crucial ways



KISS

• In anyway,




What I see in the Korea Customs:

Korea Immigration Smart Service




4.4 Wireless LANS



4.4.1 The 802.11 Architecture and Protocol Stack

802.11 architecture – infrastructure mode

AccessPoint

Client

To Network




802.11 architecture – ad-hoc mode




Part of the 802.11 protocol stack.






4.4.2 The 802.11 Physical Layer

Rate adaptation: Reduce rate if signal is bad

OFDM (Orthogonal Frequency Division Multiplexing)



4.4.2 The 802.11 Physical Layer

MIMO (Multiple Input Multiple Output)

Note that the terms input and output refer to the radio channel carrying the signal, not to the devices having antennas.



4.4.3 The 802.11 MAC Sublayer Protocol

CSMA/CA: CSMA with Collision Avoidance

DCF: Distributed Coordination Function



Two modes of operation:

DCF: distributed coordination function, no central control

PCF: point coordination function, the base station controls all activity in its cell

All implementations must support DCF but PCF is optional.





When DCF is employed, 802.11 uses a protocol called CSMA/CA (CSMA with Collision Avoidance). Two methods of operation are supported by CSMA/CA.

In the first method:1. When a station wants to transmit, it senses the channel. If it

is idle, it just starts transmitting.2. If the channel is busy, the sender defers until it is idle and

then starts transmitting.3. It does not sense the channel while transmitting.4. If a collision occurs, the colliding stations wait a random

time, using Ethernet binary exponential backoff algorithm, and try again later.





(a) The hidden station problem(b) The exposed station problem





Virtual channel sensing.

A wants to send to B. C is within range of A. D is within range of B, but not A. (NAV: network allocation vector)





Wireless networks are noisy and unreliable. If a frame is too long, it has very little chance of getting through undamaged. So 802.11 allows frames to be fragmented into smaller pieces, each with its own checksum.

Stop and Wait is used.





In PCF, the base stations polls the other stations, asking them if they have any frames to send.

The basic mechanism is for the base station to broadcast a beacon frame periodically (10 to 100 times per second).

Battery life is always an issue with mobile devices, so in 802.11, the base station can direct a mobile station to go into sleep until explicitly awakened by the base station or the user.





PCF and DCF can coexist within one cell.

SIFS: Short InterFrame Spacing, PIFS: PCF IFS, DIFS: DCF IFS, EIFS: Extended IFS









802.11 Frame Structure4.4.4 The 802.11 Frame Structure

Format of the 802.11 data frame



• Frame Control Field : – Retry: Indicates that the frame is a retransmission of an

earlier frame. – To DS, From DS (DS=Distribution System, meaning AP)– More Fragment, More Data– Power Management :Active Mode, PS Mode (Power Save)– Protected: Data are encypted– Order: Frame must arrive in order

• Duration: how long the frame and ack will control the channel (NAV)

• Address 3: the distant endpoint





Frame Control

Duration RA TA FCS

MAC Header

RTS Frame

Frame Control

Duration RA FCS

MAC Header

CTS Frame



Five distribution services and four station services

Five distribution services:1. Association: connect to a base station2. Disassociation: break the association either by the base

station or the station3. Reassociation: change preferred base station4. Distribution: how to route frames sent to the base station5. Integration: translate from 802.11 to non-802.11 (in

address scheme or frame format)

4.4.5 The 802.11 Services



Four intercell station services

1. Authentication: a station proves its knowledge of the secret key by encrypting the challenge frame and sending it back to the base station

2. Deauthentication3. Privacy: manage the encryption and decryption using RC44. Data delivery

4.4.5 The 802.11 Services



4.5 Broadband Wireless

802.16: Broadband Wireless Access

Running fiber, coaxial, or even category 5 twisted pair to millions of homes and businesses is prohibitively expensive! What is a competitor can do?

The wireless local loopThe wireless last mileThe wireless MAN (metropolitan area network)



WiMAX, the Worldwide Interoperability for Microwave Access, is a telecommunications technology aimed at providing wireless data over long distances in a variety of ways, from point-to-point links to full mobile cellular type access.

It is based on the IEEE 802.16 standard, which is also called Wireless MAN. The name WiMAX was created by the WiMAX Forum, which was formed in June 2001 to promote conformance and interoperability of the standard. The forum describes WiMAX as "a standards-based technology enabling the delivery of last mile wireless broadband access as an alternative to cable and DSL."





The 802.16 architecture




Comparison of 802.11 with 802.16

1. 802.16 provides service to buildings, and buildings are not mobile.

2. Buildings can have more than one computer in them.3. Better radios are available for buildings. So 802.16 can use

full-duplex communications.4. In 802.16, the distances involved can be several kilometers,

affect signal-to-noise ratio and need security and privacy.5. More bandwidth is needed. Hence 802.16 has to operate in

higher 10-66 GHz band, thus require a completely different physical layer.

6. Error handling is much more important in 802.16.7. 802.16 should support QoS.




Comparison of 3G with 802.16

The next step of 3G is 4G, using LTE (Long Term Evolution).

It appears that LTE has prevailed over WiMax.




The 802.16 protocol stack




The 802.16 physical layer

Frames and time slots for time division duplexing.




The 802.16 MAC layer

All connection-oriented services

4 Service Classes:Constant bit rate serviceReal-time variable bit rate serviceNon-real-time variable bit rate serviceBest efforts service




The 802.16 frame structure

(a) A generic frame. (b) A bandwidth request frame.

Encrypted or not Final checksum present or not

Encryption key



4.6 Bluetooth

Bluetooth is an industrial specification for wireless personal area networks (PANs). Bluetooth provides a way to connect and exchange information between devices such as mobile phones, laptops, PCs, printers, digital cameras, and video game consoles over a secure, globally unlicensed short-range radio frequency. The Bluetooth specifications are developed and licensed by the Bluetooth Special Interest Group.



4.6 Bluetooth

Architecture

Piconets can be connected to form a scatternet.

10 meters

7 active slaves and 255 parked nodes


Chapter 4 The Medium Access Control SublayerChapter 4 The Medium Access Sublayer

4.6 Bluetooth

Profiles



4.6 Bluetooth

Protocol stack

The Bluetooth protocol architecture.

Logical Link Control Adaptation Protocol



4.6 Bluetooth

• SCO (synchronous connection oriented)– fixed-bandwidth channel between a master and

a slave– slots spaced by regular intervals– up to 3 SCO links per master– SCO packets are never retransmitted!

• bandwidth-guaranteed, but not error-free-guaranteed



4.6 Bluetooth

• ACL (asynchronous connectionless)– a point-to-multipoint link between a master and

ALL its slaves– only on slots NOT reserved for SCO links

• but the communication can include a slave already involves in a SCO link

– packet retransmission is applicable. • packet-switched style

– a slave can send only when it is addressed in the previous master-initiated slot



4.6 Bluetooth

Detailed Connecting Steps

• inquiry:– used by master to find the identities of devices within

range• inquiry scan:

– listening for an inquiry message• page:

– used by master to send PAGE message to connect to a slave by transmitting slave’s device address code (DAC)

• page scan:– slave listening for a paging packet with its DAC



Slave’s Four Mode in Connection State

4.6 Bluetooth

• Active:– actively participates in the piconet by listening,

transmitting, and receiving packets.– the master periodically transmits to the slave to

maintain synchronization

• Sniff:– only wake up in specific slots, and go to

reduced-power mode in the rest of slots




4.6 Bluetooth

• Hold:– goes to reduced-power mode and does not

support ACL link any more• may still participate in SCO exchanges

– while in reduced-power mode, the slave may participate in another piconet




4.6 Bluetooth

• Park:– does not participate in the piconet

• but still wants to remain as a member and remain time-synchronized

– the slave gets a parking member address (PM_ADDR), and loses its AM_ADDR

– by so doing, a piconet can have > 7 slaves



4.6 Bluetooth

Typical Bluetooth data frame at (a) basic, and (b) enhanced, data rates.

Flow control (slave buffer full)

Piggyback ack

Stop-and-wait sequence number



4.7 RFID (Radio Frequency Identification)

• A means of storing and retrieving data through electromagnetic transmission to an RF compatible integrated circuit.

• Basic components: – RFID readers: read data emitted from RFID tags– RFID tags: use a defined radio frequency and

protocol to transmit and receive data• Passive: without a battery, reflect the RF signal

transmitted to them from a reader and add information by modulating the reflected signal, replace barcode, less expensive, unlimited operational lifetime, but read ranges are limited

• Active: contain both a radio transceiver and a button-cell battery to power the transceiver, longer range
















An Electronic Product Code (EPC) is one common type of data stored in a tag




技術應用




實際運用上的問題




Example message exchange to identify a tag




Format of the Query message

Define the range of slots over which tags will respond, from 0 to 2Q-1



A

Bridge

CLANs can be connected by devices called bridges, which operate in the data link layer. Bridges do not examine the network layer header.

4.8 Datalink Layer Switching



A

Router

C

RouterIn contrast, a router examines network layer headers.





Multiple LANs connected by a backbone to handle a total load higher than the capacity of a single LAN.



Why a single organization may end up with multiple LANs? (to need bridges)

1. Autonomy of departments to choose their own types of LANs2. Cheaper to have separate LANs than to run a single large LANs3. Load splitting4. Physical distance is too great. (For example, >2.5km in 802.3)5. More reliable6. More secure





Bridge connecting two multidrop LANs

Cut-through switching (wormhole routing)




Bridges (and a hub) connecting seven point-to-point stations

Backward learning algorithm: learn and forget




Operation of a LAN bridge from 802.11 to 802.3.




Bridges with two parallel links




A spanning tree connecting five bridges. The dotted lines are links that are not part of the spanning tree




(a) Which device is in which layer. (b) Frames, packets, and headers




A building with centralized wiring using hubs and a switch




Two VLANs, gray and white, on a bridged LAN




The IEEE 802.1Q Standard

Bridged LAN that is only partly VLAN-aware. The shaded symbols are VLAN aware. The empty ones are not




The IEEE 802.1Q Standard

The 802.3 (legacy) and 802.1Q Ethernet frame formats



Exercises:

Page 368: Problem1Page 369: Problems 4, 9, 11Page 370: Problems 15, 25Page 371: Problems 35, 39

Local Area Networks by R.S. Chang, Dept. CSIE, NDHU 1 Chapter 4 The Medium Access Control Sublayer 4.3 Etherent IEEE 802 Group.

Documents

local area networks

ethernet fast ethernet

mbps ethernet

xerox ethernet

ethernet dix

dix ethernet

g ethernet slide

etherent classical ethernet