Presentation

Download Report

Transcript Presentation

CCM 4300 Lecture 2
Computer Networks, Wireless
and Mobile Communication
Systems
Dr E. Ever
School of Engineering Information
Sciences
CCM4300_09-10
1
Group Presentation
Topics
 Wi-Fi
Bluetooth
GPRS
GSM
Wired Equivalent Privacy (WEP)
Wi-Fi Protected Access (WPA and
WPA2)
Kerberos is a computer network
authentication protocol
Beowulf Clusters
Myrinet-based cluster
Supercomputer (Cray, or any other one)
CCM4300_Jan 10
2
CISCO Training




the link
http://www.cs.mdx.ac.uk/cnap/index.html
User name: Cisco
PWD: class
CCM4300_Sep 08
3
Session Content
 Recap of last session
 Lesson Objectives
 Network Topologies
 Introduction – an example of human-to-human interaction
 What is a protocol?
 ISO OSI Reference Model
- TCP/IP protocol
 Physical layer
Data link layer and access control
CCM4300
4
Recap of Last Session
defined a computer network and identified some of
the basic components.
 explored the history of computers and computer
networks and how they have evolved.
 identified some of the advantages and disadvantages
of using computer networks.
 introduced the various network standards, how they
are created and by whom
CCM 4300
5
Lesson objectives
 At the completion of this lesson you should be able to
- understand network logical and physical topologies
- understand the concept of layering and structure of
the OSI Reference Model
- describe the functionality of each layer in the OSI
Reference Model
- understand the difference between analogue and
digital communication
- understand the OSI Reference Model Physical
Layer specification
- understand the OSI Reference Model Data Link
Layer specification
CCM4300
6
Physical and Logical Topologies
The word topology means maps of nodes (dots)
and links (lines) that often contain patterns.
There are two types physical and logical

Physical locations describes the plan for wiring the
physical devices.
 A logical topology of a network shows how the hosts
communicate across the medium.

Note that a network may have one type of
physical topology and a totally different type of
logical topology.

CCM4300
7
Bus
 Each computer is connected to a single cable which
connects all of the computers.
 This is the cheapest network topology as the smallest
amount of cabling is required.
 If the network cable breaks anywhere then none of the
computers can access the network.
CCM4300
8
Ring
 Each computer is connected to the two computers on
either side of it.
 The last computer is linked to the first to form a ring.
 If the network cable breaks anywhere then none of the
computers can access the network.
 Long cable length
CCM4300
9
Mesh
 Any computer can be connected to any other computer.
 There are multiple routes. If one link fails messages can
go another way so this topology is very reliable.
 The Internet has a mesh topology.
CCM4300
10
Star
 All communication takes place via a central computer.
 If the central computer fails the whole network will break
down.
 If one of the network cables breaks only the computer
connected to that cable is affected.
CCM4300
11
Extended Star
 An extended star topology is like a star
topology except that each device that links
to the centre device is also the centre of
another star.
Each of these devices acts as the centre of
another star.
An apparent advantage of this is to extend
the length and the size of the network.
The number of devices and the numbers of
cables needed to interconnect to the central
device is still limited (QoS issues).
CCM4300
12
The Network Core: CS vs PS


mesh of interconnected
routers
the fundamental
question: how is data
transferred through net?
 circuit switching:
dedicated circuit per
call: telephone net
 packet-switching: data
sent thru net in
discrete “chunks”
CCM4300
13
What is a Protocol?
 Represent the “rules” for communication:
 how do the computers initiate communication
 what features can be negotiated at the start
 the communication vocabulary
- i.e. what requests (commands) can be
given and what the valid responses are
 what kinds of data can be exchanged
 how the communication can be closed in an
orderly way
CCM4300
14
What is a protocol? - cont
 A human protocol and a computer network protocol:
Hi
TCP connection
req.
Hi
TCP connection
reply.
Got the
time?
2:00
Get http://www.mdx.ac.uk
<file>
time
CCM4300
15
What is a protocol? - cont
Protocols also define the format of the communication
exchanges

Header
Control
Information
Data
Message
CCM4300
16
What is a protocol? - cont



For two networked devices to communicate they
must speak the same language (i.e. use the same
protocol)
The protocols must be able to:
- indicate who (what address) they want to talk to
- provide any required delivery
assurances/recovery
- control the flow of information
Must specify how to initiate, maintain and conclude
the communications exchange
CCM4300
17
What is a protocol? - cont


Several protocols are involved in a network
Operate together in a layered manner
- each layer builds upon the services of its lower
layer(s)
File Transfer
Workstation
End-to-end integrity
Router
Internetworking
Remote
Server
WAN
LAN
CCM4300
18
What is a protocol? - cont
A protocol (or more typically a protocol suite) is
needed for communications
- some organisations may use multiple suites
(e.g. Novell and TCP/IP protocols)
 Protocols are specified in the form of documents
and usually implemented in software
 A separate software package is needed for each
different protocol suite to be supported

CCM4300
19
ISO OSI Reference Model





Open Systems Interconnection
Developed by International Standards
Organisation (ISO) - 1981
- revised in 1994
Described in ISO-7498 standard
Proposes 7 layers
Provides:
- a common terminology
- a framework for networking
CCM4300
20
Principles of OSI Layering





A layer should be created where a different level of
abstraction is needed
Each layer should perform a well-defined function
The function of each layer should be chosen with a
view to defining internationally standardised protocols
The layer boundaries should be chosen to minimise
the information flow across the interfaces
The number of layers should be:
- large enough that distinct functions are not
thrown together
- small enough that the architecture is not
unwieldy
CCM4300
21
OSI Reference Model
Source :
PDU: Protocol Data Unit
CCM4300_Sep 08
Computer Networks
1996
22
OSI Layers - cont

Application
contains a number of standard protocols of
general use
-e.g. file transfer protocols, email, virtual
terminals
- any service program may define an application
level protocol that clients must use.
-
CCM4300
23
OSI Layers - cont
Presentation
- concerned with the representation of
data (e.g. between different hardware,
Operating Systems etc.)

Session
- allows clients of an OS on one machine
to establish and use sessions with clients
of an OS on another machine
CCM4300
24
OSI Layers - cont

Transport
- concerned with transmission from end
system to end system

Network
- concerned with transmitting data from a
source to destination across networks
- must determine route for data packets
and attempt to avoid congestion by
controlling the number transmitted
CCM4300
25
OSI Layers

Data Link
- Concerned with taking a raw
transmission facility and turning it into a
link that appears to be free from errors

Physical
- Concerned with transmitting
uninterrupted bits from on computer to
another and managing the connection
A common mnemonic device for remembering the
layers in the right order is

All People Seem To Need Data Processing
CCM4300
26
OSI Layer Services
Layer
Number
7
6
5
4
3
2
1
Layer Name
Layer Services
Application
Presentation
Session
Transport
Network
Data Link
Support for e-mail, file transfer etc.
Data representation
Control the dialog
End-to-end data integrity
Internet addressing, routing and segmentation
Package bits into frames and control their
delivery
Adapt bits for transmission over the medium
Physical
CCM4300
27
Question?
If OSI is a Reference Model then what is a
network architecture?

A set of layers and protocol is called a
network architecture. It defines
communication protocols, message formats,
and standards required for interoperability.
CCM4300
28
OSI Central Concepts
Service vs. Interface Vs. Protocol
Service – what is done
- defines what the layer does (but not how
entities above access it or how it works)
 Interface – how it is called above
- tells the processes above how to access it
- specifies what the parameters are and the
results to expect
 Protocol – how it “talks” to its peer layer
- how a layer works (i.e. provides the offered
services)

CCM4300
29
Advantages of Layering




Standard interfaces between layers
- allows internal developments within a particular
layer to evolve
Alternative services may be offered at a given layer
- via different options or routes through the layer
Internal mechanisms of each layer are invisible to the
other layers
Layers may be completely removed if not required, or
substituted by simpler versions
CCM4300
30
Problems of OSI
The following factors limited adoption of OSI in
practice:

Timing
- TCP/IP protocols were already in widespread use

Technology
- the 7 layers are not optimal (e.g. Session and
Presentation layers hardly perform any function)
CCM4300
31
Problems of OSI - cont

Implementation
- initial implementations were unwieldy and slow led to a lasting bad reputation

Politics
- perceived as bureaucratic organisations
attempting to impose inferior standard
BUT

valuable as a conceptual architecture
- a ‘reference model’ for comparison purposes
CCM4300
32
Mapping onto OSI
Source :
Computer Networks
2003
CCM4300
33
TCP/IP Protocols
Source :
Computer Networks
2003
CCM4300
34
OSI & TCP/IP Protocols

OSI
- model useful as a means of discussing
computer networks and educational purposes
- protocols have not become popular yet

TCP/IP
- model practically non-existent
- protocols very widely used
CCM4300
35
Question?
A. application, transport, internet, host-to-network
B. application, internet, transport, host-to-network
C. application, presentation, session, network, transport, data
link, physical
D. application, presentation, session, transport, network, data
link, physical
CCM4300_09-10
36
Client/Server Paradigm
• A client server system is more structured than
general distributed computing
• A client sends request to servers to execute tasks
• The tasks may be just to provide information, or to
perform a complex computation (perhaps returning
data, results, etc)
• A client and servers are asymmetric
• A server may be a client of another server
CCM4300
37
Client/Server Properties
• Clients and servers are separate processes
• They may run on the same or different machines
• Each process can hide internal information
• Each process can implement its own set of
business rules (integrity)
• They communicate by peer-to-peer protocols
CCM4300
38
Physical Layer
It defines everything that is required to support the
transmission and reception of signals (i.e. 1s and 0s)

The Physical layer has four functional areas:
1. Electrical – signal type, amplitude, etc
2. Mechanical – connectors, cabling, etc
3. Procedural – control and timing
4. Functional - requirements for activating, maintaining, and
deactivating a physical link between end systems.

CCM4300
39
Physical Layer - cont
The physical layer is usually a combination of software
and hardware programming and may include
electromechanical devices.

All wiring, power, cabling and connections are part of the
physical layer. Without the physical layer functioning
properly none of the upper layers will respond correctly.

It has no mechanism for determining the significance of
the bits it transmits or receives. The onus for this is
passed on to higher layer protocols

CCM4300
40
Question?
If you were given one word to describe the
physical layer what would it be?
hint. Think like an electrical engineer!
CCM4300
41
Physical Layer Function
For transmission, the physical layer generally:
 convert framed data from Data Link Layer to a
binary stream
 transmit framed data serially (that is, one bit at
a time) as a binary system
For reception, the physical layer generally:
 listens for inbound transmission that are
addressed to its host device
 accept appropriately addressed streams
 pass the binary stream up to the Data Link
Layer for reassembly into frames
42
Analogue or Digital?
The term analogue refers to any physical device or signal
that can continuously vary in strength or quantity, for
example, voltage in a circuit

CCM4300
43
Analogue or Digital? - cont
The term digital refers to any physical
device or signal that is coded in a binary
form (i.e. 1s and 0s)
CCM4300

Source :
Data Communications,
Computer Networks and
Open Systems
44
1992
What is “speed” or capacity?
In analogue communication bandwidth is the total
capacity (or theoretical capacity) of a communication
channel.
bandwidth = highest frequency – lowest frequency

The greater the bandwidth, the more signals
that can be carried
 Example:
Typical ordinary telephone lines (often called a voicegrade line) transmit frequencies from 300Hz to
3300Hz.
bandwidth = 3300Hz – 300Hz => 3000Hz, or 3kHz
CCM4300
45
What is “speed” or capacity? - cont
In digital communication, bandwidth is referred to as
data rate

Data rate – amount of data that can be transmitted over
a communications medium in a given period.

Data rates measured in bits per second (bps) an can
vary considerably from one type of channel to another.

For example, the bandwidth of dialup connections using
a modem ranges from 300bps to 33,600bps (33.6kbps) or
56kbps.

CCM4300
46
What is “speed” or capacity? - cont
If we measure data rates in bits per second, then what is
baud rate?

The speed in baud (symbol rate) is equal to the number of
times the line condition (i.e. frequency, amplitude, voltage, or
phase) changes per second. I.e., the number of distinctive
events per sec.
 Named after French engineer Jean Maurice Emile Baudot
(1845 – 1903)
 For example a communication channel transmitting at 2400
baud. If each signal is used to represent one bit, then the baud
rate is equal to the data rate - 2400bps.
 If each signal represents four bits, then the baud rate – 2400,
but the data rate is 4 X 2400bps = 9600bps.

CCM4300
47
What is “speed” or capacity? - cont

Is there any difference between bandwidth and throughput?
Bandwidth represents a theoretical capacity of a
communications channel.


The “reality rate” is known as throughput.
Just because a medium or LAN architecture is specified to
operate at a certain data rate, it is not a valid assumption to
assume that this rate will be the actual throughput achieved.

CCM4300
48
Transmission medium

Transmission medium can be:
 Simplex
- transmission in one direction only
 Half-duplex
- transmission in both direction; but not at the same
time
 Full-duplex (duplex)
- simultaneous transmission in both directions.
CCM4300
49
Twisted Pair
Twisted pair cabling comes in two varieties: shielded and
unshielded. Unshielded twisted pair (UTP) is the most popular
 Two insulated wires are twisted around each other, and
combined with others into cable
 Each pair is twisted with a different number of twists per inch eliminates interference from adjacent pairs and other electrical
devices.

RJ-45 Connector
Unshielded twisted pair
CCM4300
50
Twisted Pair - cont

Several techniques can be used to improve throughput:
 Increase the thickness of the conductor
 Increase the twist of rate
 Use several different twist rates in bundle of multiple pairs
 Shield the pairs with a metallic barrier
Type
Use
Category 1
Voice Only (Telephone Wire)
Category 2
Data to 4 Mbps (LocalTalk)
Category 3
Data to 10 Mbps (Ethernet)
Category 4
Data to 20 Mbps (16 Mbps Token Ring)
Category 5
Data to 100 Mbps (Fast Ethernet)
Cat 6a: Suitable for 10GBase-T. Defined up to 500 MHz.
51
Coaxial Cable
Coaxial cabling has a single copper conductor at its center
 A plastic layer provides insulation between the center
conductor and a braided metal shield.
 The metal shield helps to block any outside interference
from fluorescent lights, motors, and other computers

Bayone-Neill-Concelman (BNC)
Coaxial cable
CCM4300_09-10
52
Coaxial Cable
The two types of coaxial cabling are thick coaxial and thin
coaxial (refers to diameter - 0.25inch and 0.5inch).

- Thin coaxial cable is also referred to as thinnet. 10Base2
refers to the specifications for thin coaxial cable carrying
Ethernet signals. The 2 refers to the approximate maximum
segment length being 200 meters.
- Thick coaxial cable is also referred to as thicknet. 10Base5
refers to the specifications for thick coaxial cable carrying
Ethernet signals.
CCM4300_09-10
53
Optical Fibre
Optical fiber is a thin, flexible medium capable of conducting
an optical ray
 It transmits light rather than electronic signals eliminating the
problem of electrical interference
 Very high bandwidth (currently up to 100Gbps)
 Used for long-distance trunks, local area networks, highspeed transmissions

Fibre Optic Cable
SC Connector
CCM4300_09-10
ST Connector
54
Question?
Which of the following is not defined at the physical layer of
the OSI reference model?

A. hardware addresses
B. bitstream transmission
C. voltage levels
D. physical interface
CCM4300_09-10
55
Data Link Layer
Regulates and format transmission of data from software on
a node to the network cabling facilities.
 It acts like a “Glue” between the wire and the software on a
node.
 Some of the services the data link layer provides to the
network layer include:
 framing – involves partitioning data into frames with
recognized frame boundaries and exchange these
frames over the link
 frame sequencing – involves maintaining the correct
ordering of frames as they are being exchanged
 establishing and maintaining an acceptable level of flow
control as frames are being exchanged across a link

CCM4300_09-10
56
Data Link Layer - cont
 detecting (and possibly correcting) errors in the
physical layer, which includes error notification when errors
are detected but not corrected
 selecting quality of services (QoS) parameters
associated – ensuring sufficient bandwidth is available and
that transmission delays (i.e latency) are predictable and
guaranteed.
the data link layer enables data frames to be
transmitted error-free between two end nodes
over the physical layer
CCM4300_09-10
57
Data Link Layer - cont
How is the data link layer implemented within a
network?

 Typically implemented on a node as device drive (i.e.
firmware layer of the network interface card), which is a
software component that is specified to both a piece of
hardware (e.g. network interface card), and the
operating system of the computer in which it is installed.
CCM4300_09-10
58
OSI Reference Model vs IEEE 802 Model
IEEE initiated its development of the LAN standards
with an architectural model, defined in IEEE 802.1

The architectural model corresponds to the two
lowest layers of the OSI Model with the following
differences:

 The IEEE divides OSI’s data link layer into two parts –
the logical link control (LLC) and the medium access
control (MAC) sublayers
Note:
The MAC sublayer has nothing to do with Apple
Computer’s Machintosh
CCM4300_09-10
59
OSI Reference Model vs IEEE 802 Modle - cont
Network Layer
Network Layer
Logical Link Control
Data Link Layer
Physical Layer
Media Access Control
OSI Model
Physical Layer
IEEE 802 Model
CCM4300_09-10
60
LLC and MAC
The LLC sublayer (i.e. upper half of data link layer)
encompass several functions – framing, flow control and error
control

The MAC sublayer (i.e. lower half of data link layer) provides
media access management protocols for accessing a shared
medium.

CCM4300_09-10
61
Logical Link Layer - Framing
Framing enables synchronize the transmission and reception
of data since frames have detectable boundaries.
 Integrity of frames - detection and correction

Data set to be transmitted:
1101110011011011
Thus, the frame to be transmitted is:
01111110
Start of frame
1101110011011011
User Data
CCM4300_09-10
01111110
End of Frame
62
Logical Link Layer – Error Control
The term error control refers to the process of guaranteeing
reliable data delivery


Two basic strategies:
 error control through retransmission (also known as errordetecting codes) – provides enough information in the data
stream to detect errors during transmission (e.g. parity,
cyclic redundancy check (CRC))
 autonomous error correction (also known as errorcorrecting codes) – provides redundant information in the
data stream to detect and correct any errors autonomously
(e.g. hamming distance)
CCM4300_09-10
63
Media Access Control
The MAC sublayer provides the protocol that define the
manner in which nodes share the single physical transmission
medium.

The IEEE 802 specifications recognizes three different forms
of media access:
 Contention
 Demand priority (not so common anymore)
 Token passing

CCM4300_09-10
64
Media Access Control - Contention
Contention based media access is embodied in the Carrier
Sense, Multiple Access with Collision Detection (CSMA/CD)
scheme 802.3
 As its name implies:
- requires station to check the wire to determine whether
any other station is already sending data
- If the station can sense an available carrier signal on a
wire, it is free to transmit
 The sensing of carrier signal does not necessarily guarantee
a free transmission media – collide with previously transmitted
signal on what appeared to be an idle medium.

CCM4300_09-10
65
Media Access Control – Token Passing
Project 802’s specifications include three different tokenbased protocols – 802.4 Token Bus, 802.5 Token Ring, and
Fiber Distributed Data Interface (FDDI); generally pronounced
as a word; fiddy.

A token is a special frame that is passed from device to
device, in sequence along the ring.


It can circulate only when the ring is idle
A device must have this token to place data frames on the
network.

CCM4300_09-10
66
MAC Protocols: a taxonomy
Access Method
Taking-turn
Protocols /
Round Robin
Random Access
Protocols / Contention
Slotted
Aloha
Aloha CSMA/CD
Channel
Partitioning /
Reservation
TDMA
FDMA
CDMA
Token Token Slotted
Ring
Ring Bus
Goal: efficient, fair, simple, decentralised
CCM4300_09-10
67
Desirable MAC characteristic
||| When only one node tx – throughput R bps
||| When M nodes tx – each node throughput
R/M bps (average tx rate over suitable
defined interval of time)
||| Decentralised protocol – no master to bring
system down
||| Simple protocol – inexpensive to implement
CCM4300_09-10
68
Random Access protocols
||| When node has packet to send
- transmit at full channel data rate R.
- no a priori coordination among nodes
||| two or more transmitting nodes -> “collision”,
||| random access MAC protocol specifies:
- how to detect collisions
- how to recover from collisions (e.g., via delayed
retransmissions)
||| Examples of random access MAC protocols:
- pure ALOHA
- slotted ALOHA
- CSMA and CSMA/CD
CCM4300_09-10
69
Pure (unslotted) ALOHA
||| Norman Abramson – surfing & packet switching
||| unslotted Aloha: simpler, no synchronization
||| pkt needs transmission:
- send without awaiting for beginning of slot
||| collision probability increases:
- pkt sent at t0 collide with other pkts sent in [t0-1, t0+1]
||| If collision occurs – transmit after different time intervals
CCM4300_Sep 08
70
Slotted Aloha
||| time is divided into equal size slots (= pkt trans. time)
||| node with new arriving pkt: transmit at beginning of next slot
||| if collision: retransmit pkt in future slots with probability p,
until successful.
||| decentralized – independently decides when to transmit but
needs synchronization
Success (S), Collision (C), Empty (E) slots
CCM4300_09-10
71
Recap: Pure/Slotted ALOHA
||| Pure ALOHA access protocol
- Data from remote terminals to the central
computer site share a common transmission
medium, a uhf radio channel.
- Any terminals with data to transmit simply sends
a packet. Hence there are occasional collisions.
||| Slotted ALOHA access protocol
- Similar to pure ALOHA except that packet
transmission occurs only in agreed time slot
- This doubles the maximum throughput compared
to pure ALOHA, because a collision occupies no
more than one time slot
CCM4300_09-10
72
Recap ….. Cont.
||| Nodes decision to transmit is made independent of
the activity of the other nodes attached to the
broadcast channel
||| Nodes neither pays attention to whether another
node begins transmitting nor stops transmission if
another node begins to interfere with transmission
CCM4300_09-10
73
Channel Partitioning MAC protocols
What is multiplexing?
||| A mux (acronym for multiplexer) enables data of multiple
transmission channels to share a common link.
||| In Its simplest form, multiplexing involves combining data
from several relatively low-speed input channels and
transmitting these across a single high-speed circuit.
Which layer of the OSI does multiplexing
occurs?
Layer 1 - Physical Layer
CCM4300_09-10
74
Network Core: Circuit Switching

Capacity of medium exceeds the capacity
required for transmission of a single signal

How can we improve “efficiency”?


Let’s multiplex!
Divide link bandwidth into “pieces”:
 frequency division - FDMA
 time division – TDMA
CCM4300_09-10
75
Channel Partitioning MAC protocols: TDM
TDM: time division multiplexing
||| access to channel in "rounds"
||| each station gets fixed length slot (length = pkt
trans time) in each round
||| unused slots go idle
||| example: 6-station LAN, 1,3,4 have pkt, slots
2,5,6 idle
CCM4300_09-10
76
Channel Partitioning MAC protocols: FDM
frequency bands
FDM: frequency division multiplexing
||| channel spectrum divided into frequency bands
||| each station assigned fixed frequency band
||| unused transmission time in frequency bands go idle
||| example: 6-station LAN, 1,3,4 have pkt, frequency
bands 2,5,6 idle
CCM4300_09-10
77
Example: FDMA and TDMA
Example:
FDMA
4 users
frequency
time
TDMA
frequency
time
CCM4300_09-10
78
Advantages / Disadvantages
|||
Advantages
- eliminates collision
- perfectly fair; each node gets dedicated
transmission
|||
Disadvantages
- node is limited to average rate even when it is
the only node with frames to send
- node must always wait for its turn in
transmission sequence even when it is the
only node with a frame to send – only for TDM
CCM4300_09-10
79
Channel Partitioning (CDMA)
CDMA (Code Division Multiple Access)
||| unique “code” assigned to each user; ie, code set
partitioning
||| used mostly in wireless broadcast channels (cellular,
satellite,etc)
||| all users share same frequency, but each user has own
“chipping” sequence (ie, code) to encode data
||| encoded signal = (original data) X (chipping sequence)
||| decoding: inner-product of encoded signal and chipping
sequence
||| allows multiple users to “coexist” and transmit
simultaneously with minimal interference (if codes are
CCM4300_09-10
“orthogonal”)
80