CHAPTER  ONE  
Data Communication
And
Computer Networks

Data Communication Basics  

Communica)on  Basics  

v  Data  communica)ons  are  the  exchange  of  data  between  two  devices  via  
some  form  of  transmission  medium  such  as  a  wire  cable.  

v  It  includes  :  
Ø  Wri:ng  and  talking  

Ø  Nonverbal  communica:on  (such  as  facial  expressions,  body  language  or  
gestures)  

Ø  Visual  communica:on  (the  use  of  images  or  pictures  such  as  pain:ng,  
photography,  video  or  film)  

Ø  Electronic  Communica:on  such  as  telephone  calls,  electronic  mail,  cable  
television,  satellite  broadcasts.  

Sunday,  August  8,  21   2  

Communica)on  Basics  

v  For   data   communica:ons   to   occur,   the   communica:ng   devices   must   be  

part   of   a   communica:on   system   made   up   of   a   combina:on   of   hardware  
(physical  equipment)  and  soMware  (programs).    

Sunday,  August  8,  21   3  

Five  components  of  data  communica)on  

1.  Message:  is  the  informa:on  (data)  to  be  communicated.  Popular    
forms  of  informa:on  include  text,  numbers,  pictures,  audio,  and  
video.    
2.  Sender:   is   the   device   that   sends   the   data   message.   It   can   be   a  
computer,  worksta:on,  telephone  handset,  video  camera,  and  so  
on.    
3.  Receiver:     is   the   device   that   receives   the   message.   It   can   be   a  
computer,  worksta:on,  telephone  handset,  television,  and  so  on.    
 

Sunday,  August  8,  21   4  

Five  components  of  data  communica)on  

3.  Transmission   medium:   is   the   physical   path   by   which   a  

message  travels  from  sender  to  receiver.  Some  examples  of  
transmission  media  include  twisted-­‐pair  wire,  coaxial  cable,  
fiberop:c  cable,  and  radio  waves.    
4.  Protocol:  is  a  set  of  rules  that  govern  data  communica:ons.  
It   represents   an   agreement   between   the   communica:ng  
devices.  Without  a  protocol,  two  devices  may  be  connected  
but   not   communica:ng,   just   as   a   person   speaking   French  
cannot   be   understood   by   a   person   who   speaks   only  
Japanese.    
Sunday,  August  8,  21   5  
Modes  of  data  transmission  

v There   are   3   different   transmission   modes   characterized  

according  to  the  direc:on  of  the  exchanges:  
1.  A  simplex  connec)on  is  a  connec:on  in  which  the  data  flows  
in  only  one  direc)on,  from  the  transmiYer  to  the  receiver.  
–  This   type   of   connec:on   is   useful   if   the   data   do   not   need   to  
flow   in   both   direc:ons   (for   example,   from   your   computer  
to  the  printer  or  from  the  mouse  to  your  computer...).  

Sunday,  August  8,  21   6  

Contd.  
 
2.  A   half-­‐duplex   connec)on   (some:mes   called   an   alterna'ng  
connec'on   or   semi-­‐duplex)   is   a   connec'on   in   which   the   data  
flows  in  one  direc:on  or  the  other,  but  not  both  at  the  same  
:me.    
–  With  this  type  of  connec:on,  each  end  of  the  connec:on  
transmits  in  turn.    
–  This   type   of   connec:on   makes   it   possible   to   have  
bidirec:onal   communica:ons   using   the   full   capacity   of   the  
line.  

Sunday,  August  8,  21   7  

Contd.  
 
–  In   a   half-­‐duplex   transmission,   the   en:re   capacity   of   a  
channel   is   taken   over   by   whichever   of   the   two   devices   is  
transmi\ng  at  the  :me.  
–  Walkie-­‐talkies  and  CB  (ci:zens  band)  radios  are  both  half-­‐
duplex  systems.  
–  The   half-­‐duplex   mode   is   used   in   cases   where   there   is   no  
need   for   communica:on   in   both   direc:ons   at   the   same  
:me;  the  en:re  capacity  of  the  channel  can  be  u:lized  for  
each  direc:on.  

Sunday,  August  8,  21   8  

Contd.  
 
3.  Full-­‐Duplex:   In   full-­‐duplex   mode   (also   called   duplex),   both  
sta:ons  can  transmit  and  receive  simultaneously  
–  The   full-­‐duplex   mode   is   like   a   two-­‐way   street   with   traffic  
flowing  in  both  direc:ons  at  the  same  :me.  
–  In  full-­‐duplex  mode,  signals  going  in  one  direc:on  share  the  
capacity   of   the   link:   with   signals   going   in   the   other  
direc:on.    
–  The  full-­‐duplex  mode  is  used  when  communica:on  in  both  
direc:ons  is  required  all  the  :me.  
–  The   capacity   of   the   channel,   however,   must   be   divided  
between  the  two  direc:ons.  

Sunday,  August  8,  21   9  

Contd.  
–  This   sharing   can   occur   in   two   ways:   Either   the   link   must  
contain   two   physically   separate   transmission   paths,   one   for  
sending  and  the  other  for  receiving;  or  the  capacity  of  the  
channel   is   divided   between   signals   travelling   in   both  
direc:ons.  
–  One  common  example  of  full-­‐duplex  communica:on  is  the  
telephone   network.   When   two   people   are   communica:ng  
by   a   telephone   line,   both   can   talk   and   listen   at   the   same  
:me.  

Sunday,  August  8,  21   10  

Contd.  

Sunday,  August  8,  21   11  

 Interconnec)on  of  Networks:  Internetwork  

§  Today,   it   is   very   rare   to   see   a   LAN,   a   MAN,   or   a   WAN  

in  isola:on;  they  are  connected  to  one  another.    
§  When   two   or   more   networks   are   connected,   they  
become  an  internetwork,  or  internet.  

Sunday,  August  8,  21   12    

 Network,  internet,  and  Internet  

§  A   network   is   a   group   of   connected   communica:ng   devices   such   as  

computers  and  printers.  
§  An   internet   (note   the   lowercase   leYer   i)   is   two   or   more   networks   that   can  
communicate  with  each  other.    
§  The   most   notable   internet   is   called   the   Internet   (uppercase   leYer   I),   a  
collabora:on   of   more   than   hundreds   of   thousands   of   interconnected  
networks.  
§  Private   individuals   as   well   as   various   organiza:ons   such   as   government  
agencies,   schools,   research   facili:es,   corpora:ons,   and   libraries   in   more  
than  100  countries  use  the  Internet.  

Sunday,  August  8,  21   13    

Sunday,  August  8,  21   14    
 Internet    Vs.    Intranet      Vs.    Extranet      

§  Internet:   It   is   a   global   collec'on   of   networks,   both   big   and  

small.  It  is  a  giant  network.  
§  Intranet:    
•  Are   similar   to   World   Wide   Web   (WWW)   where  
centrally  stored  hypertext  documents  can  be  accessed  
using  the  web.    
•  Unlike   the   WWW,   intranets   are   available   only   to   user  
within  the  company  network  system.  Intranet  is  a  very  
common  service  in  large  networked  organiza:ons.  

Sunday,  August  8,  21   15    

 Internet    Vs.    Intranet      Vs.    Extranet      

§  Extranet:    
•  Although  similar  to  intranet,  extranet  provides  selected  
users  from  outside  the  organiza:on  to  access  data  from  
the  internal  network.    
•  Extranets   are   commonly   used   by   suppliers   to   provide  
data  to  company  clients.    

Sunday,  August  8,  21   16    

 Computer  Network  and  its  Applica)ons  
! At   its   most   elementary   level,   a   computer   network   consists   of  
two  computers  connected  to  each  other  by  a  cable  that  allows  
them  to  share  data.    
! All   computer   networking,   no   maYer   how   sophis:cated,   stems  
from  that  simple  system.  
! Computer  networking  arose  as  an  answer  to  the  need  to  share  
data  in  a  :mely  fashion.    
! Personal   computers   are   powerful   tools   that   can   process   and  
manipulate  large  amounts  of  data  quickly,  but  they  do  not  allow  
users  to  share  that  data  efficiently.    
! Before  networks,  users  needed  either  to  print  out  documents  or  
copy  document  files  to  a  disk  for  others  to  edit  or  use  them.    
Sunday,  August  8,  21   17    
 Contd.  
! If  others  made  changes  to  the  document,  there  was  no  
easy  way  to  merge  the  changes.    
! This   was,   and   s:ll   is,   known   as   "working   in   a   stand-­‐alone  
environment."    

! Copying   files   onto   floppy   disks   and   giving   them   to  

others   to   copy   onto   their   computers   was   some:mes  
referred   to   as   the   "sneakernet."   This   early   form   of  
computer  networking  is  one  that  many  of  us  have  used  
and  perhaps  s:ll  use  today  

Sunday,  August  8,  21   18    

Contd.  

§  connec:ng   together   of   computers   and   other  

devices   is   called   a   network,   and   the   concept   of  

connected   computers   sharing   resources   is  

called  networking.  

Sunday,  August  8,  21   19    

 Contd.  

§  Two   computers   are   said   to   be   interconnected  

if  they  are  able  to  exchange  informa:on  

§  Components  
Hardware: of  a  compute  networks  
§ Computer Media:
§ Network card § Cable
§ Routers § Wire
§ Modem … § Microwave …

Software: Network Design:

§  Network OS n Logical layout
§  Utilities … n Physical layout …

Sunday,  August  8,  21   20    

 Uses  of  Computer  Networks  

Sunday,  August  8,  21   21  

 Contd.  
1.  Resource  sharing  
!  the  goal  of  network  is  to  make  all  programs,  equipment,  and  especially  
data  available  to  anyone  on  the  network  without  regard  to  the  physical  
loca:on  of  the  resource  and  the  user.  
!  An  obvious  and  widespread  example  is  having  a  group  of  office  workers  
share  a  common  printer.    
!  None  of  the  individuals  really  needs  a  private  printer,  and  a  high-­‐volume  
networked  printer  is  oMen  cheaper,  faster,  and  easier  to  maintain  than  a  
large  collec:on  of  individual  printers.  
!  Informa)on  sharing  is  more  important  than  physical  resource  sharing  

Sunday,  August  8,  21   22    

 Contd.  

2.  Means  of  communica)on    

–  E-­‐mail  
–  Videoconferencing  
–  Cha\ng  
–  E-­‐commerce  
–  Game  
–  ….  
3.  Centralizing  administra)on  and  support  
–  Database  
–  Banks  
–  ….  
Sunday,  August  8,  21   23    
 Classifica)on  of  Computer  Networks  
§  Computer networks are classified based on various factors. They include:
–  Administration
–  From an administrator’s point of view, a network can be
private network which belongs a single autonomous system
and cannot be accessed outside its physical or logical domain.
A network can be public, which is accessed by all.
–  Geographical span
–  Connection Links
–  Inter-connectivity (Topology)
–  Architecture

Sunday,  August  8,  21   24    

 Network  Types    -­‐    Geographical  Span    
"  Geographically a network can be seen in one of the following categories:
"  It may be spanned across your table, among Bluetooth enabled
devices, Ranging not more than few meters.
"  It may be spanned across a whole building, including intermediate
devices to connect all floors.
"  It may be spanned across a whole city.
"  It may be spanned across multiple cities or provinces.
"  It may be one network covering whole world.
§  Generally there are three broad categories of networks based on its size and
geographical coverage
1.  Local Area Network (LAN)
2.  Metropolitan Area Network (MAN)
3.  Wide Area Network (WAN)

Sunday,  August  8,  21   25    

 LAN  
"  A   local   area   network   (LAN)   is   the   basic   building   block   of   any  
computer  network.    
"  A   LAN   can   range   from   simple   (two   computers   connected   by   a  
cable)   to   complex   (hundreds   of   connected   computers   and  
peripherals  throughout  a  major  corpora:on).    
"  The  dis:nguishing  feature  of  a  LAN  is  that  it  is  confined  to  a  limited  
geographic  area.  
"  A  local  area  network  (LAN)  is  usually  privately  owned  and  links  the  
devices  in  a  single  office,  building,  or  campus.    
"  Depending  on  the  needs  of  an  organiza:on  and  the  type  of  
technology  used,  a  LAN  can  be  as  simple  as  two  PCs  and  a  printer  
in  someone's  home  office;  or  it  can  extend  throughout  a  company  
and  include  audio  and  video  peripherals.    
"  Currently,  LAN  size  is  limited  to  a  few  kilometers  

Sunday,  August  8,  21   26    

Contd.  

Sunday,  August  8,  21   27    

 Contd.  
#  LANs   are   designed   to   allow   resources   to   be   shared   between   personal  
computers   or   worksta:ons.   The   resources   to   be   shared   can   include  
hardware   (e.g.,   a   printer),   soMware   (e.g.,   an   applica:on   program),   or  
data.  
#  In  addi:on  to  size,  LANs  are  dis:nguished  from  other  types  of  networks  
by  their  transmission  media  and  topology.    
#  In  general,  a  given  LAN  will  use  only  one  type  of  transmission  medium.  
The  most  common  LAN  topologies  are  bus,  ring,  and  star.  
#  Early   LANs   had   data   rates   in   the   4   to   16   megabits   per   second   (Mbps)  
range.  Today,  however,  speeds  are  normally  100  or  1000  Mbps.    
#  Wireless  LANs  are  the  newest  evolu:on  in  LAN  technology.  

Sunday,  August  8,  21   28    

 WAN  
§  A   wide   area   network   (WAN)   provides   long-­‐distance  
transmission  of  data,  image,  audio,  and  video  informa:on  
over  large  geographic  areas  that  may  comprise  a  country,  
a  con:nent,  or  even  the  whole  world.  
§  A  WAN  can  be  as  complex  as  the  backbones  that  connect  
the  Internet  or  as  simple  as  a  dial-­‐up  line  that  connects  a  
home  computer  to  the  Internet.  We  normally  refer  to  the  
first  as  a  switched  WAN  and  to  the  second  as  a  point-­‐to-­‐
point  WAN  

Sunday,  August  8,  21   29    

 Contd.  
 
§  The   switched   WAN   connects   the   end   systems,   which   usually  

comprise   a   router   (internetworking   connec:ng   device)   that  

connects  to  another  LAN  or  WAN.    

§  The   point-­‐to-­‐point   WAN   is   normally   a   line   leased   from   a   telephone  

or  cable  TV  provider  that  connects  a  home  computer  or  a  small  LAN  

to  an  Internet  service  provider  (lSP).  This  type  of  WAN  is  oMen  used  

to  provide  Internet  access.  

Sunday,  August  8,  21   30    

Sunday,  August  8,  21   31    
 MAN  

§  A   metropolitan   area   network   (MAN)   is   a   network   with   a   size   between   a   LAN   and   a  
WAN.  
§  It  normally  covers  the  area  inside  a  town  or  a  city.  
§  It  is  designed  for  customers  who  need  a  high-­‐speed  connec)vity,  normally  to  the  
Internet,  and  have  endpoints  spread  over  a  city  or  part  of  city.    
§  A   good   example   of   a   MAN   is   the   part   of   the   telephone   company   network   that   can  
provide   a   high-­‐speed   DSL   line   to   the   customer   (we   will   discuss   DSL   in   later  
chapters)  .    
§  Another  example  is  the  cable  TV  network  that  originally  was  designed  for  cable  TV,  
but  today  can  also  be  used  for  high-­‐speed  data  connec:on  to  the  Internet.  

Sunday,  August  8,  21   32  

 Summary  of  Networks  Types

§  As   a   general   we   can   dis:nguish   the   difference   of  

network  type,  based  on  the  following  key  points  
•  The  geographical  span  they  cover  

•  Type  of  media  they  used  

•  Data  transmission  rate  

•  Type  of  service  they  provide  

•  Cost  

Sunday,  August  8,  21   33  

 Network  Type  Based  on  Connec)on  Links  

§  A  network  is  two  or  more  devices  connected  through  links.    
§  A   link   is   a   communica:ons   pathway   that   transfers   data   from   one  
device  to  another.    
§  For  visualiza:on  purposes,  it  is  simplest  to  imagine  any  link  as  a  line  
drawn  between  two  points.    
§  For   communica:on   to   occur,   two   devices   must   be   connected   in  
some  way  to  the  same  link  at  the  same  :me.  
§  There   are   two   possible   types   of   connec:ons:   point-­‐to-­‐point   and  
mul)point.  

Sunday,  August  8,  21   34  

 1.  Point-­‐to-­‐Point  

§  A   point-­‐to-­‐point   connec:on   provides   a   dedicated   link   between   two  

devices.    
§  The  en:re  capacity  of  the  link  is  reserved  for  transmission  between  those  
two  devices.    
§  Most  point-­‐to-­‐point  connec:ons  use  an  actual  length  of  wire  or  cable  to  
connect  the  two  ends,  but  other  op:ons,  such  as  microwave  or  satellite  
links,  are  also  possible.  
§  When  you  change  television  channels  by  infrared  remote  control,  you  are  
establishing  a  point-­‐to-­‐point  connec:on  between  the  remote  control  and  
the  television's  control  system.  

Sunday,  August  8,  21   35    

 2.  Mul)point  
§  A   mul:point   (also   called   mul:drop)   connec:on   is   one   in   which  
more  than  two  specific  devices  share  a  single  link.  
§  In  a  mul:point  environment,  the  capacity  of  the  channel  is  shared,  
either  spa)ally  or  temporally.    
§  If   several   devices   can   use   the   link   simultaneously,   it   is   a   spa'ally  
shared  connec:on.    
§  If  users  must  take  turns,  it  is  a  'meshared  connec'on.  

Sunday,  August  8,  21   36    

 Network  Topology  (Inter  -­‐  Connec)vity)    

§  Components   of   a   network   can   be   connected   to   each   other  

differently  in  some  fashion.    
–  By   connectedness   we   mean   either   logically,   physically,   or  
both  ways.    
–  Every  single  device  can  be  connected  to  every  other  device  
on  network,  making  the  network  mesh.    
–  All     devices     can     be     connected     to     a     single     medium     but    
geographically  disconnected,  created  bus-­‐like  structure.    

Sunday,  August  8,  21   37    

 Network  Topology  (Inter  -­‐  Connec)vity)    

–  Each    device  is    connected    to    its    leD    and    right    peers    
only,    crea'ng    linear  structure  
–  All    devices    connected    together    with    a    single    device,    
crea'ng    star-­‐like  structure.    
–  All  devices  connected  arbitrarily  using  all  previous  ways  to  
connect  each  other,  resul'ng  in  a  hybrid  structure.  

Sunday,  August  8,  21   38    

 Cont  .  .  .  

§  A Network Topology is the arrangement with which

computer systems or network devices are connected to each
other.
§  The term topology refers to the way in which a network is laid
out physically. (instead Physical  layout,  Design,  Diagram,  Map  
are  also  used)  
§  Topologies may define both physical and logical aspect of the
network.

Sunday,  August  8,  21   39    

 Cont  .  .  .  

§  Both logical and physical topologies could be same or

different in a same network.
§  The topology of a network is the geometric representation of
the relationship of all the links and linking devices (usually
called nodes) to one another.
§  There are four basic topologies possible: mesh, star, bus, and
ring

Sunday,  August  8,  21   40    

 Cont  .  .  .  

§  The  type  of  topology  you  use  affects  the  type  and  capabili:es.  
In  general  the  choice  of  one  topology  over  another  will  have  
an  impact  on  the:  
•  Type  of  equipment  that  the  network  needs  
•  Capabili:es  of  the  network  
•  Growth  of  the  network  
•  Way  the  network  is  managed  

Sunday,  August  8,  21   41    

 Cont  .  .  .  

§  Developing  a  sense  of  how  to  use  the  different  topologies  is  a  
key  to  understanding  the  capabili:es  of  the  different  types  of  
networks.  
§  Before   computers   can   share   resources   or   perform   other  
communica:on  tasks  they  must  be  connected.  Most  networks  
use   cable   to   connect   one   computer   to   another.   However,   it   is  
not   as   simple   as   just   plugging   a   computer   into   a   cable  
connec:ng  to  other  computers

Sunday,  August  8,  21   42    

 A.  Bus  Topology  

§  A  bus  topology,  is  mul)point  connec)on.  

§  One  long  cable  acts  as  a  backbone  to  link  all  the  devices  in  a  
network  

§  Nodes  are  connected  to  the  bus  cable  by  drop  lines  and  taps.    

§  A   drop   line   is   a   connec:on   running   between   the   device   and  

the  main  cable.  

Sunday,  August  8,  21   43    

 Contd.  
§  A  tap  is  a  connector  that  either  splices  into  the  main  cable  or  
punctures   the   sheathing   of   a   cable   to   create   a   contact   with  
the  metallic  core.    
§  As  a  signal  travels  along  the  backbone,  some  of  its  energy  is  
transformed  into  heat.    
§  Therefore,  it  becomes  weaker  and  weaker  as  it  travels  farther  
and  farther.  
§  For   this   reason   there   is   a   limit   on   the   number   of   taps   a   bus  
can  support  and  on  the  distance  between  those  taps.  
Sunday,  August  8,  21   44    
 Advantages  of  Bus  Topology.  
§  Advantages  of  a  bus  topology  include  ease  of  installa:on.    
§  Backbone  cable  can  be  laid  along  the  most  efficient  path,  then  connected  
to  the  nodes  by  drop  lines  of  various  lengths.    
§  In  this  way,  a  bus  uses  less  cabling  than  mesh  or  star  topologies.  
§  In   a   star,   for   example,   four   network   devices   in   the   same   room   require   four  
lengths  of  cable  reaching  all  the  way  to  the  hub.    
§  In  a  bus,  this  redundancy  is  eliminated.  Only  the  backbone  cable  stretches  
through  the  en're  facility.    
§  Each   drop   line   has   to   reach   only   as   far   as   the   nearest   point   on   the  
backbone.  

Sunday,  August  8,  21   45    

 Disadvantages  of  Bus  Topology.  
 
1.  Difficult   reconnec:on   and   fault   isola:on.   A   bus   is   usually  
designed  to  be  op:mally  efficient  at  installa:on.  
•   It  can  therefore  be  difficult  to  add  new  devices.    
2.  Signal  reflec:on(bouns)  at  the  taps  can  cause  degrada:on  in  
quality.   This   degrada:on   can   be   controlled   by   limi:ng   the  
number  and  spacing  of  devices  connected  to  a  given  length  of  
cable.    
•  Adding   new   devices   may   therefore   require  
modifica:on  or  replacement  of  the  backbone.  
Sunday,  August  8,  21   46  
 Disadvantages  of  Bus  Topology.  
 
§  In   addi:on,   a   fault   or   break   in   the   bus   cable   stops   all  
transmission,  even  between  devices  on  the  same  side  of  the  
problem.    
§  The   damaged   area   reflects   signals   back   in   the   direc:on   of  
origin,  crea:ng  noise  in  both  direc:ons.  
§  Bus   topology   was   one   of   the   first   topologies   used   in   the  
design  of  early  local  area  networks.  Ethernet  LANs  can  use  a  
bus  topology,  but  they  are  less  popular  now.  

Sunday,  August  8,  21   47  

 B.    Star  Topology  

§  In   a   star   topology,   each   device   has   a   dedicated   point-­‐to-­‐point   link   only   to  
a  central  controller,  usually  called  a  hub/switch.  
§  The   term   dedicated   means   that   the   link   carries   traffic   only   between   the  
two  devices  it  connects.  
§  The  devices  are  not  directly  linked  to  one  another.    
§  Unlike   a   mesh   topology,   a   star   topology   does   not   allow   direct   traffic  
between  devices.    
§  The   controller   acts   as   an   exchange:   If   one   device   wants   to   send   data   to  
another,  it  sends  the  data  to  the  controller,  which  then  relays  the  data  to  
the  other  connected  device  

Sunday,  August  8,  21   48    

 Contd.  
 
§  A  star  topology  is  less  expensive  than  a  mesh  topology.    
§  In  a  star,  each  device  needs  only  one  link  and  one  I/O  port  to  
connect  it  to  any  number  of  others.  
§  This  factor  also  makes  it  easy  to  install  and  reconfigure.    
§  Far   less   cabling   needs   to   be   housed,   and   addi:ons,   moves,  
and   dele:ons   involve   only   one   connec)on:   between   that  
device  and  the  hub.  
§  Other   advantages   include   robustness.   If   one   link   fails,   only  
that   link   is   affected.   All   other   links   remain   ac:ve.   This   factor  
also   lends   itself   to   easy   fault   iden)fica)on   and   fault  
isola)on.   As   long   as   the   hub   is   working,   it   can   be   used   to  
monitor  link  problems  and  bypass  defec:ve  links.  
Sunday,  August  8,  21   49    
 Contd.  

A  SIMPLE  STAR  TOPOLOGY    

A  HIERARCHICAL  STAR  TOPOLOGY  
Sunday,  August  8,  21   50    
 Contd.  
§  One   big   disadvantage   of   a   star   topology   is   the   dependency   of   the  
whole  topology  on  one  single  point,  the  hub.  
§  If  the  hub  goes  down,  the  whole  system  is  dead.  
§  Although  a  star  requires  far  less  cable  than  a  mesh,  each  node  must  
be  linked  to  a  central  hub.    
§  For   this   reason,   oMen   more   cabling   is   required   in   a   star   than   in  
some  other  topologies  (such  as  ring  or  bus).  
§  The  star  topology  is  used  in  local-­‐area  networks  (LANs).  
§  High-­‐speed  LANs  oMen  use  a  star  topology  with  a  central  hub.  

Sunday,  August  8,  21   51    

 C.  Ring  Topology  

§  In   a   ring   topology,   each   device   has   a   dedicated   point-­‐to-­‐point  

connec:on  with  only  the  two  devices  on  either  side  of  it.  
§  A  signal  is  passed  along  the  ring  in  one  direc:on,  from  device  
to  device,  un:l  it  reaches  its  des:na:on.  
§  Each  device  in  the  ring  incorporates  a  repeater.    
§  When  a  device  receives  a  signal  intended  for  another  device,  
its  repeater  regenerates  the  bits  and  passes  them  along  

Sunday,  August  8,  21   52    

 Contd.  
 
§  A  ring  is  rela:vely  easy  to  install  and  reconfigure.    
§  Each  device  is  linked  to  only  its  immediate  neighbors  (either  
physically  or  logically).  
§  To  add  or  delete  a  device  requires  changing  only  two  
connec:ons.    

Sunday,  August  8,  21   53    

 Contd.  
 
§  The   only   constraints   are   media   and   traffic   considera:ons  
(maximum  ring  length  and  number  of  devices).    
§  In  addi:on,  fault  isola:on  is  simplified.  
§  Generally  in  a  ring,  a  signal  is  circula:ng  at  all  :mes.    
§  In  a  simple  ring,  a  break  in  the  ring  (such  as  a  disabled  sta:on)  
can  disable  the  en:re  network.    
§  This  weakness  can  be  solved  by  using  a  dual  ring  or  a  switch  
capable  of  closing  off  the  break.  
§  Ring   topology   was   prevalent   when   IBM   introduced   its   local-­‐
area  network  Token  Ring.    
§  Today,  the  need  for  higher-­‐speed  LANs  has  made  this  topology  less  popular.  

Sunday,  August  8,  21   54    

 D.  Mesh  topology  

§  In   a   mesh   topology,   every   device   has   a   dedicated   point-­‐to-­‐

point  link  to  every  other  device.    
§  To  find  the  number  of  physical  links  in  a  fully  connected  mesh  
network  with  n  nodes,  we  first  consider  that  each  node  must  
be  connected  to  every  other  node.  
§    Node   1   must   be   connected   to   n   -­‐   I   nodes,   node   2   must   be  
connected   to   n   –   1   nodes,   and   finally   node   n   must   be  
connected  to  n  -­‐  1  nodes.  We  need  n(n  -­‐  1)  physical  links.  
 

Sunday,  August  8,  21   55    

 D.  Mesh  topology  

§  However,   if   each   physical   link   allows   communica:on   in   both  

direc:ons   (duplex   mode),   we   can   divide   the   number   of   links  
by   2.     In   other   words,   we   can   say   that   in   a   mesh   topology,   we  
need  n(n  -­‐1)  /2  duplex-­‐mode  links.  
§  To   accommodate   that   many   links,   every   device   on   the  
network   must   have   n   –   1   input/output   (I/O)   ports   to   be  
connected  to  the  other  n  -­‐  1  sta'ons.  

Sunday,  August  8,  21   56    

 Contd.  
§  A  mesh  offers  several  advantages  over  other  network  topologies  
§  First,  the  use  of  dedicated  links  guarantees  that  each  connec:on  can  carry  
its   own   data   load,   thus   elimina:ng   the   traffic   problems   that   can   occur  
when  links  must  be  shared  by  mul:ple  devices.  
§  Second,  a  mesh  topology  is  robust.  If  one  link  becomes  unusable,  it  does  
not  incapacitate  the  en:re  system.    
§  Third,  there  is  the  advantage  of  privacy  or  security.  When  every  message  
travels   along   a   dedicated   line,   only   the   intended   recipient   sees   it.   Physical  
boundaries  prevent  other  users  from  gaining  access  to  messages.    
§  Finally,   point-­‐to-­‐point   links   make   fault   iden:fica:on   and   fault   isola:on  
easy.    
§  Traffic  can  be  routed  to  avoid  links  with  suspected  problems.  This  facility  
enables  the  network  manager  to  discover  the  precise  loca:on  of  the  fault  
and  aids  in  finding  its  cause  and  solu:on.  
Sunday,  August  8,  21   57    
 Contd.  
§  The   main   disadvantages   of   a   mesh   are   related   to   the   amount   of  
cabling  and  the  number  of  I/O  ports  required.    
§  First,   because   every   device   must   be   connected   to   every   other  
device,  installa:on  and  reconnec:on  are  difficult.    
§  Second,   the   sheer   bulk   of   the   wiring   can   be   greater   than   the  
available  space  (in  walls,  ceilings,  or  floors)  can  accommodate.  
§  Finally,  the  hardware  required  to  connect  each  link  (I/O  ports  and  
cable)  can  be  prohibi:vely  expensive.  
§  For   these   reasons   a   mesh   topology   is   usually   implemented   in   a  
limited   fashion,   for   example,   as   a   backbone   connec:ng   the   main  
computers   of   a   hybrid   network   that   can   include   several   other  
topologies.  

Sunday,  August  8,  21   58    

Contd.  
§  One   prac:cal   example   of   a   mesh   topology   is   the   connec:on   of  
telephone   regional   offices   in   which   each   regional   office   needs   to  
be  connected  to  every  other  regional  office  

Mesh  topology  
 
Sunday,  August  8,  21   59  
 E.  Hybrid  Topology  
§  A   network   can   be   hybrid.   For   example,   we   can   have   a  
main  star  topology  with  each  branch  connec:ng  several  
sta:ons  in  a  bus  topology  (star-­‐bus  topology)  

Sunday,  August  8,  21   60  

 Network  Models  (Architecture)  by  Capability  

•  Computer  networks  can  be  discriminated  into  various  types  such  as  Client-­‐
Server,  peer-­‐to-­‐peer  or  hybrid,  depending  upon  its  architecture.    
ü There    can    be    one  or    more    systems    ac:ng    as    Server.    Other    being    
Client,  requests  the  Server  to  serve  requests.  Server  takes  and  
processes  request  on  behalf  of  Clients.    
ü Two  systems  can  be  connected  Point-­‐to-­‐Point,  or  in  back-­‐to-­‐back  
fashion.  They  both  reside  at  the  same  level  and  called  peers.    
•  There  can  be  hybrid  network  which  involves  network  architecture  of  both  
the  above  types.  
1.  Client/Server Model
2.  Peer-to- Peer Model

Sunday,  August  8,  21   61    

 Network  Models  by  Capability  
1.  Client/Server Model
§  Microcomputer users, or clients, share services of a centralized
computer called a server.

Sunday,  August  8,  21   62    

 Client-­‐server  model  

Fig.  Two  clients  using  a  single  server  at  a  

:me  
Sunday,  August  8,  21   63    
 Client/Server    

Ø  A   dis:nc:on   exists   between   computers   that   make   available   network  

resources  (servers)  and  those  computers  that  use  the  resources  (clients  or  
worksta:ons).  
Pros:  
§  Very  secure  
§  Centralized  servers  easy  to  manage  
§  Physically  centralized  
§  Secure  OS  
§  BeYer  performance  
§  Centralized  backups  
§  Reliability  
§  Simple  job  to  do  plus  built  in  redundancy  
Cons:  
§  Require  professional  administra:on  
§  More  hardware  intensive  
Sunday,  August  8,  21   64    
Network  Models  by  Capability  
2.  Peer-to-Peer Model

§  Computers share equally with one another without having to rely on a

central server.

Sunday,  August  8,  21   65    

 Peer-­‐to-­‐Peer  (p2p)  

§  Computers  on  the  network  communicate  with  each  others  as  
equals  and  each  computer  is  responsible  for  making  its  own  
resources  available  to  other  computers  on  the  network.  
Pros:
§  Uses less expensive computer networks
§  Easy to administer
§  No NOS (network operating system) required
§  More built-in redundancy
§  Shared resources – some machine will have what you need

Sunday,  August  8,  21   66    

 Peer-­‐to-­‐Peer  (p2p)  

Cons:
§  Individual user performance easily affected
§  Not very secure
§  Tragedy of the commons – no guarantee others will
administer their resources properly (almost guaranteed with
over 10 machines)
§  Hard to back up.

Sunday,  August  8,  21   67    

 Client/Server  vs.  Peer-­‐to-­‐Peer:    Advantages  and  
Disadvantages    
Client/Server Model Peer-to-Peer Model
Advantages:   Advantages:  
§  Very  secure  OS.   §  Uses  less  expensive  networks.  
§  BeYer  performance.   §  Easy  to  administer.  
§  Centralized  servers,  easy  to   §  Contain   both   network   opera:ng  
manage.   system  and  applica:on  soMware.  
§  Centralized  backups.   §  Ideal   for   small   business   and   home  
§  High  reliability.   users  (up  to  10  computers).  
Disadvantages:   Disadvantages:  
§  Expensive  administra:on.   §  Individual  user  performance  easily  
§  More  hardware  intensive.   affected.  
§  Not  very  secure.  
§  Hard  to  back  up.  

Sunday,  August  8,  21   68    

 Summary  of  Networks  Models

§  We  can  dis:nguish  network  models/architecture  difference  

based  on  the  following  key  points.
» Their  func:ons
» Security  level
» Scalability
» Type  of  opera:ng  system  they  used  
» Backup  process

Sunday,  August  8,  21   69  

 Network  Criteria    
§  A  network  must  be  able  to  meet  a  certain  number  
of  criteria.    
§  The  most  important  of  these  are  :-­‐  
–  Performance,  
–  Reliability  
–  Security.  
 

Sunday,  August  8,  21   70  

 Network  Criteria    
§  Performance    
–  Can  be  measured  in  many  ways,  including  transmit  time  
and  response  time.    
–  ︎    Transmit  time  is  the  amount  of  time  required  for  a  
message  to  travel  from  one  device  to  another.  
–   Response  time  is  the  elapsed  time  between  an  inquiry  
and  a  response.    
–  ︎Depends  on  a  number  of  factors:    
»   the  number  of  users  
»  the  type  of  transmission  medium  
»  the  capabilities  of  the  connected  hardware  
»   the  ef7iciency  of  the  software.    
Sunday,  August  8,  21   71  
 Network  Criteria    
§  Reliability    
–  In  addition  to  accuracy  of  delivery,  network  reliability  is  
measured  by  the  frequency  of  failure,  the  time  it  takes  
a  link  to  recover  from  a  failure.    
§  Security    
–  Network  security  issues  include  :  
•  protecting  data  from  unauthorized  access  
•  protecting  data  from  damage  and  development  
•  implementing  policies  and  procedures  for  recovery  
from  breaches  and  data  losses.    

Sunday,  August  8,  21   72  

Standards

# Most  networks  are  a  combina:on  of  hardware  and  soMware  from  a  

variety  of  vendors.    
# The   ability   to   combine   the   products   manufactured   by   different  
vendors  is  made  possible  by  the  existence  of  industry  standards.  
# Standards   are   guidelines   that   vendors   adhere   /s)ck   on   to  
voluntarily   in   order   to   make   their   products   compa:ble   with  
products  from  other  vendors.    
# In  general  those  standards  address:  
Ø Size.    
Ø Shape.    
Ø Material.    
Ø Func:on.    
Ø Speed.    Etc.  

Sunday,  August  8,  21   73  

Contd.

#  More   specifically,   the   standards   define   physical   and   opera:onal  

characteris:cs  of:  
Ø Opera:ng  systems  (NOS)    
Ø   Communica:on  SoMware.    
#  For   example,   standards   make   it   possible   to   buy   a   NIC   manufactured   by  
one   vendor   for   a   computer   manufactured   by   another   vendor   with  
reasonable  assurance  that  the  card  will:  
Ø  Fit  into  the  computer.    
Ø  Work  with  the  network  cabling.    
Ø  Translate  signals  from  computer  and  send  them  out  onto  network.    
Ø  Receive  data  from  the  network  and  deliver  it  to  the  computer.    
#  When   a   vendor   subscribes   to   a   set   of   standards,   it   means   that   the   vendor  
is  agreeing  to  make  equipment  that  conforms  to  the  specifica:ons  of  the  
standard.  

Sunday,  August  8,  21   74  

Layered  communica)on  model    

# Layers   model   is   a   theore:cal   approach   that   describe   how   data   is  

going  to  be  transmit  over  the  network.    
# There   are   several   layer   models   available,   among   those   following  
two  models  are  the  most  popular:      
# OSI  Reference  model  
# TCP  /  IP  Reference  model  
# A   communica:on   architecture   is   a   strategy   for   connec:ng   host  
computers   and   other   communica:ng   equipment   by   define   all  
necessary  elements  for  data  communica:on  between  devices.  

Sunday,  August  8,  21   75  

Layered Tasks

# We  use  the  concept  of  layers  in  our  daily  life.    
#  As  an  example,  let  us  consider  two  friends  who  communicate  through  postal  
mail.  

#    The   process   of   sending   a   leYer   to   a   friend   would   be   complex   if  

there  were  no  services  available  from  the  post  office.  

Sunday,  August  8,  21   76  

Contd.

#  Layer  architecture  simplifies  the  network  design.  

#  It  is  easy  to  debug  network  applica:ons  in  a  layered  architecture  network.  
#  The  network  management  is  easier  due  to  the  layered  architecture.  
#  Layering   communica:on   func:ons   adds   flexibility,   for   example,   we   do   not  
need  to  modify  the  en:re  host  soMware  to  include  more  communica:on  
devices.  
#  Network  layers  follow  a  set  of  rules,  called  protocol.  
#  The   protocol   defines   the   format   of   the   data   being   exchanged,   and   the  
control  and  :ming  for  the  handshake  between  layers.  

Sunday,  August  8,  21   77  

Open Systems Interconnection (OSI)

# Interna:onal   standard   organiza:on   (ISO)   established   a  

commiYee   in   1977   to   develop   an   architecture   for   computer  
communica:on.  
# Open   Systems   Interconnec:on   (OSI)   reference   model   is   the  
result  of  this  effort.  
# In   1984,   the   Open   Systems   Interconnec:on   (OSI)   reference  
model   was   approved   as   an   interna:onal   standard   for  
communica:ons  architecture.  
# Term    “open”    denotes  the  ability  to  connect  any  two  systems  
which   conform   to   the   reference   model   and   associated  
standards.  
# The   OSI   model   is   now   considered   the   primary   Architectural    
model  for  inter-­‐computer  communica:ons.  
Sunday,  August  8,  21   78  
Contd.

§  It  is  a  set  of  protocols  that  allows  any  two  different  
systems  to  communicate  regardless  of  their  
underlying  architecture.  
§  The  Purpose:  
–   to  show  how  to  facilitate  communica:on  between  
different  systems  without  requiring  changes  to  the  
logic  of  the  underlying  hardware  and  soMware    
§  It  is  not  a  protocol;    
–  it  is  a  model  for  understanding  and  designing  a  
network  architecture  that  is  flexible,  robust,  and  
interoperable.    
Sunday,  August  8,  21   79  
Contd.

# The   OSI   model   describes   how   informa:on   or   data   makes   its  

way  from  an  applica:on  programmes  (such  as  spreadsheets)  
in   computer,   through   a   network   medium   (such   as   wire)   to  
another  applica:on  programme  located  in  another  computer  
or  on  another  network.  

# The  model  also  helps  to  troubleshoot  problems  by  providing  a  

frame   of   reference   that   describes   how   components   are  
supposed  to  func:on  
Sunday,  August  8,  21   80  
Contd.

# The   OSI   reference   model   divides   the   problem   of   moving  

informa:on   between   computers   over   a   network   medium   into  
SEVEN  smaller  and  more  manageable  problems.  
# This   separa:on   into   smaller   more   manageable   func:ons   is  
known  as  layering.  
# The   main   objec:ve   of   dividing   into   seven   layers   include  
crea:ng  a  manageable  layers  by  breaking  down  the  complex  
network   opera:ons   and   allowing   specializa:on   of   research  
and  development  on  each  component  of  the  OSI  layer  
Sunday,  August  8,  21   81  
Contd.

#  In   developing   the   model,   the   designers   dis'lled   the   process   of   transmiPng  

data  to  its  most  fundamental  elements.    

#  They   iden:fied   which   networking   func:ons   had   related   uses   and   collected  
those  func:ons  into  discrete  groups  that  became  the  layers.    

#  Each  layer  defines  a  family  of  func:ons  dis:nct  from  those  of  the  other  layers.    

#  By   defining   and   localizing   func:onality   in   this   fashion,   the   designers   created  

an  architecture  that  is  both  comprehensive  and  flexible.    

#  Most   importantly,   the   OSI   model   allows   complete   interoperability   between  

otherwise  incompa:ble  systems.  

Sunday,  August  8,  21   82  

Contd.

Purpose  of  OSI  model  

# OSI  model  presents  a  structured  set  of  seven  layers  interconnec:ng  
as  a  stack.    
# Each   layer   in   this   model   was   developed   to   maintain   the   ability   of  
connec:ng  the  devices,  computers  and  network  segments.    
# OSI  model  was  created  with  following  purposes:-­‐    
# To   create   a   common   plaRorm   for   soDware   developers   and   hardware  
manufactures  that  encourage  the  crea'on  of  networking  products  that  
can  communicate  with  each  other  over  the  network.  
# To   help   network   administrators   by   dividing   large   data   exchange  
process   in   smaller   segments.   Smaller   segments   are   easier   to  
understand,  manage  and  troubleshoot.  With  layer  approach  they  only  
have  to  troubleshoot  the  devices  which  are  working  in  faulty  layer.  

Sunday,  August  8,  21   83  

Contd.

#  In general, the Benefits of the OSI Reference model is given below

# Big picture of communication over network is understandable through this OSI
model.
# OSI model distinguishes well between the services, interfaces and protocols.
# Protocols of OSI model are very well hidden.
# We see how hardware and software work together.
# Reduces complexity:- Dividing network communication process in smaller
component makes software development, design, and troubleshooting easier.
# Standardizes interfaces:- It describes what functions occur at each layer of the
model that encourages industry standardization.
# Facilitates modular/modification engineering:- It prevents changes in one layer
from affecting other layers.
# Ensures interoperable technology:- It allows different types of network
hardware and software to communicate. And can be used to compare basic
functional relationships on different networks

Sunday,  August  8,  21   84  

Contd.

#  In general, the Benefits of the OSI Reference model is given below

# Accelerates evolution:- Standardization of network components allows
multiple-vendor development.
# Quick troubleshooting: Enable network administrators to troubleshoot
issues more quickly and effectually by looking in layer that is causing
issue rather than finding it entire network.
# Simplifies teaching and learning:- We can understand new technologies
as they are developed.
§  Demerits of OSI reference model:
•  Model was devised before the invention of protocols.
•  Fitting of protocols is tedious task.
•  It is just used as a reference model.

Sunday,  August  8,  21   85  

 The Seven Layers
 

Sunday,  August  8,  21   86  

 OSI  Model    
•  enables  the  user(human  or  soMware)  to  access  the  
Applica)on  Layer   network.  

•  concerned  with  the  syntax  and  seman:cs  of  the  

Presenta)on  Layer   informa:on  exchanged  between  two  systems.  

•  The  session  layer  is  responsible  for  dialog  control  and  

Session  Layer   synchroniza:on.    

•  responsible  for  process-­‐to-­‐process  delivery  of  the  

Transport  Layer     en:re  message.    

•  responsible  for  the  source-­‐to-­‐des:na:on  delivery  of  a  

Network  Layer     packet,  possibly  across  mul:ple  networks  (links).    

•  the  delivery  of  the  packet  between  two  systems  on  the  
Data  Link  Layer     same  network  (links)  
•  It  deals  with  the  mechanical  and  electrical  
Physical  Layer     specifica:ons  of  the  interface  and  transmission  
medium.  

Sunday,  August  8,  21   87  

 OSI  Model    
•  it  provides  user  interfaces  and  support  for  services  such  
Applica)on  Layer   as  email,  Mp,  shared  database  management,  Telenet  

Presenta)on  Layer   •  Transla:on  ,Encryp:on  ,  Compression      

Session  Layer   •  Dialog  control  ,Synchroniza:on    

•  Service-­‐point  addressing  ,Segmenta:on  and  reassembly,  

Transport  Layer     Connec:on  control,  Flow  control  ,Error  control:    

Network  Layer     •  Logical  addressing  ,Rou:ng.    

•  Framing,  Physical  addressing,  Flow  control,  Error  

Data  Link  Layer     control  ,Access  control    
•  Physical  characteris:cs  of  interfaces  and  medium,  
Representa:on  of  bits  ,Data  rate  ,Synchroniza:on  of  
Physical  Layer     bits  ,Line  configura:on  ,Physical  topology,  Transmission  
mode    
Sunday,  August  8,  21   88  
 Contd.
#  The   figure   shows  
the   layers   involved  
when   a   message   is  
sent   from   device   A  
to  device  B.    
#  As   th e   mes s age  
travels   from   A   to   B,  
it   may   pass   through  
many   intermediate  
nodes.  
#    These   intermediate  
n o d e s   u s u a l l y  
involve  only  the  first  
three   boYom   layers  
of  the  OSI  model.  

Sunday,  August  8,  21   89  

Contd.

§  Layer   3,   for   example,   uses   the   services   provided   by   layer   2  

and  provides  services  for  layer  4.  Between  machines,  layer  x  

on   one   machine   communicates   with   layer   x   on   another  

machine.    

§  This  communica:on  is  governed  by  an  agreed-­‐upon  series  of  

rules  and  conven:ons  called  protocols.    

§  The  processes  on  each  machine  that  communicate  at  a  given  

layer  are  called  peer-­‐to-­‐peer  processes.    

Sunday,  August  8,  21   90  
Organization  of the Layers
#  The  seven  layers  can  be  thought  of  as  belonging  to  three  subgroups.    

#  Layers   1,   2,   and   3   -­‐   physical,   data   link,   and   network   -­‐   are   the   network   support  
layers;   they   deal   with   the   physical   aspects   of   moving   data   from   one   device   to  
another   (such   as   electrical   specifica:ons,   physical   connec:ons,   physical  
addressing,  and  transport  :ming  and  reliability).    

#  Layers  5,  6,  and  7-­‐  session,  presenta:on,  and  applica:on  -­‐  can  be  thought  of  as  the  
user   support   layers;   they   allow   interoperability   among   unrelated   soMware  
systems.  

#  Layer   4,   the   transport   layer,   links   the   two   subgroups   and   ensures   that   what   the  
lower  layers  have  transmiYed  is  in  a  form  that  the  upper  layers  can  use.  

#  The   upper   OSI   layers   are   almost   always   implemented   in   soMware;   lower   layers   are  
a   combina:on   of   hardware   and   soMware,   except   for   the   physical   layer,   which   is  
mostly  hardware.  
Sunday,  August  8,  21   91  
Contd.
# A  given  layer  within  the  OSI  model  communicates  with  three  other  layers:    
1.  the  layer    directly  above  it,    
2.  the  layer  directly  below  it  and    
3.  it’s  peer  layer  in  other  networked  computer  system.    

Sunday,  August  8,  21   92  

Data  Exchange  Process  

§  In  data  exchange  process,  par:cipa:ng  computers  work  in  

reverse  mode.  Layers  on  receiving  computer  perform  the  
same  task  in  reverse  mode.    
§  The  receiving  device  takes  delivery  of,  handles,  and  translates  
the  data  from  the  sending  device  at  a  par:cular  layer.    
§  For  example  on  sending  computer  presenta:on  layer  
compress  the  data,  same  presenta:on  layer  on  receiving  
computer  decompress  the  data.  

Sunday,  August  8,  21   Page  93  

Contd.
# On  sending  computer    
–  Sending application access the application layer.
–  Application provides data to the presentation layer.
–  Presentation layer format data as per network requirement and forward it's
to session layer.
–  Session layer initiate connection and forward the data to the transport layer.
–  Transport layer broke down the large data file in smaller segments and add a
header with control information, which are bits designated to describe how
to determine whether the data is complete, uncorrupted, in the correct
sequence, and so forth.
–  Segments are forwarded to network layer. Network layer add its header, with
logical address and convert it in packet and forwards packet to data link layer
–  Data link layer attach its header and footer to packet and convert it in frame.
–  Frames are forwarded to the physical layers that convert them in signals.
These signals are loaded in media.
Sunday,  August  8,  21   94  
Contd.
# On  receiving  computer    
–  Physical layer receive signals from media and convert them in frames. Frames
are forwarded to the data link layer.
–  Data link layer check the frame. All tampered frame are dropped here. If
frame is correct, data link layer strip down its header and footer from frame
and hand over packet to network layer.
–  Network layer check the packet with its own implementations. If it's found
everything fine with packet, it strips down its header from packet and hand
over segment to transport layer.
–  Transport layer again do the same job. It verifies the segments with its own
protocol rules. Only the verified segments are processed. Transport layer
remove its header from verified segments and reassemble the segments in
data. Data is handed over the session layer.
–  Session layer keep track of open connection and forwarded the receiving data
to presentation layer.
–  Presentation form the data in such a way that application layer use it.
–  Application layer on receiving computer find the appropriate application from
the computer and open data within particular application.
Sunday,  August  8,  21   95  
Contd.

#  The  figure  shown  in  the  next  slide,  gives  an  overall  view  of  the  OSI  layers,  D7  

means  the  data  unit  at  layer  7,  D6  means  the  data  unit  at  layer  6,  and  so  on.  

The  process  starts  at  layer  7  (the  applica:on  layer),  then  moves  from  layer  to  

layer  in  descending,  sequen:al  order.    

#  At  each  layer,  a  header,  or  possibly  a  trailer,  can  be  added  to  the  data  unit.  

#  Commonly,  the  trailer  is  added  only  at  Data-­‐Link  Layer  (Layer  2).    

#  When   the   formaYed   data   unit   passes   through   the   physical   layer,   it   is   changed  

into  an  electromagne:c  signal  &  transported  along  a  physical  link.  

Sunday,  August  8,  21   96  

Data exchange using the OSI model  

Sunday,  August  8,  21   97  

 The  TCP/IP  Reference  Model  

Sunday,  August  8,  21   Page  98  

The  TCP/IP  Reference  Model

#  TCP/IP  means  Transmission  Control  Protocol  and  Internet  Protocol.    

#  TCP/IP   protocol   model   is   another   popular   layer   model   that   describes  
network  standards.    
#  It   was   developed   by   Department   of   Defense's   Project   Research   Agency  
(ARPA,   later   DARPA)   as   a   part   of   a   research   project   of   network  
interconnec:on  to  connect  remote  machines.  
#  The  features  that  stood  out  during  the  research,  which  led  to  making  the  
TCP/IP  reference  model  were:  
–  Support  for  a  flexible  architecture.  Adding  more  machines  to  a  network  
was  easy.  
–  The   network   was   robust,   and   connec'ons   remained   intact   un'l   the  
source  and  des'na'on  machines  were  func'oning.  
§  The  overall  idea  was  to  allow  one  applica:on  on  one  computer  to  talk    (to  
send  data  packets)  another  applica:on  running  on  different  computer.  

Sunday,  August  8,  21   99  

Cont  .  .  .

# This  model  has  same  names  of  layers  as  OSI  reference  model  has.    
# Don't   be   confuse   with   same   name,   layers   at   both   model   have  
different  func:onality  in  each  model.    
# It   is   the   network   model   used   in   the   current   Internet   architecture   as  
well.    
# These   protocols   describe   the   movement   of   data   between   the  
source  and  des:na:on  or  the  internet.    
# These  protocols  offer  simple  naming  and  addressing  schemes.  

Sunday,  August  8,  21   100  

 TCP/IP  Model    

Applica)on  layer    

Transport  Layer    

Internet  Layer    

Network  Access  Layer    

Sunday,  August  8,  21   101  

Overview  of  Func)ons  of  TCP/IP  Layers

# Applica)on  layer:  
–  TCP/IP model combine the functionality of application layer, presentation layer and
session layer from OSI model in single application layer.
–  In TCP/IP model application layer do all tasks those are performed by upper layers
in OSI model.
–  Application layer deals with high level protocols, including data presentation,
compression and dialog control.
–  The specifications of this model describe a lot of applications that were at the top of
the protocol stack. Some of them were TELNET, FTP, SMTP, DNS
•  TELNET: allows connecting to a remote machine and run applications on it.
•  FTP(File Transfer Protocol): allows File transfer amongst computer users connected over a network. It
is reliable, simple and efficient.
•  SMTP(Simple Mail Transport Protocol): used to transport electronic mail between a source and
destination, directed via a route.
•  DNS(Domain Name Server) resolves an IP address into a textual address for Hosts connected over a
network.  

Sunday,  August  8,  21   102  

Cont  .  .  .  

§  Transport  layer:  

–  TCP protocol is used for reliable data delivery.
–  Flow control and error correction methods are used for guaranteed data
delivery.
–  It decides if data transmission should be on parallel path or single path.
–  Functions such as multiplexing, segmenting or splitting on the data
–  The applications can read and write to the transport layer.
–  Transport layer breaks the message (data) into small units so that they
are handled more efficiently by the network layer.
–  Transport layer also arrange the packets to be sent, in sequence.  

Sunday,  August  8,  21   103  

Cont  .  .  .  

§  Internet  layer:  

–  Selection of a packet switching network which is based on a connectionless
internetwork layer is called a internet layer provide all the functionality that
network layer provides in OSI model.
–  It is the layer which holds the whole architecture together.
–  It is responsible for finding the correct path for datagram [packet].
–  It helps the packet to travel independently to the destination.
–  Order in which packets are received is different from the way they are sent.
–  IP (Internet Protocol) is used in this layer.
§  Host-­‐to-­‐network  Layer  :  
–  A lowest layer and also called network access layer
–  It deals with LAN and WAN protocols and all the functionality provided by
physical and data link layer in OSI model.
–  Protocol is used to connect to the host, so that the packets can be sent over it.
–  Varies from host to host and network to network.  

Sunday,  August  8,  21   104  

 Similari)es  between  OSI  and  TCP  /  IP  
Reference  Models  
§  Both  models  are  based  upon  layered  architecture.  
§  The  layers  in  the  models  are  compared  with  each  other.    

 
 
 
 
 

§  In  both  the  models,  protocols  are  defined  in  a  layer-­‐wise  manner.  
§  In  both  models,  data  is  divided  into  packets  and  each  packet  may  
take  the  individual  route  from  the  source  to  the  des:na:on.  

Sunday,  August  8,  21   105  

 Differences  between  OSI  and  TCP  /  IP  
OSI  model     TCP/IP    
§  It  is  a  generic  model  that  is  based   §  It  is  a  protocol-­‐oriented  
upon  functionalities  of  each  layer.     standard.  
§  distinguishes  the  three  concepts,   §  TCP/IP  does  not  have  a  clear  
namely,  services,  interfaces,  and   distinction  between  these  three.  
protocols.     §  TCP/IP  protocols  layout  
Services  are  set  of  operations  and  Interface  
provides  a  way  to  implement  them  
standards  on  which  the  Internet  
was  developed.  So,  TCP/IP  is  a  
§  gives  guidelines  on  how  
more  practical  model.  
communication  needs  to  be  done  
§  the  protocols  were  developed  
§  the  model  was  developed  Girst  
Girst  and  then  the  model  was  
and  then  the  protocols  in  each  
developed.  
layer  were  developed.    
§  TCP/IP  has  four  layers.  
§  The  OSI  has  seven  layers  while  the  

Sunday,  August  8,  21   106  

 Physical  Layer

#  The   physical   layer   coordinates   the   func:ons   required   to   carry   a   bit  

stream  over  a  physical  medium.    
#  It   deals   with   the   mechanical   and   electrical   specifica:ons   of   the  
interface  and  transmission  medium.    
#  It  also  defines  the  procedures  and  func:ons  that  physical  devices  and  
interfaces  have  to  perform  for  transmission  to  occur.  

The  physical  layer  is  responsible  for  movements  of  individual  bits  from  one  hop  (node)  to  the  
next.  
Sunday,  August  8,  21   107  
 Layer  1:  The  Physical  Layer  :    

# lowest  layer  of  the  OSI  Model.  

–  It  activates,  maintains  and  deactivates  the  physical  
connection.  
–  It  is  responsible  for  transmission  and  reception  of  the  
unstructured  raw  data  
–  Voltages  and  data  rates  needed  for  transmission  is  
deGined  in  the  physical  layer.  
–  It  converts  the  digital/analog  bits  into  electrical  signal  
or  optical  signals.  
–  Data  encoding  is  also  done  in  this  layer.  

Sunday,  August  8,  21   108  

 The physical layer is also concerned with the
following:

1.  Physical  characteris:cs  of  interfaces  and  medium.  The  physical  layer  defines  the  
characteris:cs   of   the   interface   between   the   devices   and   the   transmission  
medium.  It  also  defines  the  type  of  transmission  medium.  
2.  Representa:on   of   bits.   The   physical   layer   data   consists   of   a   stream   of   bits  
(sequence   of   0s   or   1s)   with   no   interpreta:on.   To   be   transmiYed,   bits   must   be  
encoded  into  signals  -­‐  electrical  or  op:cal.  The  physical  layer  defines  the  type  of  
encoding  (how  0s  and  1s  are  changed  to  signals).  
3.  Data  rate.  The  transmission  rate  -­‐  the  number  of  bits  sent  each  second  -­‐  is  also  
defined   by   the   physical   layer.   In   other   words,   the   physical   layer   defines   the  
dura:on  of  a  bit,  which  is  how  long  it  lasts.  
4.  Synchroniza:on  of  bits.  The  sender  and  receiver  not  only  must  use  the  same  bit  
rate  but  also  must  be  synchronized  at  the  bit  level.  In  other  words,  the  sender  
and  the  receiver  clocks  must  be  synchronized.  

Sunday,  August  8,  21   109  

Contd.

5.  Line   configura:on.   The   physical   layer   is   concerned   with   the  

connec:on   of   devices   to   the   media   (point   –   to   –   point   or  
mul:point).    

6.  Physical  topology.  The  physical  topology  defines  how  devices  

are  connected  to  make  a  network.  

7.  Transmission   mode.   The   physical   layer   also   defines   the      

direc:on  of  transmission  between  two  devices:  simplex,  half-­‐
duplex,  or  full-­‐duplex.    
Sunday,  August  8,  21   110  
Data  Representa)on    

! The   word   data   refers   to   informa:on   presented   in   whatever  

form   is   agreed   upon   by   the   par:es   crea:ng   and   using   the  
data.  

! Signal:  the  electrical  wave  that  is  used  to  represent  the  data.    

! It  Can  be  analog  or  digital  signal  

Sunday,  August  8,  21   111  

Data  Representa)on  Techniques  

! The   type   of   data   to   be   transmiYed   can   be   in   the   form   of   text,  

audio,  and  video  in  the  form  of  electrical  signal,  radio,  laser,  
or  other  radiated  energy  source.  

! Data  can  be  analog  or  digital  

! The  term  analog  data  refers  to  informa:on  that  is  con:nuous;  

!   digital  data  refers  to  informa:on  that  has  discrete  states.  

! Example:    analog  clock  (with  second,  minute  and  hour  hands)  

and  digital  clock  
Sunday,  August  8,  21   112  
Analog  and  digital  signals  

! An  analog  signal  has  infinitely  many  levels  of  intensity  over  a  
period  of  :me.    
! As   the   wave   moves   from   value   A   to   value   B,   it   passes  
through  and  includes  an  infinite  number  of  values  along  its  
path.    
! A   digital   signal,   on   the   other   hand,   can   have   only   a   limited  
number  of  defined  values.    
! Although   each   value   can   be   any   number,   it   is   oMen   as  
simple  as  1  and  0.  

Sunday,  August  8,  21   113  

Analog  Signals    

!  Are   Con:nuous   wave   that   carries   informa:on   by   altering   the  

characteris:cs  of  waves.    
!  Analogue   means   that   the   original   informa:on   is   retransmiYed  
to  the  receiver/listener  without  any  manipula)on.  
!  Here   the   signal   can   take   on  any  value   (within   the   limits   set   by  
the  recording  equipment  and  the  transmiYer).  
!  It  measures  rather  than  counts  
!  For   instance,   Voice   and   all   sounds   are   analog,   traveling   to  
human  ears  in  the  forms  of  waves.  

Sunday,  August  8,  21   114  

Contd.  

 
 

!  Radio,   telephone   and   recording   equipment   historically   have   been  

analog,   but   they   are   beginning   to   change   –due   to   large   to  
computers  to  the  other  types  of  signals-­‐digitals  

!  The   disadvantage   of   analogue   signals   is   that   any   'noise',  

interference,  added  to  the  signal  at  any  point  cannot  be  removed  
from   the   audio   signal   and   this   degrades   the   audio   quality   of   the  
signal  or  causes  'hiss'.  

Sunday,  August  8,  21   115  

Digital  Signals  

! are   a   discrete   series   of   pulses   -­‐   either   high   or   low   -­‐   on   or   off   -­‐  
some:mes  expressed  as  binary  code  1s  and  0s.  
! It  counts  but  not  measures  
! Discrete  pluses  of  data  transmission  rather  than  con:nues  wave  
! More  prevalent  in  computer  based  devices  
! Flashing   light   and   telegraph   message   are   tradi:onal   examples  
of  on-­‐off  pluses,  although  not  in  binary  code.  

Sunday,  August  8,  21   116  

Contd.  

The  advantage  of  digital  signals  is  that    

1.  they   are   much   less   likely   to   be   degraded   by   interference  
(noise).    
2.  also   it   is   possible   to   send   a   lot   more   informa:on   digitally  
(e.g.   more   television   channels)   than   using   analogue  
technology.  
3.  They  can  be  transmiYed  effec:vely  along  fibre  op:c  cables.  
(on  -­‐  light  signal,  off  -­‐  no  light  signal)  and  these  are  cheaper  
to   manufacture   than   copper   wiring   and   can   cope   with  
mul:ple  signals  without  interference  between  them.  
4.  They   can   be   sent   directly   to   computers   which   use   digital  
systems  themselves.  
Sunday,  August  8,  21   117  
Contd.  

! The  signals  from  a  satellite  are  digital.  They  use  a  system  called  

'packet   switching'   which   is   the   same   as   used   in   computer  

networks   and   for   the   internet.   This   system   is   par:cularly  

effec:ve  for  dealing  with  noise  elimina:on.  

Sunday,  August  8,  21   118  

Contd.  

Which  of  the  following    signals  are  Analog    and  which  are  Digital?    

!  Volume  control  on  a  radio  

!  Traffic  lights  

!  Motor  bike  throYle  

!  Water  tap  

!  Dimmer  switch  

!  Light  switch  

!  In  Music  on  a  CD  

!  Music  on  a  tape  

Sunday,  August  8,  21   119  

 Reading  Assignment    

MODULATION  /DATA  ENCODING      

Sunday,  August  8,  21   120  

 Modula)on  
§  Modula)on  is  the  process  of  transforming  a  carrier  
signal  so  that  it  can  carry  the  informa:on  of  a  
message  signal.  
§  It  superimposes  the  contents  of  the  message  signal  
over  a  high-­‐frequency  carrier  signal,  which  is  then  
transmiYed  over  communica:on  channels.  
§  Modula:on  can  be  of  two  types  −  
–  Analog  Modula:on  
–  Digital  Modula:on  

Sunday,  August  8,  21   121  

 Digital  Data  to  Analog  Signal  Encoding  Techniques  
Ø Modula:on   involves   opera:on   on   one   or   more   of   the  
three   characteris:cs   of   a   carrier   signal:   amplitude,  
frequency,  and  phase.  
Ø  Accordingly,  there  are  three  basic  encoding  or  modula:on  
techniques  for  transforming  digital  data  into  analog  signals  
1.  Amplitude-­‐shiM  keying  (ASK)  
2.  Frequency-­‐shiM  keying  (FSK)  
3.  Phase-­‐shiM  keying  (PSK)  

   
Sunday,  August  8,  21   122  
Sunday,  August  8,  21   123  
 Analog  Modula)on  
§  Here,  the  analog  informa:on  signal  is  transformed  to  
the  analog  carrier  signal  so  that  it  can  travel  large  
distances  without  substan:al  loss.  
§  Analog  modula:on  can  be  of  three  types  −  
–  Amplitude  Modula:on  
–  Frequency  Modula:on  
–  Phase  Modula:on  

Sunday,  August  8,  21   124  

 Digital  Modula)on  
§  Digital  modula:on  is  the  process  of  conver:ng  a  
digital  bit  stream  into  an  analog  carrier  wave  for  
transmission  via  a  communica:on  channel.  
§  Digital  modula:on  is  broadly  divided  into  two  
categories  −  
     Bandpass  Modula)on  as  in  baseband  transmission:  
–  Here,  the  bits  are  converted  directly  into  signals.  
     Passband  Modula)on  as  in  passband  transmission:  
–  Here,  the  amplitude,  phase  or  frequency  of  the  carrier  
signal  is  regulated  to  transmit  the  bits.  

Sunday,  August  8,  21   125  

 Digital  
Modula)on    

Baseband   Passband  
Modula)on   modula)on    

NRZ(Non   NRZI(NRZ   Manchester   ASK(Amplitud FSK(Frequency   PSK(Phase   BPSK(Binary  

return  to  zero)   invert)   Encoding   e  Shic  Keying)   Shic  Keying)   Shic  Keying)   PSK)  

Sunday,  August  8,  21   126  

 DIGITAL  TRANSMISSION:  
§  how  to  convert  digital  data  into  digital  signals  ?  
§  It  can  be  done  in  two  ways:  
–   line  coding    
–  block  coding.    
§  For  all  communica:ons,  line  coding  is  necessary  
whereas  block  coding  is  op:onal.    

Sunday,  August  8,  21   127  

 Digital-­‐to-­‐Digital  Conversion      
§  line  coding    
–  Uni-­‐polar    
–  Bipolar    
–  Polar  :4  types    
•  Polar  Non  Return  to  Zero  (Polar  NRZ)    
•  Return  to  Zero  (RZ)    
•  Manchester    
•  Differen:al  Manchester    
§  block  coding.  
–  involves  three  steps:    
•  Division  
•  Subs:tu:on  
•  Combina:on.    

Sunday,  August  8,  21   128  

 Analog-­‐to-­‐Digital  Conversion    
§  Analog  data  is  a  con:nuous  stream  of  data  in  the  
wave  form  whereas  digital  data  is  discrete.  To  
convert  analog  wave  into  digital  data,  we  use  Pulse  
Code  Modula:on  (PCM).    
§  PCM  is  one  of  the  most  commonly  used  method  to  
convert  analog  data  into  digital  form.  It  involves  
three  steps:    
–  Sampling  
–  Quan:za:on  
–  Encoding.    
Sunday,  August  8,  21   129  
Assignment  #1  

§  Describe  the  following  digital  signal  encoding  formats  (which  format  is  

used  by  Ethernet?)  

1.  Non  return-­‐to-­‐Zero-­‐Level  (NRZ-­‐L)  

2.  Non  return  to  Zero  Inverted  (NRZI)  

3.  Bipolar-­‐AMI  

4.  Pseudoternary  

5.  Manchester  

6.  Differen)al  Manchester  

Sunday,  August  8,  21   130  

Sunday,  August  8,  21   131  
Sunday,  August  8,  21   132  
 Serial  transmission  can  be  either  asynchronous  or  
synchronous.    
§  Asynchronous  Serial  Transmission    
–  It  is  named  so  because  there  is  no  importance  of  :ming.  Data-­‐bits  have  
specific  paYern  and  they  help  receiver  recognize  the  start  and  end  data  bits.  
For  example,  a  0  is  prefixed  on  every  data  byte  and  one  or  more  1s  are  added  
at  the  end.    
–  Two  con:nuous  data-­‐frames  (bytes)  may  have  a  gap  between  them.    
§  Synchronous  Serial  Transmission    
–  Timing  in  synchronous  transmission  has  importance  as  there  is  no  mechanism  
followed  to  recognize  start  and  end  data  bits.  There  is  no  paYern  or  prefix/
suffix  method.  Data  bits  are  sent  in  burst  mode  without  maintaining  gap  
between  bytes  (8-­‐  bits).  Single  burst  of  data  bits  may  contain  a  number  of  
bytes.  Therefore,  :ming  becomes  very  important.    
–  It  is  up  to  the  receiver  to  recognize  and  separate  bits  into  bytes.  The  
advantage  of  synchronous  transmission  is  high  speed,  and  it  has  no  overhead  
of  extra  header  and  footer  bits  as  in  asynchronous  transmission.    

Sunday,  August  8,  21   133  

 ANALOG  TRANSMISSION    
§  To  send  the  digital  data  over  an  analog  media,  it  needs  
to  be  converted  into  analog  signal.  There  can  be  two  
cases  according  to  data  forma\ng.    
§  Bandpass:  The  filters  are  used  to  filter  and  pass  
frequencies  of  interest.  A  bandpass  is  a  band  of  
frequencies  which  can  pass  the  filter.    
§  Low-­‐pass:  Low-­‐pass  is  a  filter  that  passes  low  
frequencies  signals.    
§  When  digital  data  is  converted  into  a  bandpass  analog  
signal,  it  is  called  digital-­‐to-­‐  analog  conversion.  When  
low-­‐pass  analog  signal  is  converted  into  bandpass  
analog  signal,  it  is  called  analog-­‐to-­‐analog  conversion.    

Sunday,  August  8,  21   134  

 Digital-­‐to-­‐Analog  Conversion    
§  When  data  from  one  computer  is  sent  to  another  
via  some  analog  carrier,  it  is  first  converted  into  
analog  signals.  Analog  signals  are  modified  to  
reflect  digital  data.    
§  An  analog  signal  is  characterized  by  its  amplitude,  
frequency,  and  phase.    
§  There  are  three  kinds  of  digital-­‐to-­‐analog  
conversions:    
–  Amplitude  ShiM  Keying    
–  Frequency  ShiM  Keying    
–  Phase  ShiM  Keying    

Sunday,  August  8,  21   135  

 Analog-­‐to-­‐Analog  Conversion    
§  Analog  signals  are  modified  to  represent  analog  
data.  This  conversion  is  also  known  as  Analog  
Modula)on.  
§   Analog  modula:on  is  required  when  bandpass  is  
used.    
§  Analog  to  analog  conversion  can  be  done  in  three  
ways:    
–  Amplitude  modula:on    
–  Frequency  modula:on    
–  Phase  modula:on    
Sunday,  August  8,  21   136