FIRST TERM EXAMINATION [FEB.

-2016]
SIXTH SEMESTER [B. TECH]
COMPUTER NETWORKS [ETCS-306]
Time: 1.30 Hrs. Max Marks: 30
Note: Attempt Q. no. 1 and any two more questions.
Q.1. (a) What are the two broad classifications under which Networks can
be divided?
(2)
Ans. All computer networks fit in one of the two dimensions namely:
re (a) Transmission Technology, this focuses on the basic underlying physical network,
e.g. whether the nodes share a communication media or each pair of node has a
separate dedicated link. There are various topologies like mesh topology, ring topology,
star topology and bus topology etc.
‘
(6) Scale, it focuses on the scale of network how large is your network.
Q.1. (6) Explain the signficance of protocols in communication? (2)
Ans. Protocols are rules and procedures for communicating. When several
computers are networked, the rules and technical procedures governing their
communication and interaction are called protocols.
Protocols should specify rules governing the transmission. The protocol steps must
be carried out in a consistent order that is the same on every computer in the network.
In general, much of the following should be addressed: ine
1. Data formats for data exchange. Digital message bitstrings are exchanged.
Address formats for data exchange. Addresses are used to identify both the sender
and the intended receiver(s).
. 2, Address mapping: Sometimes protocols need to map addresses of one scheme
on addresses of another scheme.
3. Routing: When systems are not directly connected, intermediary systems along
the route to the intended receiver(s) need to forward messages on behalf of the sender.
4. Detection of transmission errors: Is necessary on networks which cannot
guarantee error-free operation.
5. Acknowledgements: Of correct reception of packets is required for connection-
oriented communication. :
6. Loss of information-timeouts and retries: Packets may be lost on the network
or suffer from long delays. , :
7. Flow control is needed when the sender transmits faster than the receiver or
intermediate network equipment can process the transmissions.
Q. 1. (c) Differentiate between UTP and STP cable.
Ans: Difference between STP and UTP.
Shielded Twisted pair Un-shielded Twisted pair

1, STP has a metal foil covering. 1. UTP does not have a metal foil covering.
2. STP gives better resistance to 2. UTP does not provide better resistance
electromagnetic interference to electromagnetic interference as
as compared to.UTP. compared to STP.
3. STP is little expensive the UTP. 8. UTP is less expensive than STP.
4. Grounding is not possible \-
4, Grounding is possible.
5. Possibility of cross talk is less as 5. Possibility of cross talk is more as it provide”
it resist to EMI. less resistance to EMI. i
6. Distance travelled is large and can | 6. Distance travelled is less and used in
LAN. :
be used is MAN.
 eee
por: pa
ublisher
P. versity-[B- Tech] cuted GORA
Ve puter Networks LE Uae
2, Sixth Semester, Com TDM and Statistic
=
22016 Between aSyn
chr onous
principle of ISD N? se aa
Q. 1. (d) Explaini the ih eaaa
Difs Le Statistical TDM
is a system of Synchronous TDM
ices Digital Network, This system Parameter | slots are allotted
stan ds for Integrated Serv of | In Statistical TDM
divi eee conn ch
whi i has been avai labl e for over a decade. Working = fa gyaehtortode TDM data flow dynamically. i.e.
input line is give
n slots
ections which the world using end-to- has data
ae phone
smitted simultancously across each input connecti
on is divided
in output frame if and only if it
allows voice and data to be tran basic types of ISDN service: Basic RateB into units and each
input occupies to send.
are two
end digital connectivity. There BRI consists of two 64 kb/s one output time
slot. , No. of slots in each
ry Rate Interface (PRD. No. of slots In Statistical TDM
the No. of input
Interface (BRD and Prima a total of 144 kb/s. This basic
service is intended
No. of Slots | In Sync
hronous TDM frame are less than
D channe l for pad
channels and one 16 kb/s
# aoa ee are equal to No-
individual users- ai s
to meet the needs of most ally the channel and only those input
with greater capacchan ity requirements. Typic ees P
is not done, frame is Buffering is done
t frame whose
PRI is inte nded for users nel for a total of 1536
kb/s. Buffers Buffering
interval are given slots in outpu
chan nels plus one 64 kb/s D sent after a particular one has r contains data to
send.
structure is 23 B of time whether some
buffe

operated simultaneou: contain both

of ISDN annels to be
—-+ eae : Sera Slots in Statistical TDM destination.
Some advantages L a ieee eee ronous TDM carry |
The chang e comes me Addressing | Slots in Synch there is no need data and address of the
multiple digitus' ee analog lines.
er 3
connections.
Speed: ISDN allows aoe honeec witches can support digital
wirin g data only and
as regul ar e line Itis possibl of addressing.g. Synch atio n
ronizizatio
Synchro
through the scteseee <hare a singl ionships
company S 5 Pe ssible
and pre assigned relat
when the telephone o i © .
the outputs that
= multiple ion routed toence between input and
es: ISDN allows ata Beir, ae ane addveus
C iS the info rmat
sources and have and interfer
Multiple Devt erent digital data °°" sier to keep the noise No synchronization bits
are used.
to combine many diff e tae hoe digital, itis ¢ Synchro- Synchronization bits are
used at
ee ee ee
proper destination. Sinc gending aring voltage sig
to ring the
nal begi nnin g of each fram e. bn
e signals- rate hannel nization _| the capa city of link is nor mally is less
out while combining thes a digital packet on a sepa Max. Bandwidth utilization
if all The ty of each
the phon€ eget Capacity than the sum of the capaci
Signaling: Insteadn-Bofand signal D105
inputs have data to send. channel.
bell in you r phone (“I multiplexer at
al”). ee ecas
; - : In Statistical TDM de-
- and Ban si gn
_of-B an Se TDM s each frame
“ t-of
(<Ou ? fing is Data rec eiving end decompose
Wha t is bit stu ffii ng? re of adding pits to stream ofdata. Bit stuf
Why ae tiplexer at rece
ivin g end
of eac: h dat a
Q.1 _ (e)
ie -_ ons; eis e local iaddres s eos oe
ols for the following reas aa ao
2unitrne
.
eee ted 5 re
r
munications protoc
the = ae
Sas ca e ae
extr a Ena ae
Ans. Bit stuffing: on. For example, man 1siy unit in turn . This
d Se
inf ormatiand end of a frame with device .
frame is then passed
¢ cont
d alas the rolinn
peg ing
h i i es “ i unit from
a beprere
ven t na
dat a being te ic er ps te 1 bits ine to destination device.
© To e itted has sux
ansm s
s ;
e eae | data data bei
ng tra 2 : advantage of
fram e- ba se d protocols, pc Be
a e care rded . Q. 2. (6) Name the e f four basici network topologies, and cite an
e 1 bits. must be disisca pachicenc: (4)
consecutiv uffed bit1 s :
end, the st metime mess 1 in tic .
the re aEceiving bits 4 re so |
Ans. Types of Netw ork sc hemalogy
Topo ds
d-s jze frame, gy ihe
ae en fe e 4 fi xe e Netw—ork’ nolo
Topol ogy is the schematic description of a network arrangement, , co: connecting
i
that requir size.
e
to this set bits are Pee 1 Bee
oes size equ al of data, zero i4
ee fr ameee inuo usen.stre am ae ci
make th e 4 co ntnot brok eee i
pro ’ to co ls th qu ir ed
at re stream 18 D! + 5
ac ke t beg in! :
network it has exactlhy tw endpoi
oery co er and ne
mputs, network devicenear
e For re that the pology is a . When ic ev th en it is ca ll ed Li
- certed to ensu B ected to .single cable Wh type in wh endpoint
inse’ i cousnn topology
t eapp a een beforeé the endo
loesn’ee Bus Top
s ofof Bus Topol
this sequence
ure +, appear ag
t ' ai
e sitage
A dvant ages ogy
nce ive
uensecut
bi Soeaco
oe 18- ka cat
,
10. ToS
01111110. makes
e every five g
Ai
ed after
i sert of multy iple :
gy.
aQis in e so ne ast comparred to ther network topolo
isticse
riatbe ed iisrle
Q. is 2. (@) Dedscst mnil
s Mea
ira
qul netweor.ks pa
Riried
:I | is
teasy t :
:: 3 ve ‘; By o
lexing : tworfo
r two «oe to taWT lk to an f
ea
ey M i e c }
s 2 Ne tw or ks at use multip an y ne k device r.10 This requires 8 e/ e
es
An _ le for tion for each pai L
cal
Tea l
oralke i i Fregicate a connec ther — 2 e,
I
“f
= To ma retch fur ; outed. to
i c ive resource st
, wi cause it forms a rin aesn comp
se
nnec
cohb
uter oO isneig
another computer
w av ctly rs for
nnected to ee
with ou
i
lt th e la st on e co .
ing between major me each device.
with
acllnitecabl
tete e oer otra
uplink.
i
-Bonemasa ee :
ac ro ss e u k fien
soos town
 2016-5
Sixth Semester, Computer Networks ]-AB Publisher
4-2016 LP. University-[B. Tech
it expects
the number of next frame
Advantages of Ring Topology * When the receiver sends an A! CK, it includes p of frames ending in fram
e 4,
to receive. For example in order to ac’kno wle dge the grou
as © n sender sees an ACK with
1. Transmitting network is not affected by high traffic or by adding more nodes, the receiver sends an ACK containing the number 5. Whe .
only the nodes having tokens can transmit data. number 4 have been received
number 5, it comes to know that all the frames up to
2. Cheap to install and expand.
ea eee
Star Topology 2s ea}
a | [2 Wea eee te oss
In this type of topology all the compu ters are connected to a single hub through L© 7 1°’
node. ©
cable. This hub is the central node and all o! thers nodes are connected to the central | Window
.
Advantages of Star Topology
Sliding Window*
1. Fast performance with few nodes and low network traffic. ;
2. Hub can be upgraded easily. Sliding Window on Sender Side
At the beginning of a transmission, the sender’s window contains n-l frames. .
3. Easy to troubleshoot. | _° As the frames are sent by source, the left boundary of the window moves inward,
Mesh Topology shrinking the size of window. This means if window size is w, if four frames are sent by
connection to other nodes or devices. All the network nodes are’ ponies cree a last acknowledgment, then the number of frames left in window is w-4.
It is a point-to-point
n devices.
connected to each other. Mesh has n(n-2)/2 physical channels to link : en the receiver sends an ACK, the source’s window expand i.e. (right bound:
Advantages of Mesh Topology : acetone
moves outward) tvcee all. He 2i a number of new frames equal to the number
oe of panes
frames
:
load.
1. Each connection can carry its own data :| ° For example, let the window size isis 77 (in(in fi fig. (a), if
sentieeaeee i frames 0 through 3 have been
2. It is robust. | eee << ae has been received, then the sender’s window contains
:
3. Fault is diagnosed easily.
3
° Now, if an ACK numbered 3 is recei z
4. Provides security and privacy. (2) vee received by receiver and are cide
Internet and Intranet. Snes es eet)
Q.2. (c) Differenciate between © sender’s window will ni =
Ans. buffer. At this poi point th le sender’sil window
now expand to include the next three frames in i
will contain six frames (4, 5, 6, 7, 0, 1). ie
Intranet
Internet

Itis the Network of Computer (i) In this the network is designed

@)
for specific group of uses.
which is open for all.
Internet Can’t be accessed (ii) Intranet can be accessed by This moves to th es
@ Internet But with same restrictions,
when Bn ACK is sent
from Intranet.
Unlimited Number of User. (iii) Limited number of user.
(iii)
(iv) Limited visitor traffic. ’
(iv) Visitor traffic is unlimited
with help of an exam}
Q.3. (a) Briefly explain the sliding window protocol

Ans. Sliding Window

by sender at a time befo
In sliding window method, multiple frames are sent
needing an acknowledgment.
on both se
Sliding window refers to an imaginary boxes that hold the frames ; 4
and receiver side.
Sliding Window on
¢ It provides the upper limit on the number of frames that can be
transmitt
ae
° At the Bevin ee Side
before requiring an acknowledgment. frame but not the frames. tiTansmission,
isa;
the Treceiver’s window
by receiver at any point even when window it
¢ Frames may be acknowledged ° As contains n-
full on receiver side. 4 ' the new frames
° Therefore 1 spaces for
the number the Bcc eee
of frarie the size of window shrinks
Frames may be transmitted by source even when window is not yet full onsenders€
scab) s that may stilbene not the number of
¢ The windows have a specific size in which the frames are numbe rede mod be sent ed with, out an ac frames received b ut
which means they are numbered from 0 to n-l. For eg. ifn = 8, the frames arenumbe knowledgment
ACK must
0, 1,2, 3, 4, 5, 6, 7, 0, 1, 2, 3, 4, 5, 6, 7, 0, 1,.... ‘ ee
© The size of window is n-1. For e.g. In this case it is 7. Therefore, a ™
n-l frames may be sent before an acknowledgment. so
 J
puter Netavorks e
2016-7
Sixth Semester, Com h]-AB Publisher
6-2016 It meas LP. University—| [B. Tec
is 7 as shown in diagram, (2)
size of receiver's window iu ,
e For example, let the ; fijeciondiandeworicatl
7 frames.
l offorthe first frame, the receiving
snsoneaspaces Q.3. (c) How is a hub related to a repeator?
i e
T y eet shrinks, Moving the
Ans. HUBS: It is a si ctivity device that is used to extend a network:
so the receiver may accept junction ca lled as hub there are
a centralnea
requere lel
3 types of hub passive,
boundary from space 0 to 1. requ Now, window has anaes : by one, ‘
the networks SE
i
ired send
to sen ackn owledged, the visti active and intelligent.
sixi matte: 0 thro
fram e 1 befor e iti
it
ugh is3 have arriv ed but have DOC been
ion of the n/w length.
ree Repeater: It is normally used for the expans
“ill contain three frame space
s.
ver expands to inely ie
ee es a physical
It generat signal. It receives a wea k signal. Amplified it and then
ACK, the window of the recei 5 y
7: ve receiver sends an dged fram es. retransmits it.
owle tp Re
of new frame spaces equal ts
110101. State whe ther the
y ackn
new placeholders as newl incl ude a numb er Q.4. (a) Aseven bit hamming code is received as
¢ The window expa nds to
the number of brevibus received is correct or wrong. If wrong, locate the error and correct it.
acknowledged frame minus
number of the most recently ow size is 7 and prio
if r ACK was for frame 2 &th Ans. For example: Suppose a 7-bit hamming codé is received as 1110101 and '
acknowledged frame. Fore.g., If wind
5 the window expands by
three (5-2). parity used is assumed to be even hence we can detect and correct the code as
current ACK is for frame
; Step 1: Received 7 bit hamming code is applied to hamming code format as
Receiverside
wadowo n
Shi
Sliding Window on Receiver side
Sauna
7; D7 D6 D5 D4 D3 D2 Pil

| Receiver window
[tfo[i[o[1{o]2]
TEPEEEPEEEEEEE—
Step 2: Checkbits for P4 bit
i.e. P4 D5 D6 D7
}-— Directon Directon
| 0 1 #1. 1| =odd parity hence error
This wall sores right This wall moves to the righ:
when a frame ts received when an ACK Is sent.”
So, P4=1
Step 3: Check bits for P2 bit
i.e. P2 D3 D6 D7
Therefore, the sliding window of sender shrinks from left when frames of data are sending, ~
The sliding window of the sender expands to right when acknowledgments are received. 0 1 #1 1) =odd parity hence error
¢ The sliding window of the receiver shrinks from left when frames of data are So, P2=1
received. The sliding window of the receiver expands to the right when acknowledgement
F Step 4: Check bits for P1 bit
is sent.
Q.3. (6) The codeword is received as 1100100101011. Check whether there i.e. P1 D3 D5 D7
are errors in the received code, if the divisor is 10101. (4)
0 1 1 1) =evenparity hence no error
Ans. To check codeword for errors, the received codeword is divided by the given
divisor, that is, 10101 as shown below. So, P1=0

411110001 Hence the error word is E =

10101 ) 1100100101011 F Jl eb 26
Step 5: Decimal equivalent of 110 is 6 hence 6th bit is
incorrect so invert it and the
correct code word will be, - 4 5, ae
feat

D7 D6 D5 D4 D3 D2 Di
[ifolijoliyoyay
Hence by using this method we can
detect a s well as correct the error
| “tare
transmitted coded word,
ord, But iti can locate a single
i iin th the
biti error and fails detecting the burst ©
Q.4.(b) Compare and contrast a circuit switched
network anda packet avitehaa
‘} network, el
Or be Raa Term Exam 2018 (Page No.:ny 1-2018) actsays
0101 A Iscuss the automatic repeat re bea
————— Remainder (CRC) ! i sian quest (AR x ‘
Ans. Refer Q9 of End Term Bxam 2017 (Page Nias een?)
Since the remainder obtained (CRC ) is non-zero, the received codeword contains error
 Maeeg
dems Saige
si
41+2%10+ 2 Ag
TERM EXAMINATION [APRIL-2016]
sisnntou +
SECOND [(0045) = 0000 00! are DFO00 or Ba
hexadecimal digits
Ans. (ii) Second four
SIXTH SEMESTER [B. TECH] [(DF00)=11011631111
0 =2%15+2
0000 900 8 + 1024 + header, or 88
-8 =80

COMPUTER NETWORKS [ETCS-306] = 327 68 + 84 + 4096 + 204 us the length of the

2% 8 ket min
of the whole pac
Max Marks: 30 Ans. (iii) Length
bytes. =85
ia attempt any two from rest. Length of whole pa' cke
t is calcul d ate as
2%4+ gn
g=644164+8 port), the
Note: Question No. lis compulsory and and the number 0 010 1 1000 =2*%6+ 8 (no t well-known
sing, we know th e first address [(0058) = 000 0 000
t nu mb er is 5 708
Q.1. (a) In classl ess addres If yes show the process. destination por
we find the prefix len gth? (2) Ans. (iv) Since the and SDH?
of addresses in the block. Can client to the server. tween PDH
is equivalent to the packet is from the fferentiate be ring, star,
er of addre sses in the block which Wha t is SDH technology? Di suc! h as point to point,
the numb in the e block, Q. 1. (e) topolo gie s
Ans. Yes: We know there are , 64 addresses e ts var iou s
raised to the power 0! ftwo. If prefix is 32-6 bits, or 26 bits Ans. SDH: SDH
sup por
tiplexing-
: m
‘number of digits in the suffix the TDM and octet mul are derived fro
fore there are 6 bits in the suffix, and linear bus etc. It uses e bi' t rate and
that is 246, there
—log 64 = 26 bits tip lex er’ s inputs are of sam :
n=32-logN =23 5? When is it In SDH, digital mul
is synchronous. ; at:
packet be less than are in phase. Hen ce it digital signals.
head er length field in IPV4 (2) common clock, hence SDH to combine
Q.1. (6) Can the
Inte rlea ving is used in with SDH rates
5. It is between Word (group of bits) can be integrated
exactly 5?
lengt h in an IP packe t cannot be less than r is a4 T, any of PDH rates upto 140 Mbp s
r heade As per CCIT
Ans. The value of heade is 20-60 bytes. The length of the datagram
length up to 155.52 Mbps. f
5 and 15. The header the length in unit less. SDH:
be divided by 4 to give ts or advantages of
byte word. So its value must
maxi mum value is 15. Following are the meri and dem ult iplexing technique
. :
value is 5 and tiplex ing king.
So 20/4 = 5, so minimum provided by the unrel
iable e Amore simplified mul supp orts multipoint networ
TCP, which uses the services working and SDH i
Q.1. (c) Expl ain how ; : B
e Synchronous net PDH signals.
able comm unic atio n.
they implement of tran sporting existing adm ini stration and
IP, can provide reli UDP lies in the way ° Capa bili ty
which enhanc es the
rence between TCP & pient automatia cally higher bit rates
Ans. The primary diffe rt guara nteed de livery, in which reci sender waits and e Easy growth to
de suppo the
data transfer TCP inclu age is received. In addit
ion
can > maintenance process. signals.
the sender when mess anteed message delivery. A UDP datagram sporting broadband
ifimplement guar and report e It is capable of tran ators.
retransmits the data not be able to detect supports different oper o conferencing, and
sion an e Itis multivendor and on LAN such as vide
get lost during transmis bility of TCP such as spor t serv ices
help to provide the relia by the receiver to e It provides network
tran
this.Few mechanisms a chec ksum, which is used
¢ Checksum-All TCP
segment carry interactive multimedia. out limit in SDH.
head er or data. h can be increased with
detect errors with eithe
r the TCP
ts to be duplicated in
packet © Optical fibre bandwidt
It is possible for packe on to traff ic is offe red by rings.
duplicate e Switching prot ecti
¢ Duplicate and detection- keeps track of bytes received in order to discard
from failure.
switched network; there fore TCP e SDH allows quick recovery same bit rate and
copies of data that has alrea
dy been received. t xer’s inputs (bit streams) are of
data, TCP must implemen PDH: In PDH, digital multiple ors. Each will differ within
to guarantee delivery of different oscillat
use of positive are derived from different clocks from
issi ons- In order
° Retr ansm be lost or damaged. The The ed_plesiochronous.
retransmission schemes rms successful reception
of data. e of few clock periods. Hence it is call
toleranc
receiver to the sender confi d calls for retransmission. ine digital signals.
acknowledgments by the coupled with a timeout perio Bit Interleaving is used in PDH to comb
posit ive ackn owle dgme nts, be delivered
lack of for packets to
hed networks, it is possible
¢ Sequencing -In packet switc segm ents it receives so it can deliver 44.736 Mbps
to prope rly sequ ence 64Kbps (0S0) 1.544Mbps 6.312 Mbps <2 eee
out of order. It is TCP’s job Os-2 Ds-3
plication in order. os-4 400.352Mbps
the byte stream data to anap data sent. The sending TCP
tain vario us static and dynamic timers on length of time. If the
{f os-s
© Timers-TCP main owledgment within a bounded 2 Primary | Mas
reply with an ackn we
to retransmit the segment.

ae
waits for the recei ver sende r can channels}
an ackn owle dgme nt, the xa
M23
timer expires before receiving

-
hexsadecimal format:
a dump of UDP header in (2)
Q.1. (d) The following is
0045DF0000580000. =
port number? R28
(i) What is the source 139.264 Mbps
ion port number?
(i) What is the destinat PDH Level Hierarch
the data?
(iii) What is the length of er or vice versa?
in North Am: 2
from the client to the serv
(iv) Is the packet directed
0045 which gives the decimal value
are
69, Ben ee Mees
Ans. (é) First four hexadecimal digits

—_
 LP. University-[B. Tech]-AB Publisher 2016-11
10-2016 Sixth Semester, Computer Networks
Since
‘ng are the limitations of PDH: ission and in data
layer. Hence
t differe nt frame is used for transm
Saa
is very complex. .
multiplexing and de-multiplexing
s the whole system t
o be de-multiplexed
tribut ary require nd wi dt h.
e Accessing lower is limite din ba
for PDH is 566 Mbps, which
¢ The maximum capacity
pg ene noeees
° Tolerance is allowed in bit rates. A-> E> BCost=2,4,3
configuration.
¢ PDH allows only Point-to-Point
ss ew Bcost 2
dif ferent multiplexing
own stand Brdsy PDH also has (cost = 4)
: ae ae
Similarly B has shortest path to C 9+4=6.
ate interconnecting networks|togstner. becomes A> B > C Cost =
hierarchies makingit difficult to integr figure-
i -. AC
shown jn the Cost =24+4=6
Q.2. (a) Consider the network
: 3 [ and for D
;
A> 5D Cost
E =2<0
«AD cost =2 Ais shared
1 vector of A. This update vector of
No other updations can be made in on can be made in case of each
when no updati
with B and E. The final table is obtained
possible pathA +D, (4) and every node
routing, fin dout the best Ato D using distance vector
Using distance vector d istan ce from 4 The table results in
Ans. We have to find
out the shor test
ee os Cc F|

Hlran
AINwWAOHM
ce vector to all other nodes From

QINownann
Blomranu]|
Routing. me e rin d a os
or rou tin g erat
In distance vect gnodes-)

+t+xtoQnr
ZMNOARANDE,
neig
+ to reachthe ing nodes and

No)
whi vector W1 thits neighbour
ee sch or vector share its which is shortest is found through their

Gaao
wonton
cs other path
neighbouring nodes.
be like.
Initially the table can

Therefore the distance from A to D is 2

es
q Q.2. (b) Name the ATM layers and explain their functions (4)
Ans. ATM: Asynchronous Transfer Mode (ATM) is a standard switching technique
designed to unify telecommunication and computer networks. It uses asynchronous time-
division multiplexing,(1][2] and it encodes data into small, fixed-sized cells. This differs from
distance i
This table consists of the approaches such as the Internet Protocol or Ethernet that use variable sized packets or
a». Consid eri ng Nod e A: fais - link oyey a

ATM Endpoint ATM Switch ATM Endpoint

ATM Adaptation ATM Adaptation -

it will use vectors of B and E to find out the shortest Layer
y' z j -_ Layer _
Ahas 2 neighbours B and £, so
table updates to
paths to all other nodes. The ATM Layer ATM Layer ATM Layer
ToA na 6) ae D
From
©O 2 © Physical Layer Physical Layer
o
a

Physical Layer
Ooogwn

(similarly the complete table is updated)

Physical Medium
 Se

er Networks 2016-13
12-2016 Sixth Semester, Comput
her
AB Publis
ATM
AT physical layer has fourfunctions. LP. Uni versity-
[B- Tech]-
The ATM Physical Layer: The longs
ream, the transmission a nd receipt of bits on the Physica
converted into a bitst
Ue ay address be
ageq edi, Aie (0-127). So, this IP
es are tracked, and cells are Pack 124, 240. 106. 248
|

are controlled, ATM cell boundari f class-

cells are to the ; s in the range °
appropriate types of frames for the physical medium. For
example,
| This IP address fall
Packages
differently for SONET than for DS-3/E-3 media types. to class A. }
-191)
he physical medium. d : is 147.120.12.194
gs-B 1.e- (128
The ATM physical layer is divided into two parts: t (ii) IP address
in the range of © Ja
(PMD) sublayer and the transmission convergence (TC) sublayer. “Pendens This IP address falls
administratoo)r
belongs to class sets . 10.0. The
The TC sub layer has four functions: cell delineation, header error conty,
1 So, this IP address plock 130.56
is granted
sequence generation and verification, cell-rate decoupling, and
transmissioy (Hoy Q.3. (a) An organization
subnets ; ee
adaptation. The cell delineation function maintains ATM cell boundaries
amet wants ‘~ create 1024 a caetee
of addresses 12 oe
devices to locate cells within a stream of bits. HEC sequence generation and verife (1) Find the number address of the
and last
generates and checks the header error control code to ensure valid data, q oe (2) Find the first
mask.
decoupling maintains synchronization and inserts or suppresses idle (unassign l-ratg (3) Find the subnet last subnet. default mask
ae he a Clas B address 50 the
cells to adapt the rate of valid ATM cells to the payload capacity of the tres ) AT (4) Find the first and testa eski is
a mas
ronet
- ed sub
system. Transmission frame adaptation packages ATM cells into frames acce eUssion
t 10 bits for the sub nets then oe eee
the particular physical layer implementation. Ptable ty is eee a ae
ee S e s
Oxfffftfcd, or ae ee
be specified as areata
/16 +10 =/26. (Could 55.192. gee
net mas will be 255.255.2
ATM Adaptation Layers: AAL1: AAL1, a connection-oriented service, is suitableg k
oe constani t bit rate sources (CBR), such as voice and videoconferencing. ATM trang for format) therefore, the sub t be used for the ee in = seca
(1) The remaining bits
mus ante ee
- wee using cireuit-emulation services. Circuit-emulation service also accommodate or 2°= 64 bits. Y or ee
the address component
ae nent of equipment currently using leased lines to an ATM backbone netw, a have 32-26 = 6 bits for oca tab le Each subnet mas.
not actu ally all
ao . timing synchronization between the source and the destination. Fo subnet as they are
ye
a 62 valid addresses. ‘
1depends on a medium, such as SONET, that supports clocking. 1 are
mee
The first and last allocatab le addresses 1n subnet
(2)
ae
i t nature. This is 130.56.0.1 to 130.56.0.62 :
called variable bit rate e (VB
(VBR) traffic, Th; i :
ee a services characterized as packetized voice or video Working address se the
that do not ea is found by ANDing the
Seapssay transmission speed but that do
have requirements similar to constantfc ‘ The first address for the block 0 0000000 0 (130.56.0.0). ee
0011100 0 0000000
melt ee= = suitable for VBR traffic. subnet mask 26 giving 10090010 0 00
The AAL2 process uses 44 bytes of the i ‘ address, so the next one 1S 1000001
aoe ae — 4 bytes of the payload to support the AAL2 proce But this is not an allocatable
1s characterized as either real-ti 00000000 00000001 (130.56. 0.1) so
: acl is the last one of the: 64 addresses
Lea supports both types of VBR traffic Finally, we note that the broadcast address s so it must be
ee ne that broadc ast addres
ptation Layers: . AAL3/4:le AAL3/4 supports
tati the last allocatable one is the one before
ae
both connection-oriented and _ 40000010 00111000 00000000 00111110 (130.56.0.62)
(3) The subnet mask will be 126. i.e.
10000010 00111000 00000000 00000000. ;
AnAAL3/4§ AR PDU header consists A of Type, 4 .193 Last edges in 1024 subnet
Sequence Number, and Multiplexing qa (4) First address in the 1024 subnet = 130.56.255.
Identifier er fi fields. Type fields identi
identi fy whether a cell is the begin = 130.56.255.254
ning, continuation, or (4)
_ Q. 3. (6) Explain the process of connection establishment in TCP.
Ans. The TCP three-way handshake in Transmission Control Protocol (also called
the three message handshake) is the method used to establish TCP socket connections
and tear down TCP socket connections over the network. TCP’s three way handshaking
aptatio
ATM Adapt n La Layers: 4 AALS: AALS is the pri
ation technique is referred to as the 3-way handshake or as “SYN-SYN-ACK” (or more
ae SG and connectionless data thie eae | i ~ accurately SYN, SYN-ACK, ACK). The TCP handshaking mechanism is designed so
oe ae ae
= such as classical IP over ATM and that two computers attempting to communicate can negotiate the parameters of -
LAN Emulation Ce fe soi
ee a oe and efficient adaptation layer (SEAL) becau thenetwork TCP socket connection before beginning communication. This three way
ae bites inat se Dane aa
ep e e CS-PDU and segments it into 48-oct handshaking process is also designed so that both ends can initiate and negotiate
et SAR-P DUs witho
it ut reserv
re ing separate TCP socket connections at the same time. Being able to negotiate multiple
Q.2. (c) Find the class of the followi owing TCP socket connections in both directions at the same time allows a single ph: a 1
IP address. @ network interface (such as ethernet) to be multiplexed.
(69) 01111011 11110000 01101010 11111000 (ii) 147,120.12 ; TOR ATE Se Loe
.194 , 3-Way Handshake Description |
Ans. (é) IP address is 01111011 11110000 01101010 11110
00 Host A(client) sends a TCP SYNchronize :
Decimal notation of given binary form is
Host B receives A’s SYN aA ehh
 2016-15
Sixth Semester, Computer Ne tworks LP. University-[B. Tech]-AB Publisher
14-2016

Host B sends a SYNchronize-

ACKnowledgem ent © The header checksum field is recomputed. ast one of the
a packet is a fragment if at le
Reassembly: A receiver knows that
Host A receives B’s SYN-ACK following conditions is true: except the last.)
Host A sends ACKnowledge is set. (This is true for all fragments
¢ The “more fragments” flag the
(This is true for all fragments except
Host B receives ACK * The “fragment offset” field is nonzero.
TCP socket connection is ESTABLISHED. first.)
fragments using the identification
field. The
The receiver identifies matching ents with the same identification field
from fragm
receiver will reassemble the data receiver receives
the more fragments flag. When the
using both the fragment offset and flag set to 0), it can calculate the
the last fragment (which has the “more fragme nts”
al data payloa d, by multip lying the last fragment’s offset by eight,
length of the origin
data size. In the examp le above, this calculation was
and adding the last fragment’s
495 x 8 + 540 = 4500 bytes. t order, by
it can put them in the correc
When the receiver has all the fragments, ssing.)
their offsets. It can then pass their data up t he stack for further proce
using value by 8. This
multiply th e offset
To find the number of the first byte: We
means that the first byte number is 800.
is 879
Since payload is 100 ‘bytes. So last byte number (3)
bucket algorithm.
Q.4. (6) Write a short note on leaky a small hole at the
hm: Consider a Bucket with
Ans . Leaky Bucket Algorit
bottom, whatever may be the rate of water pourin:
g into the bucket, the rate at which
out from that small hole is constant. This scenario is depicted in figure (a).
water comes
bucket is full, any addition al water enterin g it spills over the sides and is lost
Once the
hole underneath).
(i.e. it doesn't appear in the output stream through the
same idea of leaky bucket can be applied to packets, as shown in Fig.b.
The
: bucket. And the following steps
of the datagram. AU
Conceptually each network interface contains a leaky
and explain the each field are performed: -
Q. 4. (a) Draw IP datagram is 100 bytes. What the bucket.
off set is 300 and payload size host has to send a packet, the packet is thrown into
packet has arrived in which the (5) ¢ When the
network interface transmits
and the last byte? The bucket leaks at a constant rate, meaning the
is the number of the first byte receiv es a packet , it examines :
en a router rate.
Working of IP: fragmentation:Wh packets at a constant
Ans. to use. The interface has” ° Bursty traffic is converted to a uniform traffic by the leaky bucket.
mines the outgoing interface
the destination address and deter the router may fragment
the packet. 4
is bigger than the MTU, ¢ In practice the bucket is a finite queue that outputs at a finite rate.
an MTU. Ifthe packet size
31 This arrangement can be simulated in the operating system or can be built into the
hardware. Implementation of this algorithm is easy and consists of a finite queue.
8 16__19
0 4
|
- | Version| IHL | Type of Service Total-Length Whenever a packet arrives, if there is room in the queue it is queued up and if there is
4 no room then the packet is discarded.
Flags} Fragment Offset
Identification -
Header Checksum |
Time to Live | Protocol
Source IP Address
Destination IP Address
a
es
| Padding » Inflow may a S
Options
; ‘A - he bursty av
e
Fig. IP Frame Format
ts. The max size of each segment is the
™ ne router divides the packet into segmen
the IP header size (20 bytes minim um; 60 bytes maximum). The router |
MTU minus following changes:
nt packet having
puts each segment into its own packet, each fragme : Z
Constant
Outflow |
size.
¢ The total length field is the segment hichis
ts except the last one, w:
¢ The more fragments (MF) flag is set for all segmen
set to 0.
the segment in the origina
e The fragment offset field is set, based on the offset of
of eight-byt e blocks. ae
data payload. This is measured in units Fig. (a) Leaky bucket (b) Leaky bucket implementation
 "Eee

[ Y-JUNE
ION MA
-2016]
TERM EXAMINAT
puter Networks
IB. TECH]
Sixth Semester, Com
16-2016 (2) END
SIXTH SEMESTER 306]
ORKS [ETCS- Max
of NAV in CSMA/CA.
Q.4. (c) Explain the role nt of time required for
transmission
Ans. NAV allow s stati ons to indicate the amou
ely follo wing the curr ent fram e. | COMPUTER NETW Marks: 75
of required frames immediat mandatory frames. The compulsory.
rve Time: 3 Hrs. Q.No. 1 which is
This is important to rese reservation for frames stions including
al carrier se: Note: Attemp: t any five que
importance of NAV virtu col. Typic ally these are control frames, but not switching 3
n of the 802.11 proto
critical to operatio nt data and acknowle dgement Q.1. Attempt All.
n circuit swi
tching, packet
acknowledgements, subseque te bet wee
always. They include 802.11 data and ackn owledgement frames follo
wing Q.1. (a) Differentia
itching “
as part of a fragment burst
, and message switching. g and Pac ket-Sw
frames Message-Switc! hin (channel) betwee
n
an RTS/CTS exchange. part of the Ans. Circuit-Switching, d ban dwidth circuit
tion field whic h is Therefore, network establishes 4 fixe sic all y connected
the 802.11 MAC header Dura f Circuit switching if the nodes wer e phy
The NAV is carried in as discussed previously
.
may communicate, as n, as opposed to
coded at a vari able data rate
due to insu fficient nodes before the users during the connectio
MPDU payload, and is en to decode t he MPDU The pit delay is constant
the area may be able de the PLCP
E ” with an electrical circuit. varying delay.
not all stations within § i be able to deco packet queues may cause e and
ons within the area NAV protects use short
er inter- packet switching, where the circuit
c is released
e
SNR. However, all stati that the by other callers until in a dedicated circu
it then,
Since the critical frame s
still gain access to the Each circuit cannot be used takin g place
header used for CCA. those frames should a busy no communication
is
that are available
for new
s (SIFS for example), to transmit and stations will detect connection is set u p. Even if . Chan nels
frame spacing value pts NAV il able to other users
of circuit switching
befor e any other stati on attem
the adde d protection through that channel still remains unava hone net work is example
medi um However, be idle. Telep
CCA carrier sense. calls to be set up are said to
medium through e circuit
the Duration field value system. gy th at may emulat
reservation is lost. 802.11 he ader extract ket switching technolo before any packets
are
able to decode the specified (in microseconds
), Virtual circuit switching is a pac’ is established
Stations that are amo un t of time that the connection
me dium as busy
for the
s define din the 802.1
1 specification switching, in the sense in order.
and use it mark the ow the stri ct rule
transferred, and that packets
are delivered messages were
mitter 5! jhould foll mes. packet swi tching, where
Therefore, the trans be sent within fra was the precursor of
the NAV value to Message switching was first introduced in 1961. Nowadays,
for calculation of one hop at a time. It or circuit-
routed in their entirety, over packet-switched
are mostly implemented
message switching systems is example ofa message switching
system.
E-mail
switched data networks. in which packets (discrete
Packet switching is a communications paradigm with other traffic. The
nodes over data links shared
blocks of data) are routed between
from your computer is broken up
fact that the data stream
term “packets” refers to the which are then sent out onto the network.
bytes (on average),
into packets of about 200 for routing the packet from
with information necessary
Each packet contains a “header” data stream is independent.
;
packet in a
source to destination. Each it permits “statistical multiplexing”
packet-switching is that
The main advantage of can share aline,
packets from many different sources
on the communications lines. The packets are
very efficient use of the fixed capacity. With current technology,
allowing for If the network
accepted onto the network on a first-come, first-served basis.
generally
or discarded (“dropped”).
becomes overloaded, packets are delayed i peas i
adaptive p and non-ad aptive
Q.1. (6) Differentiate between
cies

These algorithms change their routing

Ans.1. Adaptive Routing Algo rithm:
and in traffic as well. These get their routing
ldecisions to reflect changes in the topol ogy
The optimization parameters
‘ormation from adjacent routers or from all routers.
time. This can be further
the distance, number of hops ani d estimated transit
sified as follows:
gets entire
(a) Centralized: In this type some central node fen the network
information about the network topology,
about the traffic and about other nodes. This
= ae been AS! to ae es routers. The advantage of this is that
ep the information. The disadvantage is that i
ntral node goes down the entire network is down, i.e. single point eee eae
 2016-19
B. Te ch ] -AB Publisher
mputer Networks LP. Univ er si ty -[
ative
Sixth Semester, Co g excellent use of altern
18-2016 routing without
seeking
s alg o rit hm has the property of m akin qu eu ed lin k.
the node decides the the sta tus ofa
,
interconnected thi sending the pac ket ont
o the lea st
thi s met hod
does no t know ab
ou t lemented by (5)
(6) Isolated: In The sending node ough a congested routes. It is usually imp thod in detail.
oth er nod es. nd thr
information fro m the packet may Q.1. (c) Explain
disadvantage is that thm for routing are:
©
king of Error Dete
ction
particular link. The Some exa mpl es of this type of algori Ans. Parity Chec in the given word inc
luding
route resulting in a del ay.
es to
,
get rid of it as fas t as it can
n par ity me an s the number of 1 ’s
de, it tri Even parityuld —- Eve
a packet comes to ano to where tha t link leads. A be even (2, 4, eu:
e Hot Potato: When out put que ue-without regard alg ori thm . _ the parity bit sho number of 1’s in
the given wo rd including the
g it on the sho rte st wit h the hot pot ato Odd parity means the
by put tin static routin g Odd par ity —
thm is to combine t both the static
weights
-).
variation 0: f this algori the routing algorithm takes into accoun par ity bit should be odd (1, 3, 5,... , on thi the type of fthe
dinng
the ~
When a packet
arr ive s,
Bit: The parit
ity bitbl can be set e to 0 andid ependi ,
lengths. Use of Pari ty par
of the links and the queue e gets modified’s _ parity required. vy ees
at each nod ; ire
Lea rni ng: In this method the rou ting tablesea
to pace learningi : marae
to 1 or 0 suc h tha t the no. of “1 bits” in the ent
e Backwa rd ng pac ket s. One a
hop cou nte r ev er this bit is set
m the incemi a
by information fro each pac et, togeth
the the $ source node innde receives a packet in
a particular line, it Pe
er with
tae (a).
-+w of aa h of “1 bits” in the entire
to include the
i i
ide nti ty of
nod e. If the aie t h i is set to 1 or 0 such that the no.
ch it from the sourcerent i one th en
; parity , y , this bit
on & ; ‘or odd 2
that is incremented
ber of hops it :
has taken to rea ; better tha han the cur word is odd. Shown
in Fig. (b).
dow n the num nod e 1s fut ure use.
notes : in the d for
ho count stored ue is better then the value 1s update ep |. Data bits Agl
revious value of apc cur rent val not recall the
e but the n it can
Sta gis don hen the best rou te goe s dow n the sto red
this is that w have to forget
The problem with all the nodes |
a particula r node. Hence
ea best route te
ally and start |
ring nodes
dic ation fr om its neighboudvantage is”
thi s the no de receives inform ck et . Th e di sa
In the pa
(ce) Distributed: sion about which way to send the paket somethin
g
and then takes
the-deci
it re ce iv es in fo rmation and sends fe
if in be tw ee n the the interval ;
3
that layed. u uting
packet may be de thsa ms do notApa base their poro
changes then the The se e e ead ction Take Place?
Routing Algori
th m: topology. Inst How Does Error Dete
2. Non-Adaptive tim ate s of the curren raffic and advance, off-line, ; er can detect the presence of an error if
the parity of the
decisi
ts and estimates
isi ons on measuremengoing from one node to the other is co
mputed in
tic Par ity ch ecking at the receiv me an s, if it is known that the
be tak en in boo ted . Thi s is als o known as sta ent fr om the expected parity. Th at
received signali
the rou te to twork is is dif fer ng to be “even” and if the
routers when the ne i Se ee
signal is always goi ¢ signal is not correct.
and downloaded to the ket is ee a a AF S a ee e rece ceiv ed
further classified as: every incoming pac ee narceta a
a
e receive byte and request
for
routing. This can be technique in which Apa ctroris Aaa the ignae
ore
ng ada pts the ob le m wi th ie ?sa me byt e to
(a) Flooding: Flo
odi
on which it arrived.
oe pr
e severalap +o|tr . ansmission of the
ng line except the one thisi a node may receiv
sent on every outgoi loo p. As are sul t of ove rco me
kets may g0 in a hniques adapted to Receiver
method is that pac esirable. Some tec ,
of a par tic ula r packet which is und
copies Sender data
as follows: . When a node
these problems are ket is giv en.a sequence number
ry pac e fin ds 1011011
e Sequence Number
s: Eve number. If the nod
see s its sou rce add ress and sequence pac ket and w i
it. it the
receives the packet then it will not transm
Reject data
same packet earlier
that it has sent the ' , Compute
s if
just discard it. ociated with it. Thi parity bit
Eve ry pac ket has a hop count ass n the hop coun !
¢ Hop Count: whi ch see s it. Whe
h node
ented) by one by eac Compute
decremented(or increm pos sib le val ue) the packet is dro
pped. parity bit
xi mu m ati o
becomes zero(or a ma y on tho se lin ks that lead to the destin
packet is sent onl transmissio
¢ Spanning Tree: The
tree routed at the source.
This avoids loops in
the network. 1011011 1 ets
ing a spa nni ng e kno wle dge of
by construct edi ate nod es hav
when all the interm
but is possible only
where hig]
topology! lications. But in cases
ng is not pra cti cal for general kinds of app ons , flo odi ng is of grea > Transmission
Bloodi
is des ire d suc h as in military applicati media
degree of robustness
i
help. by the node to one of Fig. Even-parity checking scheme
Walk : In this method a packet is sent net wor k is high l
(6) Ran dom
high ly robu st. Whe n the
algorith m is

,
 er Networks 2016-21
Sixth Semester, Comput Tech]-AB Publisher
20-2016 | Lee, University-[B. 00
10010110.110101 11.00010000.000000
p | Data bits — 450.215.016.000
Subnet Address therefore, is 150.21
5.016.000. 0 and a divisor
ee
Transmitted The subnet addres s, for a 10-b it sequence
101001111
CRC
code P Q.2. (a) Compute
the a
of 1011. 1010011110 We app
end 000 and divided
by 1011
Ans. Given Sequence:
Received code Code: 000 (1001001110
with one error 1011)1010 011 1100
1011
(5)
ISDN, broadband ISDN
between narrow band as the basic
Q.1. @ Differentiate the use of a 64 kbps channel 1011
is based on
Ans. The narrowband ISDN contribution
orientation. The major technical
t switching
1011
unit of switching and has a circui DN supports very high —
has been frame relay. The B-ISn. The major technical
of the narrowband ISDN effort a packet switching orien tatio
as
1100
) and has , also known
data rates (100s of Mbps chronous transfer mode 1011
DN effort has been asyn
contribution of the B-IS
cell relay. (5) 1110
g using an example. s on a subnet —
Q. 1. (e) Explain subnetin ing of conne cte d network devices. Node of |
1011
logical group a LAN. The pourpose
Ans. A subnet is a y to ea' ch other on |
to be loca ted in close physical proximit to keep netw ork data separated to 1010
tend s, but ©
provide extra addresse s. If your network
has
subnetting is not to to not waste addresse ' 1011
Subnetting allows you esses of a class C netw
ork.
prevent congestion. the full 255 addr
would be unwise to use 010 —Remainder
ed amount of host |
only 10 hosts then it ess with only the need
efficient to use an addr in other networks.
|
It would be much more be used 10100111100010
ing addresse s migh t So,CRC encoded sequence is n the semantics of
bits so that the remain mat of ATM Cell. Also explai
subnetting the network: ts to reduce traffic and
speed Q.2. (b) Give the header for (6.5)
The major reasons for
ork into smaller segmen ;
1. To divide a large netw i ? each filed in the header.
in the B-ISD N ATM
of your netw ork. { of inform ation transf er
up the sections
across geographical area
s. Ans. The ATM cell is the basic unit up the header
2. To connect networks of 53 bytes. Five of the bytes make
communication. The cell is comprised ation field. The following is the
belongs to. An IP address the user inform
Subnet Mask
e what subn et an IP address field and the remaining 48 bytes form :
ATM Cell Header
A mask used to determin
e network address and the host address. For example, consider
first
structure of the Network Node Interface (NNI)
has two components, th this is part of a Class B network, the
the IP address 150.215.017.009.
Assuming
address, and the second
two ATM Cell Header
represent the Class B network
two numbers (150.215) host on this network.
a particular
numbers (017.009) identify divide the host part of
administrator to further
Subnetting enables the network case, a part of the host addre
ss is reserved
or more subne ts. In this
the address into two the IP address in
parti cular subne t. This is easier to see if we show
to identify the
ss is:
binary format. The full addre
01.00001001
10010110.11010111.000100
The Class B network part is:
10010110.11010111
and the host address is: é
00010001.00001001 g the Generic Flow Control
t mask is the netwo rk addre ss plus the bits reserved for identifyin
The subne
network address are all set to 1, though ii The Generic Flow , Control (GFC) field is a 4-bit
- fi eld that wa igi
subnetwork. (By convention, the bits for the cas support eee uel of a oonBynet to shared access SoReiatecans ae
as in the network address.) In this
would also work if the bits were set exactly f e Dual Bus DB) ring. The GFC fiel i i
11111.11110000.00000000. It’s calle
therefore, the subnet mask would be 11111111.111
to which an IP address belongs b) ee clvaceathactlie ch weiss 4 bits in which to Adisbaes vir Ria poeta Ns es
a mask because it can be used to identify the subnet e cells of various ATM connections. Howe a
the IP address. The result is
performing abitwise AND operation on the mask and of the GFC field have not been standardized, and the Me hinclecsoacia done
subnetwork address: j
Subnet Mask 255.255.240.000 11111111.11111111.11110000.00000000 Payload Type (PT) / Cell Loss Priority (CLP) / Header Error Control (HEC):
IP Address 150.215.017.009 10010110.11010111.00010001.00001001 When user information
‘ k mite.is present or: th 3 ATM cell has suffered traffic congestion then
 i
ks AB Publisher 2016-23
Sixth Semester, Comp
uter Networ LP. University-[B. Tech]-
22-2016 ding byte is to be them
r the corres pon N users want to communicate, some of
tell the sy stem whethe pri ority in regard to. ‘ And due to this problem. If more than reason for lack of bandwidth
bit is used to hav e a of band widt The
h. main
The CLP
esti on . ATM cells wi th CLP=0 gestions, CLP=1 will be denied permission for lack hardly ever transmit
wor k cong urce con, been assigned a frequency band
discarded during net CLP =1. Therefore, during reso is that some of the users who have
a single channe l into static sub channels is quiet
cells with or Sigel anything. So. dividi ng
cell loss than ATM is dropped. for sensing an d correcting FDM can easily be seen from a simple
queuing
re any CLP=0 cell inefficient. The poor performance of static
cells are dropped befo fiel d and is use d n tim e del ay is T, for a cha nn' el of capacity C bps, with an
for the ce 1] header theory calculation. If the mea an exponential
each frame having a len gth drawn from
_
HEC is a CRC byte
cell errors and in del ine ati ng the cell header.
Y/VCD: The role ofthe arrival rate of lambda frames/sec, mean i/i/* bits/frame. With these parameters the
tual Channel Identifier (VP probability density func
tion with
ice rate is “C frames/s
ec.
Virtual Path Identifier / Virual Path or Virtual Channel identification numberand s, _
arrival rate is lambda
frames/sec and the a)servFor example, if Cis 100 Mbps, The mea
(AC -la mbd
n
I fiel ds is to ind ica te Virt dis tin gui she d. A uni que m que uin g theo ry. T =i c,
VPI/VC tion can be
ing to the same connecadvance to indi
is | Fro bda, is 5000 frames/se
so that the cells belong 1 cateé which type of cell and the frame arrival rate, lam
fie r is assign ed in lling channel ORS generic
; frame length, is 10,000 bits.
/VC I ide nti met asi gna
separate VPI al layer OAM cel
Is, | then 7 = 200 sec. mic channel
gned cells, physic : Unlike the static channel,dyna
following, unassi e: d for that Dynamic Channel Allocation c is nonuniform and heavy.
dcast signalin _ cha nne l. how : is QoSS negonet tiat cation: is efficient3 and is used in area s wh ere the traffi
tio n est abl ishedie and naling . Sig nal ing perform: scall allo
: d in dynamic channel alloc
ation.
ns must be considere
a l connec d sig
(ii) How is & virtua
dee
Five main assumptio of independent
through the process Cc related to the network, _ The num ber
nnection? This is done ons el: Sta tio ns are also called terminalss . lambda, and probability of a
er functi 1 Station Mod ependent constant arrival
rate
aring, and many oth j stations are N, with ind tx Jambda). Once a frame
io and call cle for ATM signaling: interval of (delta t) is (delta me
There oeare two sets of sta nda rds __ frame being generated ina time
ion does not hin g unti l the fra has successfully been
stat
UNI sign ing vate ATM switch, or . "a been generated the
nend station and a pri | ransmitted.
is point of view, all stations
are
NNI signaling k: The UNI signaling umption: From hardware
med ts eats ‘AT M net wor
duced by the ATM — 2. Single Channel Ass mun ica tio n on whi ch all stations can transmit
nal ing is per for ® sta nda rds are pro l is available for com
UNI sig te ATM switch an P
fta1996).
NI
4.0 is an addition to ’ equal. Asingle channe
routing. x Th on it and all can receive from it.
snt 't involve Q.2931 brou. ght es are transmitted simul
taneously, they will
Signaling protoco!
doe
tre beca usee it
i t d
n Assumption: If two fram t collision and it
and UN ork byin ISDN and Frame Rela
nle e 3. Coll isio ty to detec
simp y) on has the abili
e resultinI g in a false signa 1. Each stati
a
Forum oSare 118 aaded Or on
-yed from (2
the 99 4) col lid d later.
d call eri ps public ATM network.
Since a” collided frame must be retr
ansm itte
ul
must be kept in mind that N
UN pe TTU-T which
is further derived from °¢ stch es the routing becomes very i
is perf orme een ey)
stan dard s 4, Time Man age men t:
begin at any insta nt as there is no
NNI signaling oe ! 3 NNI sign aling has
two majo r
Time : Frame transmission can
rall y invo lves Cont inuo us
public net- ae gene etwork into discrete intervals.
and PNNI (Priva Network-to-N master clock diving time time slots. A slot
rtan t com) nent of the NNI signi stocol 1)which uses static routing and is designed) Fram e transmission start at
the beginning of the
successful or collision
impo rsw1 Si Slot ted Time :
espo ndin g to an idle,
TISP (inte ted Inte that uses very elab orat e
is a signaling protocol to large ATM networks may contain 0,1.
or more frame corr
2 simple signa eal le small respectively.
Interface) IISP is orks hand tran smis sion
for small private ATM netw
switches.
of thousanes of ATM 5. Sensing of Channel: e trying to use it. Ifa
al oo eee dean even tens gorithm in be sensed by station befor
dynamic routing prob lem and its solu tion Sense: A channel can use it, unitl it gols idle.
, tho ion y Carr ier atte mpt to
which can have hundreds as busy, no stati on will
and its solution. on senses the channel to use it. First
el wa ie the problem who gets to use stati the channel before trying
Q3. Explain ¢! jggue is how to determine No Carr ier Sense: Station connot sense busy or idle
where the channels is
i
Provide an example
to illustra used to determine smit and the came to
ork, oe a The protocols layer called the) the tran
link layer in detail.
Ans. In any proadcast netw of the data link the vari ous issu es in designing data HEB)
© pe Ee A ‘sub-layer i Q. 4. (a) Discuss :
the channel when there © ” functions to carry out. Thes
e
belong channél) has a number of specific face to the network layer,
next on a multi-access
e to allocate a single proadcast Ans. The data link layer
MAC (Medium AcceAllo ss Control) pe _ |
functionsinclude prov
iding a well-defined
service inter
grouped into frames,
dealing with
The Channel cation Problem 18 OW i ting a single ¢ the bits of the physical layer are slow receivers are not
dete rmin ing how or fram es so that
competing users. cal Le of ae
(F
transmission errors rs. and regulating the flow we will examine each of ‘
these issues
channel ocation: The most typi portions. sende In the following sections
See Divi tt -siz ed swap ed by fast
ultiple competin g users is Frequency into INequaa “ED
Ar turn
the panduidth is divided small and constant, : .
s: ”
among ™ are the following issue
:
numb er of users are Ther e
e ee 10n. 7 If the when the num of ork Layer
; Ee
3 Netw
a. ed one E ided to the
\ (a) a)Serviices Prov
mec han ism . ver, vat l§—( b)
ee a etricient allocationk can bea simple example ofis this 00 oPestedi
type. Howe
heavy: FDM presents a ae i
oe
é any and cons tant or the traffic of user s ere
A telephon the numb er
N regions
ey divided into spectrum
ee bt a large piece of valuable
O) tbe oY ting is Jess than N
 a
5

i
Ne tworks 2016-23
Sixt h Semester, Computer LP. University-[B. Tech]-AB Publisher
22-2016 is to be '
corresp onding byte N users want to communicate, some of them
is used to tell the system whether the ori ty in reg ard to : And due to this problem. If more than th
The CLP bit hav e a pri dth.The main reason for lack of bandwid
M cells with CLP=0
es tion .AT ns, CLP=1_ will be denied permiss ion for lack of bandwi
d a frequency band hardly ever transmi
t
discarded during network cong CLP=1. Therefore, during resource congestio is that some of the users who have been assigne
cell loss tha n AT M cell s with
anythin g. So. dividin g a single channel into static sub channels is quiet
pped. or receive easily be seen from a simple queuing
ore any CLP=0 cell is dro g and correcting — The poor performance of static FDM can
cells are dropped bef d and is us ed for sensin inefficient.
T, for a channel of capacity C bps, with
an
the cell hea der fiel If the mean time delay is
HEC is a CRC byte for .
theory calculat ion.
each frame having a length drawn from an exponential
ineating the cell header CD: The role of the arrival rate of lambda frames/s ec,
}
cell errors and in del parameters the
fie r / Vir tua l Cha nnel Identifier (VPI/V fication numbers, ~ probability density function with
mean i/i/* bits/frame. With these
Virtual Path Identi tual Channel identi frames/ sec and the service rate is AC frames/sec.
Path or Vir © arrival rate is lambda
to indicate Virtual shed. A unique and For example, if C is 100 Mbps, The
mean
VPI/VCI fields is tion can be distingui From queuing theory. T =i (AC-lambda)
ong ing to the same connec whi ch type 0 f cell is ©
rate, lambda, is 5000 frames/sec,
© frame length, is 10,000 bits. and the frame arrival
so that the cell s bel e to ind ica te
I ide nti fie r is assigned in advanc alli ng chan nel or a generic
separate VPI/VC al laye r OAM cells, metasi
gn
; | then T = 200 sec. channel
following, unassigne
' d cell s, phy sic { Allocation : U: nlike the static channel,dynamic
ae al ; Dynamic Channel form and heavy.
sig nal ing channel. oe Se re the traffic is nonuni n.
broadc ast
ablished and tt ¥ and is used in areas whe ami c cha nne l all ocatio
l con nec tio n est allocation is efficient be consid ere d in dyn
Five main assumptions mustiot ns are also called terminals. The number of indepei ndeofa
(ii) How is @ vir tua Tae Pp nt
cess called aa 7 ne
done through the pro ate
connection? This is y oth er fun cti ons rel 1 Station Model: Sta ndent constant arrival rates lambda, ? and P probabil ty '
setup and call clearing, andstaman : N, with inda epe bda). Once a frame
ndards for ATM signaling stations are e interval of (delta t) is (delta t x lam
, framé being generated in a tim n does nothing until the frame has successfully been
*
There are two sets of ;
: : tio
UNI signali ing rivate ATM switthiGn has been generated the sta ;
i ed.
NNI signaling a2 ek transmittes re péint of view, all statio
ns are
d between anpublend sta tio n and ap
k, The UNI by signaling is umptio n: Fro m har dwa nsmit
ing i ic AT M net wor — 2. Single Channel Ass n on whi ch all stations can tra
e d the e P he standar aalere produc ed the ATM
nnel is available for com
mun ica tio
UNI signaling’ e
between 4 private AT
M switch an is an add iti on to equal. A single charec eive from it. y will
4.0 (1996). (UN I 4.0
931 brought on it and all can itted simultaneously, the
doesn't involve roulin
g
Signaling protocol Q.2 n: If two frames are transm ect collision and it
simpler because it led 1 (1994) are eae me Relay). 3, Collision Assumptio n has the abil ity to det
ISD N and Fra e signal. Each statio
collide resulting in a fals collided frame must be retransmitted later.
Forum and are 1 use d in
ived from the ee wa Q93 ablic ATM network. Sin .
ce a
UNI 3.1, UNI 3.1 is der iro nae s very must be kept in mind that , \
whi ch is er derived
sew itc hes the rou tin g bec ome
by the ITU-T, between the sw? 4, Time Managemen
t:
instant as there is no
NNI signali is performed two major standards
;
me transmiss ion can beg in at any
es er t (orANG ign aling has Continuous Time: Fra .
erally inv olv etwork te intervals
diving time into discre
k gen wor k-t o-N
of the time slots. Aslot
lic net- wor Pri vat e Net
pub
t of the NNI sign ing
. ade
g and is designed master clock iss ion start at the beginning cessful or. collision
jmportant compo- nen
2 pest which uses static routin porate: Slotted Time: Frame tra nsm
ponding to an idle , suc
witch Si
IISP (Integrated Inters ple signe th
is a signaling protocol to lar tworks may contai n 0,1. or more frame corres
P isa sim ll ATM respectively.
Int erf ace ). [IS
net wor ks. P i asil y handle smausa nds of ATM swi transmission
small
for ami privating e AT M thms ghee a ten s of tho l:
jyn c rou algori 5. Sensing of Channe tion before trying to use
it. Ifa
on algorithm in detail nnel can be sensed by sta
dre ds, tho usa n‘ roblem and its soluti ) Carrier Sense: A cha tio n will att empt to use it, unitl it gols
idle.
e hun , no sta
which can hav and its solution. station senses the channe
l as busy to use it. First
l soe : roblem ermine who gets to cha nne l bef ore try ing
Q3. Explain channe how to det n con not sense the
No Carrier Sense: Statio re the channels is busy or idle
gue is
e to {Illustrate the Pp
provide an exampl etermine whi
d cam e to whe
network, ee ae ae pro toc ols use
called the! the tra nsm it and the link layer in detail.
Ans. In any proadcast r of the data link layer issues in designing data
oeiz ‘su b-l aye ‘ Q. 4. (a) Discuss the various (6)
the channel whenaccthe re is Pe ry out. These
mul ti- ess ch ; eae oe
ast ce num ber of specific functions to car
nex t ona bro adc has a lay er,
-« how to allocate 4 single
lay er wor k
, ‘Ans. The data link service interface to
the net
m Access Con tre viding @ well-defined ling with
MAC (Mediunne l Allocation Problem
18
‘o a single chant functionsinclude pro
phy sic al lay er are gro uped into frames, dea
ers are not
The Cha ; the bits of the tha t slo w rec eiv ae ues
M). If determining how ulating the flow or frames 2 so : each of these iss
competing users. ee ion err ors and reg ‘ 5 min e
Allocation: The
mos t typ ica l yey, ot
MusP ; a transm; iss
the fol low ing sections we will exa
oe a ti pence users is Freque ncy Divisi on portions. is a! ;
into N e ual-sizedtant swaped by fast sen der s. In
.
sesnaere tiple compe tinging 5 PB atdod
in q' , FDM
ng mul dth is div ide d in turn. ets issues: ‘
iis fusers are N, the pandui gma ll and con sta n™ There are the followin
;
num; ber of users are the Network Layer
isi assl Ae : the nis
portioi n If m. (a) Services Provided 3
A
e of this type. However b eam Framing
ar simple exampl heavy:of FD M presentsy 80 (b)
apaelncien Ae. the traf
tant orion fic is ber
num users currentl
:
i
or’ e e a the
picler aa ee cee
of senders is not one
: u
spectrum
piece of valuable
ee
anammuDl . ee than N a large
 Publisher 2016-25
puter Networks LP. University-[B. Tech]-AB
sixth Semester, Com two
24-2016 the fiber on and off at will. These
various wavelengths of light traveling down to provide
network bandwidth provisioning
(c) Error Control ‘ ata characteristics of fiber cable enable dyna: mic
(d) Flow Control an example. (6.5) for data traffic spikes and lulls.
detail an using enc e - Fiber has a very low rate of
bit
lai n sli din g wi ndow protocol in 3. Resistance to electromagnetic interfer rence
Q.4. (6) Exp ation 2016: so resistan t to electromagnetic interfe
First-Term Examom (4) (error (10 EXP-13), as a result of fiber being
Ans. Refer Q.2. (a) in detail. Fiber-optic transmission are virtually noise
free.
ing cab les
Q.5. Explain the fol
low secure transmissions - Fiber provides
ze 4, Early detection of cable damage and
Coaxial Cable. : net | cable, is a 50-
ohm as there is no way to detect the data being
-Q. 5. (a) Baseband Jled an nernet
Ether an extremel y secure transmis sion medium,
cable, commonly cal om energy “leaking” through the cable
Ans. A bas eba nd coax ial
<i
al. Transmission
in org transmitted by “listening in” to the electromagnetic
ed digi tal sign i : a transmissions. By constantly monitoring
‘an unmodulat e es Be as is possible with traditional, electron-based
cable that. transmits at any point propan time it takes light to reflect down the
ctional, mea nin g a signal inserted wor ks. | ~ en pee aa an optical network and by carefully measuring the
is bidire d in local area net fiber, splices in the cable can be easily detected.
are commonly use e eront ee ie
Baseband cables ulated, analog Bien -
75-ohm cab les that transmit mod by div idi ng into a Disadvantages of Fiber Optics: .
are this
can compensate for lle nan ae
ing
1. Installation costs, while dropping, are still high -
Despite the fact that fiber
pnidirectional, but transmit stti
ting oe aeae :
g fiber optic cabling
are not capable of installation costs are dropping by as much as 60% a year, installin
Baseband cables e <a expanding beyond its
restricted to data i cate OD ee is still relatively costly. As installation costs decrease, fiber is
steady oe in
in leng th experience 4 nd ca original realm and major application in the carrier backbone and
is moving into local
exceeding 0.62 mile
advantages of usin i
g baseba absence of a , etc), and
of using baseba
n d cable is the
i
loop, and through technologies such as FTTx (Fiber To The Home, Premises
benefit nic broadband access.
of use-Another Pte PONs (Passive Optical networks), enabling subscriber and end user
cost and the ease are a
bec aus e the cable’s signals : 2. Special test equipment is often required - The test equipme
nt typically and
mode m a sing:
LANss. Baseband coaxia
ansmit l cable Bara traditionally used for conventional electron-based networking is of no
use in a fiber
Abaseband aei dae te ae rhe ae ooh
le is ae e ies - ital signaling. Baseband ee p8, optic network. Equipment such as an OTDR (Optical Time Domain Reflecto
meter) is
baseba nd Eap
ar ae va Re e oe and bg cat specialized optical test equipment such as optical probes are
git al tra nsm ission’. For 1Km io c needed at most fiber endpoints and connection nexuses in order to pro; ] id
a Se ‘di data rates or per
used with low 4) testing of optical fiber. spies eahee
. it
i ca ble
coaxial 12-2018) ; 3. Susceptibility to physical damage - Fiber is a small and compact cable, and
Q.5.(b) Broadband 2017 (Page No.: tion
Ans. Refer Q4(b)
of First Term Exam (4.5) is highly susceptible to becoming cut or damaged during installation or construc
f-way for fiber optic installat ion,
le activities. Because railroads often provide rights-o
Q.5. (c) Fiber cab i ht travels down
a
n the pipe railroad car derailments pose a significant cable damage threat, and these events can
fiber op ics: Lig t) bounces dow prune service to large groups of people, as fiber optic cables can provide tremendous
Ans. Working of (particle of ligh ggin
wall s. Eac h ae ant on
i + ) m of light, travelin
a bea
repeat edl y off the you mig ht expe c
ee a ee ang é ee ners ee capabilities. Because of this, when fiber optic cabling is chosen as
an ice run. Now light hits Beer transmission medi ium, itit isi necessary to address restoration, backup and
i i
to leak out of the
edge s. But if
wee i a ' pec e eente. i
ey n—as though ha
Sa
ees) , it refle cts back in agai ecti on. It Ss
es ian 42 degr total interna 1 refl J
4. Wildlife J damage i to fiber optic cables -- M. Many birds,
i for example, find
enon is called
mirror. This phenom : a Kevlar reinforcing material of fiber cable jackets peetceianiy ai poaitng: as SaaS
de the pipe. the cable, whichi
keeps ligh t insi 7 the .structure of material, so they peck at the fiber cable jackets to utilize bits of that material
ps light in the nie e s called
le—i
The othe r thin g that kee
of the ca’ le—in the midd
f two separate par
ts. 'The mai n part
d aro und the out sid e of the co Q.6. Explain
K the
1 network
: layer in the inte: rnet and the network la: i: TM.
h. Wrappe
a that 's the bit the light travels throug ding ’s job is to keep the sign
networks in detail. Differentiate in between these two: ree atta 5)
ae called the cladding.
The clad
t type of glass to thecor
Ans. The network layer in the internet ;
i use it is made ofa differen
ee oe ae this beca refr acti ve inde x than the core. List a ‘ ‘ The main objective of the > network la yer is i to j allow endsystems
cladding has a higher to leak into the claad aren networks, to exchange information through intermediate Ce ae eae a
(More technically, the ligh t that tries
than in the core. Any ‘he unit of information in the network layer is called a packet. es
slower in the cladding
de the core.
tends to bend back insi
optics: medi Transport
Advantages of fiber ed data transmission Packets
dwidth - No other cable -bas
1. Extremely high ban —
fiber does.
offers the bandwidth that Using many of the Datalink
e increasing bandwidth - fiber ¢ in
2. Easy to accomodat ent can be added to the inert
cabl ing, new equ ipm
generations of fiber optic fiber. oe¥
city over the originally laid
that can provide vastly expanded capa ing the ability ¢
iplexing, lends fiber optic cabl
Dense Wavelength Division Mult
 2016-27
Publisher
Sixth Semester, Computer Netw
orks LP. University-[B. Tech]-AB
26-2016
¢ Flags : It has three bits -
Network Layer Functions layer include: and if M is 0, then it
performed by the n etwork more fragments on the way
Some of the specific jobs normally ° Mbit : IfM is one, then there are
device that comm uni cates over a network has is the last fragment.
° Logical Addressing: Every address. For example, not fragment such a packet.
sometimes called a layer t hree ° DF bit : If this bit is sent to 1, then we should
associated with it a logical address, rk layer pro tocol and every mach
ine
Protoc ol (IP) is the netwo
on the Internet, the Intern et ¢ Reserved bit : This bit is not used.
after it.
has an IP address. header and everything that comes
of interco! nnected networks
is probably the ¢ Total Length : It includes the IP
e Routing: Moving data across a series : Using this field, we can set the time within which the
k layer. ° Time To Live (TTL)
defining function of the networ messages packet should be delivered or else destroyed.
network layer normally encapsulates
e Datagram Encapsulation: The (also called packets) with a to which we should hand over the packet
placing them into datagr ams ¢ Protocol : This specifies the module
received from higher layers by .
(UDP or TCP). It is the next encapsulated protocol
network layer header. ges down
The network layer must send messa ATM networks:
e Fragmentation and Reassembly: technologies have limits
da’ ta link layer architecture is designed to support
to the data link layer for transmission.
Some The asynchronous transfer mode (ATM) protocol
quality of service. The user data is
on the length of any message that
can be sent. k the transfer of data with a range of guarantees for
ls are used at the networ and transported over virtual
e Error Handling and
Diagnostics: Special protoco traffic, to divided into small, fixed-length packets, called cells,
are logicall y connect ed, or that are trying to route connections. ATM operates over high data rate physica
l circuits, and the simple structure
layer to allow devices that or the devices themselves. re, which improves the speed
the status of hosts on the network of ATM cells allows switching to be performed in hardwa
exchange information about
and efficiency of ATM switches.
Packet Structure notice is that, as well
Figure 1 shows the reference model for ATM. The first thing to
Version Header |
| Length Type oF Total Length (16 bits) as layers, the model has planes. The functions for transferring user
data are located in
ervice (8 bits)|
Number
(4bits) | (4 bits) the user plane; the functions associated with the control of connections aré located ir
and planes
the control plane; and the co-ordination functions associated with the layers
eon) Flag Offset (13 bits)
ID (16 bits) are located in the management planes.

ee Header checksum (16 bits)

Time to Live (8 bits)
Source (32 bits)

Destination (32 bits)

Higher-layer protocols
Options
ATM adaptation layer
4 (0100).
current version is Version
¢ Version Number : The ple of 4bytes and so we can have ATM layer
er will alway s be a multi (20 —
e Header Length: Head axim um size of the header is 60 bytes Physical layer
field as 15, so the m
a maximum length of the
:
bytes are mandatory ).
The structure is:
* Type of Service (ToS) : y of the packets,
Fig. 1: ATM reference model
fy the precedences i.e. the priorit
First three bits : They speci The three-dimensional
1 ) representation of the ATM protocol architecture is i
Next three bits: a to portray the relationship between the different types of protocol. The ena
¢ D bit - D stands for delay. indicate the encapsulation of protocols through levels of abstraction as one layer i:
.
e T bit - T stands for throughput built onmn topto} of another, ȴwhereas the verti ertical planes indicate
indi the functi i
« B bit - R stands for reliability.
pacination of the actions taken by different layers. An etntane at ate
Last two bits: The last two bits
are never used. fone e 2 ae a user Beas is that it introduces a degree of qhanpendeee ‘
e
her will represent the fragments of a uniqu ons: the protocols for transferri
ID Field : The source and ID field toget 7 separated from the protocols for controlling Sadie
packet. So each fragment will have a different ID. ermal sce ied pS
represents wher e in the packe t, the curre nt The protocols in the ATM layer provide communication between ATM switche
¢ Offset : It is a 13 bit field that
t. So the packet size can be at while the > protocols
fragment starts. Each bit represents 8 bytes of the packe prot in the ATM ad japtation
i |. AAL
size in bytes as a multiple users. This is illustrated in the searaple A
most 64 kB. Every fragment except the last one must have its *
eae 2 Bebe ae
ce with this struc ture. ;
of 8 in order to ensure complian
 Publisher 2016-29
LP. University-[B. Tech]-AB
ks
Computer Networ ees,
Sixth Semester, and the e:
—2016
28-20 1 tween the users R) e ntries in its routing tabl
ent ifi ed i n Fig. 1. One be hes ) within A route has the following (CID
ace are id
ween the nodes (sw
itc
Two types of interf and the other bet Address/ mask
Next hop
work int erf ace ),
network (user-net Inter face 0
node interface). 135.46.56.0/22
the network (network- | Usel_ TM adapution
Inter face 1
i AAL layer 135.46.60.0/22
User | cxcleoregeere sSesshes
i p| ATM | ATM layer 192.53.40.0/23 Router 1
| sicral
Phylaye Default Router 2
' PHY | PHY i | PHY
PHY | PHY
u Phys ical medium For each of the following IP address,
Physical medi
1010
User-network (a) 135.46.63.10 10000111 00101110 00111111 0000
Network-node Network-node 100 0000 0000
User-network interface 41111111 11111 111 11111
interface interface 135.46.63.10
interface k
Fig. 2. ATM networ el, I shall 255.255.252.0 AND AND
in the ATM reference mod 10000111 00101110 00111100 0000000
cri bin g the fun ctio ns of the three layers bas ic typ es of cell . 135.46.60.0
Before des
mat of ATM cells. Fig.
3 shows the two
briefly describe the for matches found, so it’s forwarded
be 8 bits It matches entry with 135.46.60.0/22, and no other
8 bits Byte number :f
)
|
to interface 1.
eS 4 identifier virtual path
i h identifier . — fo virtual channel 10000111 00101110 00111001 00001110

Jepeey Wily
j
Generic now control wena Pal
resis
Virtua l path identifier © “identifier (b) 135.46.57.14
virtual channel 2
41111111 11111111 11111100 00000000
vietwal path identifier
ot
~ 135.46.57.14
identifier
i virtual channel identifier
sopeoy WAN

3
7 Virtual channel identifier 4 Fe vimual pane V" payload type !
CLP 255.255.252.0 AND AND
10000111 00101110 00111100 00000000
eeeHeader erroree
eh
fevieck 135.46.60.0
Header error control
5
control so its forwarded
It matches entry with 135.46.56.0/22, and no other matches found,
i)

6 Cell payload
to interface 0.
Cell payload :
53 i.
(b) (c) 135.46.52.2 10000111 00101110 00110100 00000010
(a) 11111111 11111111 11111100 00000000
large image the 135.46.52.2
(b) ATM cells at view
ls at the user- - AND
‘

Fig. 3 (a) ATM cel interface t 255.255.252.0 AND

and the rem ain ing
network-node der is five
I byt es long 135.46.52 .0 10000111 00101110 00110100 00000000
Each ATM cell consis
ts of 53 bytes: the hea
on fro m hi ghe r laye rs. The only different
carry informati use r-network interface
carryt It doesn’t matches any entry, so it’s forwarded to the one defined in default entry,
bytes (the cell pa: load) is that t he cells at the Raaly: Routan’
= ;
the two “ig 3 of ATM cell tha t onl y eig ht bits ar
between from users. This means
data Gel field for thewhtflow control of data 12 bits at the network-node
interfa
iden tifi ers, rat her tha n
available for virt ual pat h tab le. (d) 192.53.40,7 11000000 00110101 00101000 00000111
ries in its routing
following (CIDR) ent
Q.7.Arouter has the 192.53.40.7 11111111 11111111 11111110 00000000
Next Hop
Address/Mask 255.255.254.0 AND AND
Int erface 0
135.46.56.0/22 192.53.40.0 11000000 00110101 00101000 00000000
Interface 1
135.46.60.0/22
135.53.40.0/22 le . It matches entry with 192.53.40.0/23, and no other matches found, so it’s forwarded
Router 1
3 to Router 1.
Default Router 2
addresses, what does the router do
ifa pac
For each of the foll owin g IP a . (e) 192.53.56.7 11000000 00110101 00111000 00000111
with that address arrives? 192.53.56.7 11111111 11111111 11111110 00000000
; (a) 135.46.63,10 255.255.254.0 AND AND
(b) 185A6.57.14
: 192.53.56.0 11000000 00110101 00111000 00000000
(c) 185.46,52.2
(d) 192.53.40.7
(e) 192.53.56.7
 2016-31
TechJ-AB Publisher
uter Networks LP. University-[B.
Sixth Semester, Comp it has all the features
30-2016
the one define: d in
default ent, s the faded signal. In addition,
entry, 50 it’s forwarded to Hence, the repeater regenerate repeaters and the Hubs
are that only one
It doesn’t matches any on probl em betw een the es
ofa Hub. One comm cular time. If multiple devic
on the network at a parti
namely, Router 2. following:
(12.5, transmission can take place be data collision.
s on any two of the simul taneo usly, there will
Q.8. Write short note transmit data
rs
Registration and Registra
Q.8. (2) Domain Name regisby is a databas of aj
Ja the Signal)
Active (it can amplify
e Regi stration: A domain name inteor
of dthea Netw [ Hubs =
Ans. Domains Nam istyty opera
ope tor, also )calle
(DNS n e | Passive (it can’t)
in a top~ level d domain.in. Areg
A regis
domain name registered of the Doma i n Nam e Syst em
is a part
Information Center (NIC)
of domain names. d online meaning
that keeps the database have to be regis' tere
rs: Name for websites have been approved
by
Domain Name Registra also comp aine s that
which at are ion for Assigned Name jy
s
that there are websites (Inte rnet Corp orat
called ICANN is a company which
responsible organization regisbars. A register
dits domain name
Numbers) ICANN accre .
in name are registered active and passive.
also a website where doma + Hubs: There are two types of hubs:
r electrically and are usually not
Q. 8. (b) Sate llit e netw orks.
ications for today's adv
anced satellite + Passive hubs simply connect all ports togethe
one of the h ottest appl
Ans. Data networking is powered.
st
main markets: users an dclean up the signal before it is broadca
services. There are two e & Government * Active hubs use electronics to amplify
consisting of Corporat
e Commercial-mostly homes and smal busi l nesses to the other ports.
isting ofservices to computers together.
Consumer-mostly cons the use of small satel lite dishey * Hubs are devices used to link several
rapidly and both rely on * They repeat any signal that comes in on
one port and copy it to the other ports (a
Both markets are growing calle d “very small aper
ture
meter in diameter,
often less than one process that is also called broadcasting).
at the receive site,
terminals” or VSAT's. 1980s when large * It works in physical Layer of OSI Model.
| data market since the
the co’ mmercia sactions at retail busin
ess
Satellites have served "pol nt of sale” tran * Bridges
e deployed for umer market. Analyst
numbers 0 f VSATs wer to serve the cons
> They join similar topologies ard are used to
divide network segments.
s more rece! ntly begun and another ong
locations. Satellite have are about two million commercial VSATs address .
estimate that toda
y there
ou n d the world . Most of these are used for > It can filter traffic on the basis of MAC
er services deployed ar > If itis aware of the destination address, it is able to forwar
d packets; otherwise
million for consum
toc ol (IP) broadband services. e a bridge will forward the packets to all segments. They
are more intelligent than
Int ern et Pro
isprimarily due to the improved pric
ks simultaneously.
The rapid growth
of satellite broadband
operators like Te repeaters but are unable to move data across multiple networ
based solutions. Experienced
performance of today’s satellite- of satel lite VSAT sites into a comm > Unlike repeaters, bridges can filter out noise.
rate any number strong service le
can now seamlessly integ ty standards backed by > Works in Data link Layer of OSI Model.
3
ork with high quali host of ?
customer's existing netw quality experience for a
remote users to have a high > Multiple collision Domain but single Broadcast Domain.
agreements. This enables ;
> The main disadv4 yantage to bridges4 is that they can’t connect dissimi ssimilar network
applications. netw orki ng:
Other key reasons for the rapid
growth of VSAT types or perform intelligent path selection. For that function, you would need a router.
nal and/or local reach * Switches:
¢ Satellites offer global, regio C (single
¢ Satellites offer an increa: sing
range of solutions, from traditional SCP i > A network switch is a computer networking device that connects network
ased VoIP and data netw orki ng
channel per carrier) services to IP-b
segments.
vity to regions with little e2 cisting
Satellites offer direct, last mile connecti > Populates MAC address table on the basis of source MAC address of a Frame.
at a fract ion of the cost of insta lling and maintaining terrestrial solutio > Network switches are capable of inspecting data frames as they are received
infrastructure,
ly, sometimes within hours,
e Satellite networks can be implemented quick on Ee the so urce and destination
determining inati device
i of that frame, and forwarding it
es, prov idin g “inst ant infr astr uctu re”
disaster strik
Q.8. (c) Network devices. > Data Link Layer or Layer 2 Device. f
- Ans. Repeater ; > Multiple collision Domain but single Broadcast Domain.
ures. It also
i ee on a device similar to the Hub, but has additional featification of >A vital difference betweena hub and a switch is that all the nodes connected
ivi
in plac es wher e ampl h
to a hub share the bandwidtamong themse!
a ical layer. The repeaters are used
Sees seth oal alate
signal is necessary. But, the kind of amplification done by the repeater is
di
different!
port has the full bandwidth all to itecif
the regular amplification by amplifiers.
 Networks
-2016 Sixth Semester, Computer
on a 10-Mbps
usingusit a hub sea network,
bn eke
i xample, if 10 nodes are communicating EXAMINATION [FEB. 2017]
FIRST TERM
kam ake ut get a portion of the 10 Mbps if other nodes 0 the hub wana
ene eco each node could possibly communicate at the full 10 Mbps. SIXTH SEMESTER [B.TE CH]

. COMPUTER NETWORK [ETCS-306]

= e Hrs.
: 1,30
Tim M.M.:30
switch _\server Note: Q.No. 1 is compulsory and attempt any two more questions from the remaining-
—— =
Raw Q.1. (2) Explain the difference between an Internet draft and a proposed
Saget standard. (2)
Ans. Internet draft: Internet Draft is a series of working documents published by
the IETF. Typically, they are drafts for RFCs, but may be other works in progress not
intended for publication as RFCs. During the development of a specification, draft
versions of the document are made available for informal review and comment by
placing them in the IETF’s Internet-Drafts directory. This makes an evolving working
document readily available to a wide audience, facilitating the process of review and
‘Tevision. Internet-Drafts have no formal status, and are subject to change or removal at
Jany time; therefore they.should not be cited or quoted in any formal document.
Proposed Standard: The entry-level maturity for the standards track is “Proposed
‘Standard’. A Specific action by the IESG is required to move a specification onto the
‘standards track at the “Proposed Standard” level.
A Proposed Standard specification is generally stable, has resolved known design
choices, is believed to be well-understood, has received significant community
review,
‘and appears to enjoy enough community interest to be considered valuable.
ks eee aad Perea nor operational experience is required for the
; esignation of a specification as a Proposed Standard. However, such experience
is
gent devices that connect multiple network types highly desirable, and will usually represent a strong argument in favour of : Proposed
highly intelli
Router: : Routers are ding data. Bender designation.
and determine the best path ead one LAN to another. o _ Q.1.(b) In the ring topology what happens if one of the station isunplugged?
Routers are normally es ackets to determine the destination address of ; (2)
These devices examine eal routing table to choose the best path for the Ans. In a ring topology, unplugging one station interrupts the ring. However, most
the data. It then examines its ‘ switches them t0 the proper outgoing port. : ring networks use a mechanism that bypasses the station; the ring can contiite its
packet through the network, ee bridges because they are more intelligent devices; as pperation.
pen causing packet-forwarding delays. Because sth
such,oe
they analyze ev 2 oe expensive.
intelligence, theyare a
layer 3 devices.
Routers are OSI netwo! rk ast two routers used.
is set up, there will be at le
Typically, when a WAN
Repeater . Resear
Repeater
Repeater Repeater Repeater

Piil ilod
Q.1.(c coer
) Expla cra
in why te ion
collis is ani issue in random access
protocols but a
Ans. In random access methods i" there
is no access control (as there is j
88methods) and there 18 no predefined
channel (as in cannntteation Bat
transmit when it desires. This libert ten
y may create collision.
=
 uter Network -
LP. University-[B.Tech.]-AB Publisher 2017-3
Sixth Semester, Comp’ ee find which station

2-2017 tations consult one ac aA authorized by othe In burst error, it is not necessary that only consecutive bits are changed. The length
bit. As shown in fig.
d unless it has Pe of burst error is measured from first changed bit to last changed
con! trolled access
In methods, the 8 ot sen
aig i0 controle length of burst error is 8, although some bits are unchanged in between. Burst error is
aA station cann
has the right to sence: ae control (a8 th Gen) Each stati, ™most likely to occur in a serial transmission.
Q.1.(e) Describe the need for switching. (2)
ee
In aes
access methods, there iB Nanaale (asin channeliza'Jongs
snd there ig no predefine to the statio, Ans. Switching is process to forward packets coming in from one port to a port
access metho hen it desires: hole available bandwidth belo’ Weston methoj leading towards the destination. When data comes on a port it is called ingress, and
an transmitit W F he WhO . In a channe when data leaves a port or goes out it is called egress. A communication system may
needs to wait. does not hay
: . the other stations
In a random access« method ons. If a statio include number of switches and nodes. At broad level, switching can be divided into two
the statio’
. the © de +d between . Q major categories:
that wins: the contention;
dwidth is div? ains idle.
the available banav loca’ ted channelA ae rv diff sffer from a burst error? ¢ Connectionless: The data is forwarded on behalf of forwarding tables. No

a single pt ee one bit of data unit is changed from 1 ie on previous handshaking is required and acknowledgements are optional.
data to send, ore ee
QO ae error: Itmeans ony ¢ Connection Oriented: Before switching data to be forwarded to destination,
Ans. Single b¥ in fig. _ there is a need to pre-establish circuit along the path between both endpoints. Data is
from 0 to 1 as shown ' then forwarded on that circuit. After the transfer is completed, circuits can be kept for
future use or can be turned down immediately.
; Q.2.(a) Discuss the functions of all layers of OSI model. (4)
j Ans. The Open Systems Interconnect (OSI) model has seven layers. This article
_ describes and explains them, beginning with the ‘lowest’ in the hierarchy (the physical)
; and proceeding to the ‘highest’ (the application). The layers are stacked this way:
¢ Application
¢ Presentation
Data received - © Session
: ¢ Transport
: ° Network
error } ‘Data Link
Single Bit
tr an s mi s e Physical
in parallel
can happen rors are Physical Layer: The physical layer, the lowest layer of the OSI model, is concerned
;
Single bi it error parate wires.
a Single bit er
te d us in g se to Off with the transmission and reception of the unstructured raw bit stream over a physical
transmit ged from 1
ta unit are chan medium. It describes the electrical/optical, mechanical, and functional interfaces to the
re bits in dai
e 3 me ans two or mo physical medium, and carries the signals for all of the higher layers. It provides:
REBuBr! e
n in fig. ¢ Data encoding: modifies the simple digital signal pattern (1s and 0s) used by the
0 to 1 as show
PC to better accommodate the characteristics of the physical medium, and to aid in bit
and frame synchronization. It determines:
— What signal state represents a binary 1 -
— How the receiving station knows when a “bit-time” starts
— How the receiving station delimits a frame :
e Physical medium-attachment, accommodating various possibilities in the
edium: i :
— Will an external transceiver (MAU) be used to connect to the medium?
— How many pins do the connectors have and what is each pin used for?
e Transmission technique: determines whether the encoded bits will be transmitted
y baseband (digital) or broadband (analog) signaling.
¢ Physical medium transmission: transmits bits as electrical or optical signals
Length of burst error = 8 ppropriate for the physical medium, and determines: }
Burst error
 4-2017 Sixth Semester, Computer Network :
2017-5
— What physical medium options can be used LP. University-[B.Tech.}-AB P ublisher
é a ae ae volts/db should be used to represent a given signal state, using a giy Transport Layer: The transport layer ensures that meee are delivered error,
g a layer protocols
free in sequence, and with no losses or duplications. It relieves ie
payerca menu transfer of data between them and their pe re ee
Data Link Layer: The data link layer provides error-free transfer of data fran, from any concern with the
from one node to another over the physical layer, allowing layers above it to a8sun, The size and complexity ofa transport protocol serene OO res Z ee ability, a
virtually error-free transmission over the link. To do this, the data link layer provide, get from the network pata. For Z aS arreliable
yer provide; ithe network ee
layer is with v1liable and/or
network layer only supportss
* Link establishment and termination: li i ical |i minimal transport laye' : detectionae ‘and recovery. .
exmretion:ontablities a tenninaiee oo a Fyeicrams, thet Latiaeed photaied should include extensive error
between two nodes.ane Se
* Frame traffic control: tells the transmitting node to “back-off when no fran The transpo rt layer provides: e (session) layer above it,
buffers are available = ; py Message sogmentatiaes
into units (if Sra
smallerSr not alreadyoosmallaf enare anny passes the
pee :
pis iy ie network layer. The transport layer at the destination station
transmits/receives frames sequentially. Se iicc units
* Frame sequencing:
e Frame acknowledgment: provides/expects frame acknowledgments. Detects anireassembles the message. : :

recovers from errors that occur in the physical layer by retransmitting non-acknowledgg —. Message acknowledgment: provides reliable end-to-end message delivery with
frames and handling duplicate frame receipt. acknowledgments.
* Frame delimiting: creates and recognizes frame boundaries. ° Message traffic control: tells the transmitting station to “back-off ” when no
e Frame error checking: checks received frames for integrity. message buffers are available.
onto one
to use ¢ Session multiplexing: multiplexes several message streams, or sessions
¢ Media access management: determines when the node “has the right” belong to which sessions (see session
logical link and keeps track of which messages
the physical medium.
., layer).
ion of the subnet, decid;
eee Aa See ark endl conditions, priority of servi,a Typically, the transport layer can accept relatively large messages, but there are
which physical pa’ a: strict message size limits imposed by the network (or lower) layer. Consequently, the
*
and other factors. Jt provides: transport layer must break up the messages into smaller units, or frames, prepending
¢ Routing: routes frames among networks. 3 a header to each frame.
© Subnet traffic control: routers (network layer intermediate sy! eae si instru © Session Layer: The session layer allows session establishment between processes .
asending station to “throttle back” its frame transmission when the router's uffer fill running on different stations. It provides:
up. 2 —_—° Session establishment, maintenance and termination: allows two application
° Frame fragmentation: if it determines that a downstream router's maximuyprocesses on different machines to establish, use and terminate a connection, called a
transmission unit (MTU) size is less than the frframe
am size, aa router can fragment a framsession. these eee eee
destinat ion station . ecealan eappart pastes tis funetien tee lies
for transmi ssion and re-asse mbly at the
ity, amon are oestune
es, or names, injover the network, performing s
* Logical-physical” address mapping: translates logical address

ee Presentation Layer: The presentation layer formats the data

viewed as the eines for the netw ee
* Subnet usage accounting: has accounting functions to keep track, of framthe application layer. It can be
forwarded by subnet intermediate systems, to produce billing information. Eeanslate data from a format used by the application layer into a common format at the
Communications Subnet: The network layer software must build headers so oe station, then translate the common format to a format known to the application
recogni yer at the receiving station.
the network layer software residing in the subnet intermediate systems can
them and use them to route data to the destination address. The presentation layer provides:
¢ Character code translation: for example, ASCII to EBCDIC
This layer relieves the upper layers of the need to know anything about the dat
transmission and intermediate switching technologies used to connect systems.| ° Data conversion: bit order, CR-CR/LF. integer-floating point, and s
: > 0 on.
establishes, maintains and terminates connections across i
communications facility (one or several intermediate systems in the a. number of bits that need to be transmitted on the
subnet), I ° Data
7 encryption: : en encrypt data for i
In the network layer and the layers below, peer protocols exist between a nodearncryption. security purposes. For example, password
its immediate any
neighbour, 9 but the neighbor may be a node through which dataisroutdi Application La: yer: The a swage
~ not the destination station. The source and destination stations may be separatedapplication processes to Sahel eneg aye Serves as the window for users and
many intermediate systems. mmonly needed functions: ork Services. This layer contains a variety of
* Resource sharing and
device redirection
 2017-7
Publisher
Sixth Semester, Comput
er Network LP. University-{B. Tech. |-AB
ons
6-2017
e stand of glass fiber with
a diameter of 8.3 to 10 micr
Single Mode cable is a singl
Fibe r with a relat ively narrow diameter,
e Remote file acces Single Mode
that has one mode of transmission. 1550 nm. Carries higher
e Remote printer access will propagate typically 1310 or
through which only one mode sou rce with a narrow spectral
tion but requires a light
¢ Inter-process communica bandwidth than multimode fiber, -mod e optic a | fiber, single-mode optical
also called as mono
e Net wor k man agm ent width. Single-mode fiber is
e Directory services waveguide, unimode fiber.
as mail)
« Electronic messaging (such v: arious propagation
modes
ission medium? Discuss (4)
Q.2. (b) What is transm
in the fiber optics.
Ans. The path through
called channel. It is also
which data i s transmitte
known as communicatio
d from one plac e
n mde dia or transmission
to another in
media.
- TU Source
NV
s of transmittion media . Destination
These are different type
(i) Guid ed medi a
(ii) Unguided media > | Q.2.(c) Compare the throughp ut of a pure ALOHA network with a slotted
other by (2)
Guided media directly connected with each ALOHA network.
ication device @ re
In guided media commun call ed bounded media.
a like wire s. It also - Ans. Throughput of pure ALOHA channel
using some physical medi
system during one frame
Unguided Medi Let G be the average number of frames generated by the
or spa that the average no of successful transmissions
each other through air _ transmission time. Then it can be proved
media is us for pure ALOHA is S=Gxe6, The maximum throughput S,,,,, is
0.184, for G= 1/2. i.e. if
infrared signals unguided d using media i
:
ligh t alon g opti cal percent
be transfer all over
the worl I agating
modes for prop | Scale a frame isgenerated during one one frame transmission time, then 18.4
There are two different ic types0: f fiber: multimode frames reach their destination successfully. This is an expected result because
:
Propagation Mode: single mode. There are two basi F
and the vulnerable time is 2 times the frame transmission time. Therefore if a station
channels: multimode _
fiber
fiber and single-mode generates only one frame in this vulnerable time, the frame will reach its destination

|
i
; successfully.
Propagation Mode The throughput for pure ALOHA is S = G x e?¢
The maximum throughput S,,,,, = 0.184 when G = (1/2)
Throughput of slotted ALOHA channel
The average number of successful transmissions for slotted ALOHA is S = G x eS
_ The maximum n throughput S ‘max 18 0.368, when G = 1. In other words, i " is
Single Mode
Steed during one frame transmission time, then 36.8 percent ofthee ae oes
ee r A iee successfully, This result can be expected because the vul vabl eae

beca u! se mult iple beams from a light sourd FE a a the frame transmission time. Therefore, if a station gene’ ise arc
ode is so named e of glass fibers, with e in this vulnerable-time, the frame will reach its Ae cA pret
Multimode: Multim Mul tim ode cabl e is mad y
core in different paths. y component (the mo
move through the ran ge for the ligh t carr The throughput for slotted ALOHA is S = G x e-&
1n the 50-100 micron
common diameters
62. 5). distance ' The maximum throughput S,,,, = 0.868 when G = 1
com mon size is high speeds 0 ver medium
you high bandwidth at oug!
Multimode fiber gives ero us pat hs, or mod es, as they travel thr Q.3.(a) Why channel allocation is a difficult task? Explain the random access
persed into num 0
fiber core diameters are method that tries to avoid collision.
_ Light waves are dis 0 nm. Typical multimode
cable’s core typ ica lly 850 or 130 (5)
rs.
62.5, and 100 micromete fiber and a high ly focu sed sour ce 0 Ans. Refer Q.3 End term june — 2016.
e uses step -index
Single Mode: Single mod e to the horizontal. Random Access, which is to i
l range of angles, all clos , Which is to issue a completely random time, relies on the Aloha
that limits becams to smal ethod

ee
In this me tho d, h n a coupler
when has informati 0) mn. to transmit, i
t
sends it without
{ a about other users, If there is a collision, that is to say su; iti
perposition of two signal
‘or more users, the signals become indecipher able
and are lost. They are Aibeeqeenits
Source
Destination »
 i

fe 2017-9
isher
Sixth Semester, Computer Netw
ork LP. University-(B.Tech.-AB Publ
8-2017 : ne
Couple, 10001100
ers 1, 2 and 3 collide. The
Figure, in which the coupl module 2 cal
transmitted, as shown in the smallest timer. Then, the
se he shot
transmits its field first becau random time of retransmigg; 1000
the coupler 1. Both derive a roof1oo01tto
and its signals collide with 2 are silent, so that the frame
of
jroo!
while the couplers 1 and m cee,
The coupler 3 is listening the origin of all the rando
Technical aloha is 000!
coupler 3 passes successfully. 0000
methods. i piaaa
s
e,
when the number ofcoupl 0000
system becomes very small
The flow rate of such a ifthe number of stations goes to
infinity — 2
mathe matic ally that
increases. It can be shown To reduce th
a certai n momen t, the system is more stable. ¢ i ‘ 0
the flow becomes zero. From ts of this technique have
bee, aoe
en users, various improvemen
likelihood of conflict betwe ;
(ior!
100)
proposed.
1000
er cd 1001
Eze =----~7 She ieee
Sg 0010
Coupler 1 0000
4
0100
0000

Coupler 2 100 (remainder)

y
(modulo 2
10011101000-100=10011101100
= -7 — ene 4
Gh ------------
Actual frame transmitted:
Ben

- time
Coupler 3 subtraction) is received as
the left is garbled and the frame
e Now suppose the third bit from by the polynomial generator we geta remainder of
a
Operating Principle of Pure Aloh 0. Hence on dividing this
1011110110 error free we
received frame been
error has occured. Had the
4 * 100 which shows that an below .
d inder of zero. See
Slotted aloha, aloha or slice to would have got a rema
time slots, or slots, and
was to cut the time into on time of a frame
Improved technical aloha first, the trans missi 10101000
of frames that slice collision ifa single frame
authorize the issuance way, there is no
111to1000
of time. In this tool1o
requiring exactly a slice start transmittiInng thein oie
d at the begin ning of slice. Howev er, if several frames along the slot.
transmitte are superimposed Bae
the frames emissions | 1
the beginning of slice, a random time. i
case, there has retransmission after 0000
latter remai n
the start-up perio d but eeka
the throughput during of the devices,
This method improves cost from a complication 1001
le. In additi on, there is an additional
unstab
be synchronized.
since all emissions must 0100
dard CRC method.
) Abit strea m 10011 101 is transmitted using the stan mitte d,
0000
Q.3.(b bit string being trans
is 3+ 1, Show the actual
The generator polynomial this error is detected 1001
1001
g transmission. Show how (5)
Suppose a bit is inverted durin
;
at the receivers end. 0001
as 1001. Because the generator polynomial
Ans. Our generator G(x) = x? + 1encoded the lower end of the frame to be
0000
d three zeros to
is of the degree three we appen 101000. On
“001
the 3 zeros the bit stream is 10011 000 °
transmitted. Hence after appending
ding three zeros to the frame we geta
dividing the message by generator after appen
after of the remainder from the
remainder of 100. We do modulo 2 subtraction there
0100
1 frame transmitted is 100111011 0000
stream with the three zeros appended. The actua
See below. 1 0 0 (remainder indicating error)
 2017-11
k
Compu ter Networ .Tech.}-AB Pub
lisher
Sixth Semester, LP. U!niversity-{B ociated
ass
10-2017 data link layey s uence number
issues involved in th e The last three bits
he seq
are the
SS-Bits of the
various desi, gn (4) Poll/Final bit. indicated d in the
Q.4.(a) What are The fifth bit is REJ and SREJ ervisory
pro toc ol. states RR, RNR, either on a sup
Explain sliding
window
functions to carry
out. These with the supervisory (RR) can : be sent han d RNR, REJ
of spe cifi c an ac Jcnowledgement the oth er
layer has 8 number k layer, control field. Note
that ned earlier. On
Ans. The data link erface to the networ as mentio
:
viding a wel l-defined service int frames, dealing with frame or piggyback
e don an I-Frame frame.
functions include pro h a supervisory
layer are grouped into only thr oug
bits of the physical that slow receiver
s are not and SREJ are sent
determining how the ulating the flow
or fra mes SO ram e
rs and reg Control field of S-f
transmission erro
ers. ree
swapped by fast send 6 7 8
3 4 5

Eee
1 2

eee
g issues.
There are the followin
(a) Services Pro vid ed to the Netowrk Lay
er
sal eot
>! ~E ~<----- CON
ADDRESS- >:
TROL FIELD - - --
- >> 1 -Fcs
(b) Framing READY (RR)
00 RECEIVE (RES ) > 01 REJECT
(c) Error Control oy DY (RNR)
: 10 RECEIVE NOT REA (SREJ)
(d) Flow Control > 11 SELECTIVE REJECT
NUM BER
Examination 2016. ACKNOWLEDGMENT
Refer Q.3.(a) of First Term
N(R) :
(3) > POLL/FIN AL BIT
me types in HDLC. PIF
various fra
Q.4.(b). State and explain
are thr ee typ es of HDLC frames:
Ans. There Unnumbered Frame (U-Frame) for link
frame (I-Frame) mes. They are used
Information transfer have data field as S-Fra Controls field of
The U-Frames do not and other control functions.
rame) mode settin g
e Supervisory frame (S-F establishment, termination,
e Unnumbere d fram e (U-F rame) an unnumbered frame is shown. fy an unnumbered frame.
l field are 11 which identi
Frame (-FRAME) ledgement of the The first two bits of the contro er bits. They specify the
Information Transfer also carries acknow The rest five bits are called modifi
for tra nsp orting user data. It as shown in Fig.. The first bit is 0 The fifth bit is Poll/Final bit. ;
I-Frame is used I-F ram e is
control field of the are the sequence
number control function.
received frames. The The next three pits
frame as an I-Fra me. Cntrol Field of a U-Frame
which identifies the
N(S) of the frame. 5 6 7 8

|
3 4

Eber
(P/F) bit. 1 2
The fifth is Poll/Final gement (RR) Seven tates
N(R) of the acknowled
bits are the sequ ence number
The last three
on the I-Frame- \ << -FCS
which is piggy packed
ADDRESS- =>) & --~ >> tS
Control field of I-frame Peas MMMMM __ : MODIFIER BITS
ae 450 O20. G : POLUFINAL BIT
ago We P/F

0 No) PIF ) NR}

by a receiver i

|
4 1
If the 7-bit Hamming i
code word i is received
Q.4.(c)
=> 1€ - DATA
Se aeS ~ = ~~ - CONTROL FIELD = © ==
See ack
the even parity state, find whether the received code
ADDRES peo
ats:
N(S) FRAME SEQU ENCE NUMBER wrong. If wrong, locate the error and correct the code word.
N(R) ACKNOWLEDGMENT NUMBER - <
P/F POLLIFINAL BIT Ans,
d6 a5 na ae
P
d7 ;
) 0
Supervisory Frame (S-FRAME 1 0 1 1
1
wledgements,
and is used to carry only ackno 5
S-Frame does not have data field
requests forretransmission, etc. It is identi fied by the first two bits of the control fiel P4 =d5 d6 d7
ate four
eee oe ee a 10 in anPee The next two bits (SS) are used to indic p4isli
j = 1=> 101
Ready (RR), Receive N
pervisory states. Receive
Selective Reject (SREJ).
Sa athe rahi ipa ee a
P2 = p2 d3 dé d7
 k
Sixth Semester, Computer Networ
12-2017
RM EX AM IN AT IO N [MAY-JUNE 2017]
= 0=>00 1 p2=0 END TE [B.TECH]
SIXTH SEMESTER
Pi = pl d3 p5 d7 RK [ETCS-306]
1=> 011 COMPUTER N ETWO : 75
M.M.
Pl=1. Time : 3 Hrs.
rest.
mpt any five . Questions from the
P4 p2 pl=101=5
Note: Q. No. 1is compulsory. Atte
(3)
different from congesitio
Error at 5 bit. n control?
the 5'® bit will be : 1001011 Q.1. (a) How flow control is
Corrected answer after changing that the sink
the source from sending data
Ans. Flow control means preventing is fairl y easy with a sliding
. This
of broad band coaxial cable. will end up dropping because it runs out of b
uffer space
Q.4.(d) List the characteristics ’s Wi ndow is no large r than the free space in
media that carnes the signal of
guided
higher frequency window protocol—just make sure the source its free buffe r spac e in the
Ans. Coaxial cables are the g the sink advert ise
. Coaxial cables are also calle
d coax. (short form), the sink’s buffer. TCP does this by lettin
ared to twis ted pair cable
range comp cable and 75 ohm cable. 50 ohm window field of the acks.
widely used: 50 ohm
Two types of coaxial cables are for analog transmission. vent) the source from sending
sion and 75 ohm cable is used Congestion control means preventing (or trying to-pre
cable is used for digital transmis noise immu nity. It has a large becau se its queue is full. This is more
this cable has excellent data that will end up getting dropped by a router
Due to the shield provided, are often installed ling different paths can
Co axial cables are easy to
install. They complica ted, because packets from different sources travel
bandwidth and low losses. star (ARC net).
daisy chain (Ethernet) or a converge on the same queue.
either in a device to device r shields
l cabl es in a pro tective cover. The cove Explain with example.
Acoaxial cable consists of man y smal
trom agne tic interference. Within — Q.1.(b) Difference between bit rate and baud rate?
ers as well as from elec
the cable from physical dang with one another. Coaxial (3)
are shielded from interference
the cover, the various cables orks that requ ire many simultaneous |
ication netw Bit rates
cables are used in commun than 5000 links. It has a- Ans. Data rate is the capacity of a channel in bits per second.
coaxial cable can provide more 1?s) transmit ted in one second ina
communication links. Each ease in diam eter of the inne r measure the number of data bits (that is 0?s and
can be increased with the incr s © Seas channel. A figure of 2400 bits per second means 2400 zeros or ones can
data rate of 10 Mbps which of node s on a thin net segment is 30 node
numb er
conductor. The specified max
imu m Ee ees in one second, hence the abbreviation “bps.” Individual characters (for
net it is 100 node s. p letter: s or numbers) that are also referred to as bytes are composed of several
and on a thick amplifiers which are. a ample
and is used for large 2rea, i t requires for
Since bro adb
bles run together, one used
systems two identical ca
unidirectional. In dual band n for inbound and Baud
; rate represents the number of times the e line li i
condition (i.e., frequen:
data. Different bandwidths are give
outgoing data, one for incoming und and 40- 300 MHz for outbound, amplitude or phase) changes each second. Data rate = n bits x hainiieace =
, 5-30MH z for inbo
outbound cables. Eg: for 300MHz 10 BASE 2 > eo to “ number of bits one signal represents. A baud rate is the went of
bus LANs namelv 10 BASE 5 and
There are two systems that use the signal in a communications channel changes state i
2400 baud rate means that the channel can change as states up Se non ti renoesm were
to 2400
The term “change state” means that i it can change from 0 to 1‘ or fro: Siues per sowpad:i
this case, , 2400) times per second. : It alscso refe Se ee oe;
such as voltage, frequency, or phase level. Se ee ee coe

Q.1.(c) What is difference between Classfull addressing and Classl ess IP

addressing? (4)
Ans. Classfull Addseeaing:
IP addresses have a network and host portion. In classful addressin g,
athe network portion ends on onee 0 f the separati i
parating dots in
Ns Mi eieer euanae Meena ae an_IP address into the ene ener a
¢ i In the c assful addressin, Sab erste
available are divided into the five classes A,B ,C,
Cin maaan:
: cotA,Baii
ch class and C
address are frequent} y used because cl i 8 for Multicast and is rarely used and
class E is reserved and is not currently nL :
 2-2017 Sixth Semester, Computer Network
a
In controlled access methods, the stations consult one another to
has the right to send. A station cannot send unless it has been g Which, LP. University-{B.Tech.}-AB Publisher 2017-3
stations. rized by
In burst error, it is not necessary that only consecutive bits are changed. The length
In random access methods, there is no access control (as thers ei of burst error is measured from first changed bit to last changed bit. As shown in fig.
access methods) and there is no predefined channels (as in channelization) Bact length of burst error is 8, although some bits are unchanged in between. Burst error is
can transmit when it desires. most likely to occur in a serial transmission.
In a random access method, the whole available bandwidth bel Q.1.(e) Describe the need for switching. (2)
that wins the contention; the other stations needs to wait. Inachannelizata” Ans. Switching is process to forward packets coming in from one port to a port
the available bandwidth is divided between the stations. If q station as leading towards the destination. When data comes on a port it is called ingress, and
data to send, the allocated channel remains idle Rot hy When data leaves a port or goes out it is called egress. A communication system may
Q.1.(d) How does a single bit error differ from aburst errory include number of switches and nodes. At broad level, switching can be divided into two
Major categories:
Ans. Single bit error: It means only one bit of data unit is chan :
from itp * Connectionless: The data is forwarded on behalf of forwarding tables. No
0 to 1 as shown in fig.
from
previous handshaking is required and acknowledgements are optional.
ae PRS Pan ERE er * Connection Oriented: Before switching data to be forwarded to destination,
0 9 1 0 6 4 Dace there is a need to pre-establish circuit along the path between both endpoints. Data is
' Sene then forwarded on that circuit. After the transfer is completed, circuits can be kept for
| l J future use or can be turned down immediately.
Q.2.(2) Discuss the functions of all layers of OSI model. (4)
c prs Ans. The Open Systems Interconnect (OSI) model has seven layers. This article
? describes and explains them, beginning with the ‘lowest’ in the hierarchy (the physical)
and proceeding to the ‘highest’ (the application). The layers are stacked this way:
aaron . — a * Application

Sea a Me] e © | cumcccchal : amare

Ae HS ek ee 3 * Traneport
Single Bit error . ; mee :
Single bit error can happen in parallel tranemiseion where all the date z "
transmitted using separate wires. Single bit errors are the least likely type
of * Physica!
serial tranamiasion, Physical Layer: The physical layer, the lowest layer of the OSI model, is concerned
Burst EB It t ; changed from with the transmission and reception of the unstructured raw bit stream over a physical
0 to 1 :as shown
rnaeh Tt AdDs
in fig, Sao eemnaee heads Gone wean i medium. [t describes the electrical/optical, mechanical, and functional interfaces to the
physical medium, and carries the signals for all of the higher layers. It provides:
* Data encoding: modifies the simple digital signal pattern (1s and 0s) used by the
to better accommodate the characteristics of the physical medium, and to aid in bit
frame synchronization. It determines:
— What signal state represents a binary 1
— How the receiving station knows when a “bit-time” starts
Bits changed
by Gurst error — How the receiving station delimits a frame
* Physical medium attachment, accommodating various possibilities in the

— Will an external transceiver GAAL) ba eundincennaed tatiana

— How many pins do the connectors have and what is each pin used for?
_ * Transmission technique: determines whether the encoded bits will be transmitted
baseband (digital) or broadband (analog) signaling.
Length
of burst error = § << Physical medium transmission: transmits bits as electrical or optical signals
Burst error
1 iate for the physical medium, and determines: toe
 ——=
—————

Computer Network
Sixth Semester,
4-2017 2017-5
. }-AB Publisher
] medium options can be
used LE. University-[B.Tech
— What physi resent a i given signa l ; es are delivered erro
r,
db should i be use d to re gnal state, USing g gi r ensures ; tha t messag
transpo rt layea layer protocols
— How many volts/ois P
x Transport Layer: The ns. t reli eves the hig her
and with no losses or duplicatio .
physical medium i between them and their peers
ides error-free t
of data fran, ey snscora wit te transfer of data of service it can
Data Link Layer: ee Airis ea col depends on the type
one nod e to ano therosover t enti
this, the data link iae‘ The size and complexity ofa tr ansport proto layer with virtual circu
ircuiit cap: ability, a
from
nsmission over the link. To do ec ork
For a reliable netw
virtually error-free tra rth ney : d , a Provides get cs the network yee d/or r only supports
18 unreliable and/o
t layer is required. If the network layer n ‘and recovery.
include extensive error detectio
t and termina’ logicalliy mini mal tran spor
e Link establishmen Hon: ota oe e
e col should
nod es. i datagrams, the transport proto
between two ue
:
mitting node to "back-of” when ng fray The transport layer provides: the (session) layer above it,
e Frame traffic contro |: tells the trans * Message segmentation: accepts a message from enough), and passes the
buffers are available. er units (if not already small
smits/receives frames sequential
ly, splits the message into small at the destination station
netwo rk layer. The transport layer
° Frame sequencing: tran
frame acknow ledgme nts, Detects an aie down to the
smaller units
e Frame acknowledgment: provides/exp ects : the message.
d
non-acknowledge reassembles delivery with
the physical layer by retransmitting des re! liable end-to-end message
recovers from errors that occur in ( ¢ Message acknowledgment: provi
dupli cate frame receip t.
frames and handling acknowledgments.
frame boundaries. to “back-off ” when no
e Frame delimiting: creates and recognizes ol: tells the transmitting station
contrble
: .
rity. i ° Mess
butte ard-ac valla
agers traffi
5 s recei. ved frame s for integ
Frameerror checking: check
l m essage streams, Or sessions onto one
e Session multiplexing: multiplexes severa
Media access management: determines when the node “has the right” to use be long to which sessions (see session
logical link and keeps track of which messages
the physical medium.
ion of the subnet, decid; layer). relatively large messages, but there are
Network Layer: The network layer controls the operat Typically, the transport layer can accept
on network conditions, priority of seryj rk (or lower) layer. Consequently, the
which physical path the data should take based & strict messag e size limits impose d by the netwo
other factors . Jt provid es: break up the messa ges into smaller units, or frames, prepending
and : transport Galayer
Bree cack: must
frarne
ks.
e Routing: routes frames among networ ses .
Session Layer: The session layer allows session establishment between proces
¢ Subnet traffic control: routers (network layer intermediate systems) can instru
bufferfill running on different stations. It provides:
a sending station to “throttle back” its frame transmission when the router’s allows two application
up. e Session establishment, maintenance and termination:
on diffe machines to establish, use and terminate a connection, called a
¢ Frame fragmentation: if it determines that a downstream router’s maximu processes rent
can fragment a framsession.
transmission unit (MTU) size is less than the frame size, a router
¢ Session support: performs the functions that allow these processes to communicate
for transmission and re-assembly at the destination station.
security, name recognition, logging, and so on.
. ages address mapping: translates logical addresses, or names, inover the network, performing
3 Presentati on Layer: The presentation layer formats the data to be presented to
physical addresses.
layer. It can be viewed as the translator for the network. This layer may
4 Shine usage accounting: has accounting functions to keep track, of framethe application
forwarded by subnet intermediate systems, to produce billing information. ‘translate data from a format used by the application layer into a common format at the
Cpeneeations S &:"The network layer software must build headers 60 nding station, then translate the common format to a format known to the application
eo enc
the network layer software residing in the subnet intermediate systems can recognit
them and use them to route data to the destination address. The presentation layer provides:
Seen eee eee ee ee eee eG
This layer relieves the upper layers of the need to know anything about the da
ord
bit order,
° Data conversion: : bit i
CR-CR/LF, integer-floa ti: i i
intermediate switching technologies used to connect s ystems, {|
transmission and inte ser
fe that need e
to be transmitte d on the
establishes, maintains and terminates connections across the intervenil| ° Datacompression: : reduces reduces th the number of bits
iene facility (one or several intermediate systems in the communicati poner:
* Data encryption: encrypt data for security purposes. For example, password
subnet). , i i |
eh In thenetwork layer and the layers below, peer protocols exist between a node a epcrvedon, :
its immediate neighbour, but the neighbor may be a node through which data is rou Application Layer: The application layer serves as the wind
ted fapplication procegses to access network services. . This layer eae vad 0
’ not the destination station. The source and destination stations may be separa’
mmonly needed functions:
many intermediate systems.
_ © Resource sharing and device redirection
 6-2017 Sixth Semester, Computer Network
° Remote file acces _
e Remote printer access
: Lee University-(B.Tech. -AB Publisher
2017-7
¢ Inter-process communication
Single Mode cable is a single stand of glass fiber with a diameter
¢ Network managment of 8.3 to 10 microns
_ that has one mode of transmission. Single Mode Fiber
with a relatively narrow diameter,
* Directory services
. © Electronic messaging (such as mail) through which only one mode will propagate typically 1310 or 1550 nm. Carries higher
bandwidth than multimode fiber, but requires a light source with a
Q.2. (b) What is transmission medium? Discuss various peeiaad narrow spectral
= Berens eri is also called as mono-mode optical fiber, single-mode optical
in the fiber optics. Pagation Mode, guide, unimode fiber.
Ans. The path through which data is transmitted from one ace 4)
called channel. It is also known as communication mdedia or
transmios nother in
These are different types of transmittion media 10n m .
(i) Guided media
a | L LU NK,
(ii) Unguided media ; Source
Guided media OM
: Destination
In guided media communication device are directly
using some physical media like wires. It also called bounded connected wit
media. h each Other by , Q.2.(c) Compare the throughput of a pure ALOHA
network with a slotted
Unguided Media: In unguided media communication devices
communicates y; eC network:
each other through air or space using broadcast radio signals (2)
microwave signal with “Ans. Throughput of pureA LOHA channel ~
infrared signals unguided media is used where
it is impossible wuided
to instal l a data‘ia cay
be transfer all over the world using media it is also Let G be the average number of frames generated by the system during
j : called un one frame
Media. transmission time. Then it can be proved that the average no of successful
aoe ete are fwo different modes for propagati transmissions
S m : ne le = single mode. There are two basic types ng light along opti for pure ALOHAis S=Gxe 2°. The
of fiber: Multimode one - halfa frame is generated duringmaximum throughput S,_. is 0.184, for G=/2. ic. if
a
er and single-mode one one frame transmission time, then 18.4 percent
fiber : ; _ of these frames reach their destination successfully. This is an expected result because
:
_ the vulnerable time is 2 times the frame transmission time. Therefore,
ie if a station
‘ _ generates only one frame in this vulnerable time, the frame will reach its destination
Propagation Mode " successfully. : i
; The throughput for pure ALOHA is S = G x e2¢
yi vines __ Themaximum throughput S,,,,
= 0.184 when
G = (V2)
J aq __ Throughput of slotted ALOHA channel
The average number of successful transmissions for slotted ALOHA is S = Gxe.
Multimode ~ Single Mode The maximum throughput Sax 18 0.368, when G = 1. In other words, if a frame is
_ | generated during one frame transmission time, then 36.8 percent of these frames reach
f 3 their destination successfully. This result can be expected because the vulnerable time
Multimode: Multimode is so named because multiple beams
from a light sourt js equal to the frame transmission time. Therefore,
move ie if a station generates only one
ae Seen paths. Multimode cable is made of glass fibers, with? frame in this vulnerable-time, the frame will reach its
ae ae = me bs e ee range for the light carry component (the m Br stirsiat tine teres secs destination successfully _

Multimode fiber gives you high bandwidth at high speeds over medium
distanced The maximum throughput S__,. = 0.368 when G=1
. Light waves are dispersed into numerous paths, or modes, as they
travel through tht
cable's core typically 850 or 1300 nm. Typical multimode fiber core diameters are 5) Be :
_@-3.(a) Why channel allocation is a difficult task? Explaincs the random access
62.5, and 100 micrometers. method that tries to avoid collision. : (5)
Single Mode: Single mode uses step-index fiber and a highly focused source of figh
Ans. Refer Q.3 End term june — 2016. ; ;
that limits becams to small range of angles, all close to the horizontal.
Random Access, which is to issue a completely random time, relies on the Aloha
me — ~ method.
' In this method, when a coupler has information to transmit, it sends it without
worry about other users, If there is a collision, that is to say superposition of two signals
Source or more users, the signals become indecipherable and are lost. They are subsequently
 8-2017 Sixth Semester, Computer Network
transmitted, as shown in Figure, in which the couplers 1, 2 and 3 collide, The eo,
transmits its field first because he shot the smallest timer. Then, the modu] “Ue LP. University-[B.Tech.]-AB Publisher 2017-9
and its signals collide with the coupler 1. Both derive a random time of re} ‘ Zen,
The coupler 3 is listening while the couplers 1 and 2 are silent, so that the— 10001100
coupler 3 passes successfully. Technical aloha is the origin of all the rando loo} 10011101000
methods. , 1001

The flow rate of such a system becomes very small when the number of 0

oo
oo
0
increases. It can be shown mathematically that if the number of stations goes tone
the flow becomes zero. From a certain moment, the system is more stable. To reduc

soo

o=-
oo

o-
likelihood of conflict between users, various improvements of this technique haveh
proposed. bee

o-

oo
o-
oo

on
eo
——
Coupler 1

oso
eo
10

oo
00

Coupler 2 oo

eo
o-
00
1 0 0 (remainder)
Coupler 3 =e
Actual frame transmitted: 10011101000-100=10011101100 (modulo 2
Operating Principle of Pure Aloha
Now suppose the third bit from the left is garbled and the frame is received as
Slotted aloha, aloha
or sliced 10111101100. Hence on dividing this by the polynomial generator we get a remainder of
100 which shows that an error has occured. Had the received frame been error free we
Improved technical aloha was to cut the time into time slots, or slots, and j would have got a remainder of zero. See below.
authorize the issuance of frames that slice first, the transmission time of a fram
requiring exactly a slice of time. In this way, there is no collision if a single fram 10101000
transmitted at the beginning of slice. However, if several frames start transmittingiy Horiti1101000
the beginning of slice, the frames emissions are superimposed along the slot. In the 1001
4
latter case, there has retransmission after a random time.

oso
oo
o-

on
This method improves the throughput during the start-up period but remai
the devia
unstable. In addition, there is an additional cost from a complication of

—-——
oo
o-
since all emissions must be synchronized.
Q3.(b) Abit stream 10011101 is transmitted using the standard CRC mei

oso

oso
o-

oo
The generator polynomial is x*+ 1. Show the actual bit string being transmitted
Suppose a bit is inverted during transmission. Show how this error is dete

-——
——
oo
oo
at the receivers end.
o1
Ans. Our generator G(x) = x* + 1 encoded as 1001. Because the generator voir
oo
Jos 00
is of the degree three we append three zeros to the lower end of the frame to
transmitted. Hence after appending the 3 zeros the bit stream is 10011101000, 01 10
os
os

oo

remainder of 100. We do modulo 2 subtraction thereafter of the remainder from theb Mt 100
stream with the three zeros appended. The actual frame venemlned is 10011101100 000
See below.- 3
1 0 0 (remainder indicating error)
 :

10-2017 Sixth Semester, Computer Network

;
in the data link 1p University-{B.Tech.-AB Publisher
Q.4.(a) What are various design issues involved lay,
2017-11
Explain sliding window protocol.
i are the h sequi ence number associated
Ans. The data link layer has a number of specific functions to carry out Thy4 «+ ig i bit.i The last three bits
Poll/Final
K E RNR, REJ and SREJ indicated in the SS-Bits of the
ude
functions include providing
idi a well-defined
ei service
ice interface
i to the network la
, with Paice
control
thefield, alaccy
Note
supervisory stateo
thatstates RR,
an acknowledgement (RR) can be sent either on a supervisory
REJ
determining how the bits of the physical layer are grouped into frames, dealin, ve,
frame or piggybacked on an I-Frame as g menti me oned earlier. On the other hand RNR,
SW}
transmission errors and regulating the flow or frames so that slow receivers
: } aeny and SREJ are sent only through a supervisory frame.
swapped by fast senders.
There are the following issues. Control field of S-frame
(a) Services Provided to the Netowrk Layer i 1 Aig pao eG Cacia 58
oe cane ee Ras T aoe ae
Cl be ie
(b) Framing
pple ays ve peal a
(c) Error Control ; FCS
1. -- CONTROLFIELD ----- oe ee
ADDRESS- >>: ~€ CONTROE oo ee
(d) Flow Control : 00 READ’
d
Refer Q.3.(a) of First Term Examination 2016. ee: 10 ReeRECE! ENO? READY (RNR)
(8) : 11 SELECTIVE REJECT ee)
Q.4.(b). State and explain various frame types in HDLC. N(R) : ACKNOWLEDGMENT NUMBE
es : + POLL/FINAL BIT
Ans. There are three types of HDLC frames :
¢ Information transfer frame (I-Frame)
Unnumbered Frame (U-Frame) oe
¢ Supervisory frame (S-Frame) are used for li
not have data field as S-Frames. They
The U-Frames do
°U; a establishment, termination, mode setting and other control functions. Controls field of
SS
Frame (I-FRAME) an unnumbered frame is shown. ;
Information Transfer
identify an unnumbered frame.
I-Frame is used for transporting user data. It also carries acknowledgement of the The first two bits of the control field are 11 which .
. The fifth bit is Poll/Final bit. The rest five bits are called modifier bits. They speci ify the
received frames. The control field of the I-Frame is as shown in Fig.. The first bit is 0
control function.
which identifies the frame as an I-Frame. The next three bits are the sequence number
N(S) of the frame. Cntrol Field ofa U-Frame.
The fifth is Poll/Final (P/F) bit. Gernot se
See ee Sa ae
acknowledgement (RR)
The last three bits are the sequence number N(R) of the ; M
. ; 1)1/ MIMI PF MyM
which is piggy backed on the I-Frame ee ee ees
. ees.
:
Control field of I-frame
MMMMM__ : MODIFIER BITS
;
ee Os Goce
2 PIF : POLUFINAL BIT
0 Is) PIF iN
ee fans wooee
: Q.4.(c) If the 7-bit Hamming code word is received by a receiver is 1011011.
ADDRESS = => <€ ~- = ~ CONTROL FIELD=~ == =>; - DATA Assuming the even parity state, find whether tne received code word is correct
N(S) FRAME SEQUENCE NUMBER or wrong. If wrong, locate the error and correct the code word. (3)
N(R) ACKNOWLEDGMENT NUMBER
P/F POLL/FINAL BIT Ans.
a7 d6 d5 p4 d3 p2 pl
Supervisory Frame (S-FRAME)
cL 0 1 1 0 1 1
$-Frame does not have data field and is used to carry only acknowledgements
requests for retransmission, etc. It is identified by the first two bits of the control fiel P4=d5 d6 d7
Bes 1 A 1 .
These two bits are 10 in an S-Frame. The next two bits (SS) are used to indicate fol
we pete
supervisory states. Receive Ready (RR), Receive Not Ready (RNR), Reject (REJ) an
Selective Reject (SREJ).
a P2 = p2 d3 d6 d7
12-2017 _ Sixth Semester, Computer Network
XAMINATION [MAY-JUNE 2017]
= 0=>00 1 p2=0
END TN ee SEMESTER [B-TECHI
Pi = pl d3 p5 d7
sir aes 01 . COMPUTER NETWORK [ETCS-306]
MLM. : 75
ere ;
Time : 3 Hrs- is compulsory. Attempt any five. Questions from the rest.
P4 p2 pl=101=5 cro.No.1
Me
control? (3)
Error at 5" bit.
How flow control is different from congesition
sg:
GO). v
be : 1001011 y 4 : e:

Corrected answer after changing the 5" bit will

Q.1. (a)
oO
ns preventing the source from sending data that the sink
This is fairly easy with a sliding
coaxial cable. Ans. Flow control raed ee censout of buffer space.
Q.4.(d) List the characteristics of broad band make sure the gource’s window
r is no larger
es than the free space
: in
the signal of higher frequeng justbecare
willi end up dropping
Ans. Coaxial cables are the guided media that carnes in the
he sink advertise its free buffer space
called coax. (short form) window Pe TCP does thi s by letting t
range compared to twisted pair cable. Coaxial cables are also ohn the sink’s is elk Sacks:
and 75 ohm cable. 50
Two types of coaxial cables are widely used: 50 ohm cable sending
transmission, window field 0 ns preventing (or trying to-prevent) the source from
cable is used for digital transmission and 75 ohm cable is used for analog is full. This is more
Due to the shield provided, this cable has excellent
noise immunity. It has a large Congestio:
ee nee ened by a router because its queue
to install. They are often installed data that will end up travelling different paths can
bandwidth and low losses. Co axial cables are easy ackets from different sources
or a star (ARC net). complicated, because e P
either in a device to device daisy chain (Ethernet) The cover shields Converge 0 the same queue-
;
small cables in a protective cover. te and paud rate? Explain with example.
Acoaxial cable consists of many
interference. Within Q.1.(b) Difference between bit ra’
the cable from physical dangers as well as from electromagnetic (3)
one another. Coaxial
the cover, the various cables are shielded from interference with ous
many simultane
are used in communication networks that require l in bits per second. Bit rates
cables
5000 links. It has a Ans. Data rate é: 3 the capacity of a channe
communication links. Each coaxial cable can provide more than measure the number 0 £ data bits (that is 0?s
and 1?s) transmitted in one second in a
of the inner
Mbps which can be increased with the increase in diameter
data rate of 10 bits per second means 2400 zeros or ones can
on a thin net segment is 30 nodes communication channel. A figure of 2400
conductor. The specified maximum number of nodes ed in one second, hence the abbreviat ion “pps.” Individual characters (for
: be transmitt of several
and on a thick net it is 100 nodes. to as bytes are composed
it requires amplifiers which are example letters or numbers) that are also referred
Since broadband is used for large area, one used for bits.
unidirectional. In dual band systems two identical cables run together, cy,
s are given for inbound and Baud rate represents the m umber of times the line condition (i.e., frequen
outgoing data, one for incoming data. Different bandwidth bits x baud rate, where
z for outbound amplitude or phase) changes each second. Data rate =n
outbound cables. Eg: for 300MHz, 5-30MHz for inbound and 40-300MH 10 BASE 2 n is equal to the number of bits one signal represents. A baud rate
is the number of
There are two systems that use the bus LANs namely 10 BA SE 5 and or varies. For example, a
: times a signal in a communications channel changes state
times per second.
2400 baud rate means that the channel can change states up to 2400
0 up to X (in
The term “change state” means that it can change from 0 to 1 or from 1 to
this case, 2400) times per second. It also refers to the actual state of the connection,
: such as voltage, frequency, or phase level.
Q.1.(c) What is difference between Classfull addressing and Classless IP
addressing? _ ; (4)
Ans. Classfull Addressing:

All IP addresses have a network and host portion. In classful addressing

the network portion ends on one of the separating dots in the address (on an BctEE
Rene Classful addressing divides an_IP address into the Network and Host portions
Re are Eee the classful addressing system all the IP addresses that are
; ce, . Renee into en five classes A,B,C,D and E, in which class A,B and C

class E is reserved aaa enc ean se pe ee clea

 ‘
—15
k
Computer Networ 3 2017
Sixth Semester, J-A B Publisher
14-2017 y— [B.Tec h. ry can be
LP. Uni versit the
nt, where
elec tron ic equl. pme
Class A 31 th 1 arge
factory wi
3 such as 4
bit# 01 interference, d an d ma in tained.
alle ;
properly inst ‘UTP is not
.
6 ielded while st. ie
= Dee beets sh are no
Sarasa even
Network W: hil e UTPs ee
vy duty use
ee, 2. STPs are for hea
dwidth whi le 's
pees co
. ot (2.5)
Class B 8 Bytes 31 3. STP s allo w max imum pan
15 16
tw ee n Int ranet, an d ea se
2 iate be ent hardware
.1.(e) Different
bit# O
en with differ people attach
e
the world, oft icate with
Many networks e xist in mm un y
wor k Num ber —— > Ans.
: two rk oft en wan t to co
dif fer ent , and frequentl
er—> <— Net 16 Bytes to one ne gether
<+— Network Numb People connected
requ ires connecting to ga te ways to make
the
desi re cal led
16 Bytes different one. This using machines ware and
Class C 23 24 31 , sometimes b y terms of hard
incompatible networks ation, both in t
the nece ssary transl an in te rnetwork or jus
ele
bit# O 3
connection and prov ide ks is cal led
ae +_ Host —>
software
internet.
. A colle ction of inte rconnected networ!
rks using the sam
e
Number-————> le smaller netwo:
<——__ Network Number
made up of multip d to the
24 Byte s 8 Bytes It is simply any netwo' rk “i”) isn’ t nec ess arily co: nnecte
s. An internet (little its internet wor ‘kin
g protocol.
internetworking protocol
why they are classfj rily use TCP/IP as
ess belon gs to a particu! lar class that’s e Inte rnet (big “I”), nor does it nec essa
t have any name,but when classl
Each of the IP addr e internets.
essing system did no! There are isolated corporat began its life as the
addresses.Earlier this addr as Classful addressiy to the network th: at
em came into exist en' ce then it is named liy The Inte rnet , with a capital “I,” refe rs n © f all TCP/IP networks
addressing syst ful addressing is that it limit
ed the flexibi the confederatio
disa dvan tage of class AKP Ane t and cont inues today as, roughly, s. Seen up close, it’s actually
system.The main maj backbone
t o any device.One of the ected to commercial U.S.
s that can be assigned directly or indirectly conn IP backbones, corporat
e and U.S.
and number of addresse no’ t send subne t infor mation buti orks -com merc ial TCP/
addressing is that it does quite a few different netw terconnected
disadvantage of classful orks , and TCP/ IP netw orks in other countries-in
ork address. government TCP/IP netw
will send the complete netw
by high-speed digital circuits. ,
Classless Addressing: to connect universities
ork and hos e internet that is widely used
a varia ble number of bits for the netw ‘ It means a specific worldwid vidu als.
Classless addre ssing uses s, and of late, priv ate indi
government offices, companie ther. In our
portions of the address. inct networks are connected toge
as a 32 bit stream of ones and zerog z An internetwork is formed when dist ,
Classless addressing trea
ts the IP address
ere betwe en be ng LAN and a WAN or conn ecti ng two LANs forms an internetwork but
can fall anywh view, connecti a
ne’ twork and host portions ogy in this arga.
where the boundary between known as CIDR( Class less Inte
there is little agre emen t in the indu stry over term inol
addressing sy: stem is also .
0 and bit 31.Classless to allocate and specify the Intern
e lable to the world outside of the Intranet
ss addressing is a way Intranet: This is a network that is not avai
Domain Routing).Classle more flexi bly than with the origin al syste m
anet netw ork is conn ected to the Internet, the Intranet will reside beh
ind a
main routi ng If the Intr
addresses used in inter-do Internet Domain Routing , be an Extranet. The firewall
ss classes. CIDR (Classless firewall and, if it allows access from the Internet will
Internet Protocol (IP) addre significanbi t
y-siz ed subne ts solel y by base address and number of helps to contr ol acces s betw een the Intranet and Internet to permit access to the Intranet
defines arbitraril pany or organisation. . ;
in the address. only to people who are members of the same com
(2! be ‘ a net wor ked PG : wi
es of STP over UTP? lest form, an Intr anet can set up on
Q.1.(d) What is major advantag In its simp ro
Internet eS
ng shielded made of metallich on the network having access via the Intranet network to the
Ans. Shielded twi sted pair cable have a conducti 7
block s the elec trom agnetic interfer ence allo wil ple, cons ider an offic e with a few PC: i in
encas ing the twist ed wire pairs , which For exam , ked
ielded twisted pair howeré d not be connecteyd to webs tha! eae for Detiers all. networ
at faste r rate of speed rathe r than unsh toget her. The netw ork woul outs ide worl d. On one of the drives
it to carry data
cting interference tof Pap TES EC: bare would bo a dixector of
they have sever al disa dvan tage s STP cable s work by attra
fate bee see ser we tntranete.
shield them running it off into a grounded cable. It the cable is impr
operly groun de Other PCs.on the network could access this Intranet y pointing their brow (Netscap
ely comp romi sed addit ional ly, STP cl or Inte rnet Expl orer) to this directory - for example
ee } oa pron are sever , voile
and are more expensive. F i
igger cable, U:\inet\index.htm ;
UTP cables, as the shielded must be kept ee ee Hee cromore aaa t
em to work prop erljf then onwards the y would navi avigate around the Intranet in the same way as
setti i i From
ing with high amounts of electromagt#
best use for STP cable are in indus trial net.
the Inter
they would get around
 2017-17
Network
Sixth Semester, Computer Uni versity 1B ‘Te
ch. ]-AB Publisher
16-2017 LP.
of the network
baseband transmission and broad bag
es are more the performance
Q.1.(f) Difference between
(2.5) 2. If network traffic is heavy or nod

trans: mission? decreases.

hn
an Ethernet cable, is a 50-o
cable, commonly calle d 3. Itis slower than the ring topology.
Ans. A baseband coaxial signal . Tran smis sion in baseb and cable,
dulated digita l
cable that transmits’an unmo point propagates in both
directions, RING Topology
mea nin g 4 signal inserted at any ial cable s ures of Ring Topology.
is bidirect iona l,
area networks. Broadb and coax Feat by having 2
com mon ly use d in loca l es are is unidirectional, bu tit can
be made bidirectional
Bas eba nd cabl es are Bro adb and cabl 1.The trans missi on gy.
analog signals. Node, it is ¢ alled Dual
Ring Topolo
transmit modulated, connections between each Network
are 75-ohm cables that this by divi ding into different channels. bit by bit. Data transmitted,
compensat e for that is
unidirectional, but can one sign al at a time , . 2. Data is transferred in a sequential mann er .
ing more than k, til 1 the destination node
not capable of transmitt eband cables has to pass through each node of the networ
Baseband cables are ta and voic e. Add iti ona lly , bas
s are resti ricted to da ion rates unless an Advantages of Ring Topology
and the transmission ad ly drop in transmiss as
gth exp eri enc e a ste ated low ic or by adding more nodes,
exceed ing 0.6 2 mil e in len
of usi ng bas eba nd cabl e are the ass oci 1.Tran smitting network is not affected by hi gh traff
ever, advantage s is the absence of a
amplifier is used. How ing baseb and cable only the nodes having tokens can transmit data.
eas e of use . Ano ther benefit of usi
cost and the already digital. 2. Cheap to install and expand
cable’s signals are speed. A
modem because the gle sig nal at a time at very high Disad vantages of Ring Topology
sin ports frequency
cable transmits a coaxial cable sup
Abaseband co axial Ns . Ba se ba nd 50 ohm 1.Trou bleshooting is difficult in ring topology.
m ainly used for LA l cables are
baseband cable is sig nal ing . Baseband coaxia ger rk activity.
dig ita l Gbp s. Lon or deleting the computers disturbs the netwo
and are used for bandwidth is 1-2 2.Addi ng
range of a-4kHz ’. For [K m cab les the
gital transmission am plifiers. 3.Failure of one computer disturbs the whole network.
cables used for ‘di dat a rates or periodic video
d wit h low e of voi ce,
cables can be use erge nc STAR Topology
orks: The conv networks
oad band netw mmunications
rol e of co ax ial cable in br in br oa db an d co Features of Star Topology
Th e re is the trend nal capability
ngle infrastructu er broadband sig
and data over a si cable (hf c) ne tw or ks off 1.Every node has its own dedicated connection to the hub.
id fiber/coaxial
worldwide. Hybr e an d ec on om y.
gged performanc as a repeater for data flow.
the p roblems and 2.Hub acts
combined with ru or k To po logy? Discuss
you mean by Ne tw (4.5 ) 8.Can be used with twisted pair, Optical Fibre or coaxial cable.
Q.1.(g)- What do gi es ? k
three topolo nected in & net wor Advantages of Star Topology
benefits of any t of com pute: rs con
top olo gy ref ers to the arrangemen T
oloo p gy gene rall y determines” 1.Fast performance with few nodes and low network traffic
Ans. Network optical fiber ete. Top
me phy sic al me dium such as cable, or k top olo gie s are as follows: 2. Easy to troubleshoot. é
through so ious types of netw
sha pe of the network. The var
the 2. Bus topology 3. Easy to setup and modify.
pology
1s Hierarchical to Rin g top olo gy Disadvan tages of Star Topology
4.
3. Star topology top olo gy 1. Cost of installation is high.
6. Hybrid
5. Mesh topology 2. Expensive to use.
1. BUS Topology
t
{ 8. If the hub fails then the whole v network is sto pped because all the n odes depend
po lo gy
Features of Bus To .
nm the hub.
1. Ittr ansmits
data only in one direct
le
ion
cable "Q.1.(h) What is routing? ED Di g? Diff erentiiate between dista . 2
nce vector routing and
ce is con nec ted to a sing
2. Eve ry devi - (3)
pology Ans. Routing is the process of forwardin g ofa packet in - a netw
Advantages of Bus To h
ork so that it reacches
B its intended destination. The main goals of routing ar a:
1. Itis cost effective.
ed to oth er net wor k top ology. 1. Correctness: Th: e routing
par i
2. Cable required is least com should be done properly and correct] ctly so that the
‘ ackets may reach their proper destination
3. Used in smal l netw orks .
2. Simplicity: The routing should be done ina sim
Disadvantages of Bus Topology ple manner so that th e overhead
3 as low as possible. With inoreasin
1. Cables fails then whole network fails. so increases.
g complexi ty of the routi
: uting
algorithms the overhead
ER

ork 2017-19
uter Netw IP. Universit: y-[B-Tech,
Sixth Seme
ster, Comp cted , ech. |-AB Publisher
it may be expe
18-2017 bec o} me s operative, ed forrO Uting | Distance——How far |
maj or network gorithms de
sign
ss: Once ar Jures. The al ld be abje |
Vector—In which direction
3. Ro bu st ne an y fai s an d sh ou
s W1 thout software fai
jure
hosts tj
usly for year rdwar' e and jobs in all
run continuo gh to han dle ha o ut re quiring all .
bu st en ou ffic wi th
should be ro in th e to pology and tra e ro ut er goes down.
change s ery time som
to cope with k rebooted ev der all
POSSib]y
d the networ d be stable un i eB
be aborted an ms sh ou l a |Routing Routin 9 ; 2
in g algorith Routin 9 Routing
y: The rout table
table
ee
table
4, Stabilit renal
s.
circumstance ques.
routing techni Routers pass periodi ic copie:
euie a = 0 f theiri routing
i table to neighbouring rout uters
Co mp ar e the following
|
uting |
e Link state ro To maintain the sta’ te of all links i
tor ro uting two factors: the distance,
« Distan ce vec
because i
{ invo: ves
get there. Routing consume less overhe ad at the expensei ofeer limited ona
one . Fe Because
cach routers
soaker
to take to
is SO named have only a
ctor routing or direc! tion This means ul limited view of the netw ork, ti i ; -
Distance ve vec t or, prevent routing loops. ools like split horizon and poison reverse are needed
ded to
of a de stination, an
d the
dir ect ly com nected neighbours. where that
ric , know
or met d between ne d, but it does
not
only exchange te W as lear This aspect of
information is ig hb ou r & rou neighbours. Link-State: : Now, let’
et’s | i
from which ne beyond its own like at the topology running a link-s ta
Orne protocol (in
router knows te; a rou! “routing by rumor.” Measures a single area). Recalled i record s the state of link eae
all
arned the rou router
neighbour le
uting is some
time’
av oi d routing loops. can construct a short:rtest-path tree from itself to all known wn d destin
Bee .
ations
vec tor ro em pl oy ed to
distance e are about the pathy
poison revers rout ers know ning § link-
Although maintaining link-state
li informationi for the entire area typically requi
split horizon and re qu ires that all ion is flooded
i equires
st, rm at more overhead than does processing
routing, in
con tra
k. Link-stat
e in fo
er crovide ior! sshaaieeeet ata eerti sements only from direct neighbours, but
adverti
to ensure air rout
Link-state the ne tw or
ters in
all other rou O SPF or IS- 1S) s commoy
reachable by domain (an area in ase . Fr om thi
datab
throughout the
link-state a’s link-sta te tree, wi th itself
as th { Q.2.(a). Calculate
; th e throughput of aa
copy of the are shor’ test-path apes at OHA channel )
§
posses 2 ynchronized its own re lat ive Which Channel gives better urwegiene Gee y? 2 (6)
router constructs
database, each
e

tes. Ans. Refer Q.2. (c) of First Term Examination 2017

all kn ow n rou
root, for
lowing topology: Slotted aloha gives b etter perfo sos
Consider the fol Seen es The numbe: ee that can possibly
is reduced by a huge ee
a o
k
Client Os networ take place in :sl ee
ee .

we Perens
Pure Aloha. Collisions may onl mak
™
Network 1 fe e place with nodes
rooeee Fee) that are ready to d ‘beak aeat theto same time. But cachet
, this is a substantial
* reduction.
Q.2.(b). Define Sto p and Wait AR
from stop and: wait AR: Q protocol to Caen
th e GO-Back-N
eNomARQ protocol? as (6.5)
Ans. Stop and Wai‘ait transmission i 5 reliabili ;
si ssion is the simplest
adequate for a very Sat e mit
protocol transmits
eee
C
Data e e x
a Protoc ol
of informatio n and then waits fora ver
receivi es each PDU and PDU fa peabe The recei
Network 2 an Acknowledgement (ACK)
Roriectiy ail a et
PDUiffytheseidt
(NACKly) identi
ementto reliab aaa
corey anda Negative Acknotnwledgable
Network 3
Netware file server se Hee received.
(“=p
Mcsiveduendcne pe : er a PDU has bee =
nt ae er will usuall sneer y als implement a timer to recov.
link-state routing p rotoc
ols are suitable for deploy from th e condition where the receiver does
Both distance vector and in g routes in a different manner, |
go about propagat
on this network, but each will Under normal transmission th
e sender will receive an ACK es data and then
ocol like
Distance Vector: If we were
to run a distance vector routing prot commence transmissi ss next data block. For a long dela:
uming eachlink a eee the
w Rl woul d see the netw ork, ass have to wait an
EIGRP on this topology, here’s ho ciable time for this response, While it is oe ite eta
ing the sender is said
been omitted): ° to be ini the “idle”
metric of 1 (locally connected routes have “idle” state and is unable to send further data
:
Network 2017-21
, Computer IP. University-[B.Tech.]-AB Publisher
Sixth Semester t a
20-2017 Stop-and-wai not fit any
of ac on ne ction than is ARsti l 7. OSI modelhas a problem of fitting 7. TCP/IP model does
use on
is a more efficient et, the connecti
Go-Back-N ARQ t for each pack the protocols in the model protocol
sf ahedaa als, is not
since unlike wait avng
s, duri aE
ert the O Protocols are hidden in OSI model In TCP/IP replacing protocol
e packeee nie tilebesini gwordsent. However, this methi
are bein od g. 8.
ing, B moere
ket s
ita
pac
being utilized as nt wae ts are $ e e e and are easily replaced as the easy.
i , ee
iti mes-if ape Ea me
ames multiple ti
technology changes.
i ai e fr separated
cae d, the if tha g. OSI model defines services, inter- 9. In TCP/IP it is not clearly
be
thesout error) will: To avol this, protocols.
re-sent. faces and protocols very clearly its services, interfaces and
we re re oed wiyat
niv
ce
rd r if th
(ae he ey at e. o
ie
Sdn P
thSelective ARQ can be used.
Repeat models?
and makes clear distination
TCP/IP networking between them.
s t and Co mp ar e ISO-OSI an da (6.5)
It has 4 layers
Q.3.(a). Contra 10. It has 7 layers 10.
] Protocol/
ansmission Contro algorithm? Explain in detail?
sys tem Int erc onn ection) T CP/IP (Tr Q-3-(P) tis the need for adaptive routing
Ans. OSI (Open (6)
ernet Protocol).
Internet Protocol/Int available
Ans. Adaptive routing is driven purely by the current state information
TCP/IP the network’s ability to adapt to condition s
Osi a ai} for the network. The term adaptive refers to
Application and find a route optimal to the present conditions. Unlike time-dependent routing,
Lee have no dependence
alternative routes are evaluated on a real-time basis and as such
t—~__Not present is computati on intensive at the switches since the
_on the time of day. Adaptive routing
Presentation (ee IN the model
6 eS
a
state information must be examined frequently but it is also more responsive to local
oe Session oe network conditions.
5
Transport Time-dependent routing is not considered adaptive because routing alternatives
4 Transport es
e eee e
e
remain fixed during a constant time period, such as one hour.
eRe | ——
rece
Internet
1. Centralized: In this type some central node in the network gets entire information
je
about the network topology, about the traffic and about other nodes. This then transmits
2 Data link this information to the respective routers. The advantage of this is that only one node is
ere Se Host-to-network
Physical required to keep the information. The disadvantage is that if the central node goes
eer down the entire network is down, i.e. single point of failure.
Pee n Control |
TCP/IP (Transmissio |
2. Isolated: In this method the node decides the routing without seeking information
OSI (Open System ProtocoolV/Internet
Protocol)
from other nodes. The sending node does not know about the status of a particular link.
Interconnection)
ieee TCP/IP model is mor
e based on | The disadvantage is that the packet may be send through a congested route resulting in
ing and 1.
1. OSI provides layer function protocols and protocol are
s not with a delay. Some examples of this type of algorithm for routing are:
tion s of all the
also defines func r laye rs. ° Hot Potato: When a packet comes to a node, it tries to get rid of it as fast as it can,
flexible othe
layers. r
2s In TCP/IP model the transport laye by putting it on the shortest output queue without regard to where that link leads. A
In OSI model the transport layer
2.
ets does not guarantees delivery of variation of this algorithm is to combine static routing-with the hot potato algorithm.
guarantees the delivery of pack When a packet arrives, the routing algorithm takes into account both the static weights
packets. of the links and the queue lengths.
3. Follows vertical approach.
3. Follows horizontal approach
4. TCP/IP does not have a separate ¢ Backward Learning: In this method the routing tables at each node gets modified
4, OSI model has a separate by information from the incoming packets. One way to implement backward learning is
presentation layer
presentation layer
to include the identity of the source node in each packet, together with a hop counter
5. OSlis a general model. 5. TCP/IP model cannot be used in any
that is incremented on each hop. When a node receives a packet in a particular line, it
other application. notes down the number of hops it has taken to reach it from the source node. If the
6. Network layer of OSI model 6. The Network layer in TCP/IP previous value of hop count stored in the node is better than the current one then
provide both connection oriented provides connectionless service. nothing is done but if the current value is better then the value is updated for future use
and connection less service. The problem with this is that when the best route goes down then it cannot’recall
the
Roa

t ; A
LP. University-{B.Tech.]-AB Publisher thsi7-23
‘ Sixth Semester, Computer Network pa: e
Bee icul de, Hence all the nodes have to forget the sto,., functions: eaaiiael
The ATM Physical Layer: The ATM physical layer has four the
; physical me
converted into a bitstream, the transmission and receipt of bits on
second best route to a particular t node. again.
ly and star all over are controlled, ATM cell boundaries are tracked, and cells
are packeuss ie
information pe riodical fi tion from its neighbouring node, ; kaged
this t
eives informatio ;
ese a Deane
ii i
For examplele, cells are pa’
3. Distributed: In
way to send the packet. The disadvantage, _ apProP tty for SO!
he decision Hane est i d sends the packet something |; diffe ; 7
: medium-de
and then takes the de6!9" ‘ The ATM physical layer is divided into two parts: the physical pendent
it receives information an P
that ifin between the the interval
| blayer and the transmission convergence (TC) sublayer.
changes then the pac ket may be delayed. all fixed 1 | (PMD)su ne
reference model? Why does ATM use sm: “a : The PMD sublayer provides two key functions. First, it synchronizes transmission
Q.4.(a) Explain ATM See
; ae. and reception by sending and receiving a continuous flow of bits with associated
cell? What is ATM signaling?
Ans. ATM: Asynchronous Transfer Mode (ATM) is a standard switching information. Second, it specifies the physical media for the physical medium VEE
The TC sub
. layer has four functions: cell delineation,
and computer networks. It uses cluding connector types and cable
: ns. designed to unify telecommunication
cell-rate decoupling,
, [1)[2] and it encodes data into small, fixed. header error control (HEC) sequence generation and verification,
a time-division multiplexing
ke are and transmission frame adaptation.
as the Internet Protocol or Ethernet that
Avs tila This differs from approaches such p y AAL1: AAL1, a connection-oriented service, is suitable
i
; data link layer services that run) ATM Adaptation Layers:
use variablei sized
i p ackets or frames. ATM provides A’
for handling constant bit rate sources (CBR), such as voice and videoconferencing.
over a wide range of OSI physical Layer links. transports CBR traffic using circuit-emulation services. Circuit-emulation
service also
ndpoint ;
ATM Endpoint ATM Switch a - ; accommodates the attachment of equipment currently using leased lines to an ATM
ATM Adaptation ATM Seen backbone network. AAL1 requires timing synchronization between the source and the
Layer destination. For this reason, AAL1 depends on a medium, such as SONET, that supports

ATM Layer ATM Layer ae. clocking.

5 i Layer ATM Adaptation Layers: AAL2: Another traffic type has timing requirements
Physical Layer Physical Layer like CBR but tends to be bursty in nature. This is called variable bit rate (VBR) traffic.
This typically includes services characterized as packetized voice or video that do not
have a constant data transmission speed but that do have requirements similar to
: constant bit rate services. AAL2 is suitable for VBR traffic. The AAL2 process uses 44
Oe bytes of the cell payload for user data and reserves 4 bytes of the payload to support the
ATM layers: AAL2 processes.
The ATM reference model is composed of the following
* Physical layer VBR traffic is characterized as either real-time (VBR-RT) or as non-real-time
(VBR-NRT). AAL2 supports both types of VBR traffic. ;
¢ ATM layer ATM adaptation layer (AAL) *
* Physical layer - Analogous to the physical layer of the OSI reference model, the ATM Adaptation Layers: AAL3/4: AAL3/4 supports both connection-oriented and
connectionless data. It was designed for network service providers and is closely aligned
ATM PI physical layer manages the medium-dependent transmission. The ATM reference with Switched Multimegabit Data Service (SMDS). AAL3/4 is used to transmit SMDS
model is coniposed of the following ATM layers.
packets over an ATM network.
* ATM layer - Combined with the ATM adaptation layer, the ATM layer is roughly
An AAL 3/4 SAR PDU header consists of Type, Sequence Number, and Multiplexing
analogous to the data link layer of the OSI reference model. The ATM layer is responsible”
for the simultaneous sharing of virtual circuits over a physical link (cell multiplexing) Identifier fields. Type fields identify whether a cell is the beginning, continuation, or
and passing cells through the ATM network (cell relay). To do this, it uses the VPI end of a message. Sequence number fields identify the order in which cells should be
VCI information in the header of each ATM cell. reassembled. The Multiplexing Identifier field determines which cells from different
traffic sources are interleaved on the same virtual circuit connection (VCC) so that the
¢ ATM adaptation layer (AAL) - Combined with the ATM layer, the AAL is rougl i
correct cells are reassembled at the destination.
analogous to the data link layer of the OSI model. The AAL is responsible for isolati
higher-layer protocols from the details of the ATM processes. The adaptation k ATM Adaptation Layers: AAL5: AALS is the primary AAL for data and supports
prepares user data for conversion into cells and segments the data i both ction-oriented and ctionl data. It is used to transfer mast non-.
48-byte ce
payloads. ra a SMDS data, such as classical IP over ATM and LAN Emulation (LANE). AAL5 also is
known as the simple and efficient adaptation layer (SEAL) because the SAR sublayer
simply accepts the CS-PDU and segments it into 48-octet SAR-PDUs without reserving
any bytesin each cell.
 Network
Sixth Semester, Computer
94-2017 LP. University-[B.Tech.|-AB Publisher 2017-25
m ultiplexing of information on a given medium than
1. Allows better statistical © tn terface (BRI) and Primary Rate Interface (PRI). BRI
consists of two 64 kb/s B
ets of varia ble length. | channels and one 16 kb/s D channel for a total of 144 kb/s. This basic service is
the use of larger pack sah | intended
packets, hig! speed switching would be more comp)
; 3 , » meet the needs of most individual users. ‘
2. With longer varib le-le ngth
PrOoCeGaa |
the header allows ATM cells to be
3. The limited amount of info rmation in : . PRI is intended for users with greater capacity requirements. Typically the channel
at a very high rate from 150 Mbps to several Gbps. - structure 1s 23 B channels plus one 64 kb/s D channel for a total of 1536 kb/s.
4. Provides speed, flexibility and compatibility with any different applications ang Some advantages of ISDN
physical media.
tees a minimal
; delay and dete Speed: ISDN allows multiple digital channels to be operated
guaran’
5. The reduced size of the internal buffers simultaneously
consta nt rate traffic (audio, video) and variably) through the same regular phone wiring used for analog lines. The change comes about
jitter. This is why ATM can handle both
when the telephone company’s switches can support digital connecticns.
rate traffic (data) easily.

6. To limit the queuing delays in internal buffers. Multiple Devices: ISDN allows multiple devices to share a single line. It is possible
ATM Signalling : Protocol: : many different digital data sources and have the information routed to the
to combine ination.
proper destina’ Since the line is digital,
gital, it it isi easier
i to keep the noise
i and interfere
i
Protocol consists of two parts nce
out while combining these signals.
-> User-Network Interface (UNI)
e defines how hosts talk to switches Signaling: Instead of the phone company sending a ring voltage signal to ring the
bell inof-your D A channelé
Rout Bandphone (“In-Band signal”), it sends a digital packet on a separate
slgaaifl
_» Network-Network Interface (NNI)
° defines hi how swi' itches talk to other switches Q.4.(c) What is Tunneling? @)
¢ Cell formats slightly different
ATM Signalling Operation:- : 4 Ans. Tunneling: also Op known as “port forwarding,” is the transmission of data
© Connection requests proceed hop-by-hop through the switches of the networken| intended for use only within a private, usually corporate network through a public
route to destination network in such a way that the routing nodes in the public network are unaware that the
Control (CAC ) based on traffic descripto,| transmission is part of a private network. Tunneling is generally done by encapsulating
¢ Switches perform Call : Admission : the private
i lable resources at that switch vas network
: data and protocol
, information within the publicpubli networ
‘twork
QOS ieee fad aver an Pee eaida
d on, otherwise “reject” js » transmission units so that the private network protocol information appears to the
¢ Ifconnection is acceptable, then req public network as data. Tunneling allows the use of the Internet, which is a public
returned ali eG * | network, to convey data on behalf of a private network.
¢ Ifdestination accepts
s
connection,
i
then “accept” 1s
One approach to tunneling is the Point-to-Point Tunneling Protocol (PPTP) developed
Example of ATM Signalling: by Microsoft and several other companies. The PPTP ee data ee hably
secure, even though part of the path(s) between or among end users exists in public
| communication channels. The PPTP makes it possible for authorized users to gain
‘| access to a private network - called’a virtual private network (VPN) -through an Internet
| service provider (ISP) or online service. Another commonly used tunneling protocol is
| generic routing encapsulation (GRE), developed by Cisco Systems. There are numerous,
less common tunneling protocols. :
Tunneling, and the use of a VPN, is not intended as a substitute for encryption/
CR : Connection Request
OK : Connection Accept decryption. In cases where a high level of security is necessary, the strongest possible
encryption should be used within the VPN itself, and tunneling should serve only asa
Q,4.(b). Write short notes on ISDN?
convenience.

digital phone connections which has been available for over a decade. This sys Q.5.(a) Given the network topology below use the Dijkstra’s algorithm to
allows voice and data to be transmitted simultaneously across the world using end Compute the shortest path from A to all other nodes. Make sure to show the
end digital connectivity. There are two basic types of ISDN service: Basic ‘Tesults of the computations at each step. - (6.5)
 twork
Sixth Se mester,
Computer Ne LP. University-[B.Tech.]-AB Publisher 2017-27
26-2017 i wait befo re retransmitting increases with the number of collisions occurring on the
work.
net GSMA is a network access method used on shared network topologies such as *
pthernet to control access to the network. Devices attached to the network cable listen
(carrier sense) before transmitting. If the channel is in use, devices wait before
pransmitting. MA (multiple access) indicates that many devices can connect to and
access to use the network when it is
ghare the same network. All devices have equal
clear. Even though devices attempt to sense whether the network is in use, there is a
ood chance that two; stations will attempt to access it at the same time. On large
time between one end of the cable and another is enough
net works, thez transmission
that one station may access the cable even though another has already just acc -ssed it.
Method for avoiding these collisions is as below:
«QSMA/CD (carrier sense multiple access/collision detection) C?, (collision
d evection) defines what happens when two devices sense a clear channel, then attempt
(w, ADE)
to transmit at the same time. A collision occurs, and both devices stop transmission,
used to access
wait for a random amount of time, then retransmit. This is the technique
er the edge AD.
so we will consid the 802-3 Ethernet network channel. This method handles collisions as they occur, but
smallest is (1, AD) drops
if the bus is constantly busy, collisions can occur so often that performance
Now we are at Node D drastically. It is estimated that network traffic must be less than 40 percent of the bus
capacity for the network to operate efficiently. If distances are long, time lags occur that
may result in inappropriate carrier sensing, and hence collisions.
Q.6.(a). Draw and explain packet format of Transmission control protocol?
Explain various steps that are followed in releasing a TCP connection? (8)
Ans. Refer Q.4.(a) Second Term 2016.
(2, ADC) (@, ADE) The basic steps followed by TCP to establish the connection are as follows :
(2, AC)
Release can be asymmetric(one sided, where data may loss ) or symmetric (both
take any node ADB
or ADC. sides agreed) Steps for symmetric Release:
Smallest is 2 so can
e Step 1: Client end system sends TCP FIN control segment to server
Let us consider C Node. ° Step 2: Server receives FIN, replies with ACK. Closes connection, sends FIN.
(1, AD)
(2, ADB) e Step 3: Client receives FIN, replies with ACK.
Enters “timed wait” - will respond with ACK to received FINs
° Step 4: server, receives ACK. Connection closed.

Cdeiient server ih
Close IN
D detect and correct collisi mn
Q.5. (b). What is collision? How does CSMA/C
(
Close
two or more compulél
Ans. In computer networking, a condition that occurs when
ns is one ofth
ona network try to transmit signals et the same time. Handling collisio
main functions of a networking access method. “a
timed wait
How Collision Works: When a station begins transmitting a signal and dei ect Closed
collision, the station stops transmitting and issues a jam signal to tell the other stall!
that a collision has occurred, Both stations then stop transmitting and wait a ra Cc
Symmetric Release
o
a

interval of time before retransmitting their signals. The amount of time the 5
 .
k
Computer Networ : j
LP. Universit y-(B-Tech At pun:
98-2017 :
Sixth Semes ster,
ference
et Sala 2017-29
a

* : pif
Connection releases: ee
Lea e
Host 2
nance
UD
Host 1 et |
f poe eer Leal OR q
Host 4 onnection] ess
me eOR tart timer is Unreliable packet deliv ,
[ Send DR -—~——— iS!
esaae .
2 start timer ot
get oat ian 1 fas flow com No flow control me eu
Send OR ferred for real time ew:
4 OS" om t pre Pre ferred for real tim Fags
oR = + start timer, _ - Not P € applications
| Release : | , plications
Rs ps a:
a ~ iconnection . | 7
Has considerable overhead Lightweight protocol
elease
° ‘
connection Eas ‘ol with less overhead
: FTP, HTTP ete pplication: DNS, DNS
| Ree
.:om oe Ap plication
send ACK ——__ACK
Se ee ease Send ACK _ ” Waive i Q.7.(a) Explain the Leaky bucket algorithm to cont
connectio | hea | (Timeout) | pow the drawbacks of this are overcome in token bck e > Shoal
orithms? nr acl
|
release © g
| oo 5
connection’ Ans. Leaky Bucket Algorithm: Consider a Bucket with asmall hole at the bottom,
Bee a — | natever may be the rate of water pouring into the bucket
i es ie tate at which water
(b) comes out from that small hole is constant. This scenario
ee entering it spi a epicted in figure (a). Once
the bucket is full,é any additional water Ting it spills over (i.e iti
theth)sides and isis lost lost (7.2.
(a) Host2 doesn't appear in the output stream through the hole undernea
Host 2 Host 1 a 3
wai
os! sul ———-}
SendDR — .-R idea ofof leak,
The same idea leaky bucket can be applied
e
as shown in Fig.b.
ee VBR DR |
Send timer ; Sa “|Li Send DR \ | conceptu ally each network interface contains eek ee
aSea ——t____ gstart it start timer
ucket. And the following steps
por Os) ho ee ~~ jastart timer) | re performed:
wt dans iiaan* ‘ i
eout). =kz
Sarees : |iy et : When the host has to send a packet, the packet is thrown into the bucket
. i
¢ The bucket leaks at a constant rate , Meaning . j i
& start timer. © i } (Timeout) jv o the network interf: ace transmits
——_. Send DR &; j ost
pec rie Jaa ts packets at a constant rate.
at start timer }
| cael peas rena. s ¢ Bursty traffic
‘
is converte . traffic by the|eaky bucket :
: d to a uniform
t i ° °
connection POY Ses . ° In practice the bucket is a finite queue that outputs at a finite rate
: 3 j .
et ees be built into the
arrangement can be simulated in the operating system or can
Send ACKGs -——__ACK c wi Release | (N Timeouts) Timea This
elease | | hardware. Implementation of this algorithm is easy and consists of a finiteeas
| fF tease |
in the queue it is queued up and
| connection,
jeonnection | connection} | Whenever a packet arrives, if there is room ‘ ee
: no room then the packet is discarded .
es © ce
case ofa
a connection. (a) The normal
Four protocol scenarios for releasing and sue a.
lost. (c) Resp onse lost. (d) Response lost
way handshake. (b) final ACK
lost.
DRs (Disconnection Requests) ie & :
protocols like T'
Q.6.(b). Why transport layer
Inflow may 7
a
between them? he bursty ae
end protocols. What is the difference
s whiche
protocols typically mean protocol
Ans. J assume end-to-end transport
ty from one end-point to another end-point. From
handling the full interconnectivi
port protocols are residing inl4
computer to Any Web server for instance. The trans
netwo rk model . ‘
4 in the OSI
The Transport Control Protocol (TCP) running on top of the Internet Protocol{l
often therfore called TCP/IP, is the most used of the end-to-end transport pro ooo
offers reliable transport with automatic resending of lost packets. Anothe isl
Datagram Protocol (UDP) which also is end-to-end but without reliability, no r
of lost data packets. ae
 _—

Network
LP, University-(B.Tech,|
Sixth Semester, Computer -AB Publish 4 er
30-2017 jadding the last fragment
case? é ’s data size. In the e. Xan]0ple
number of subnets in each above, thisi calculation was
Q.7.(b) What is the maximum ) 495°8 4540 = 4500 bytes. ae
(i) Class A: mask 255.255.192.0 when the receiver has all the fragments, it ¢ i
a 7 1v Can put t]
(ii) Class B: mask 255.255.192.0 ysin8 their‘ offsets. It can then pass their data up fee Zz = correct order, by
urther processing.
(iii) Class C: mask 255.255.192 The following table lists the important differenc es between [Py4
and IPv6.
(iv) Class C: mask 255.255.255.240 IPv4
(b) 4 IPv6
Ans. (a) 1024
(c) 4 (d) 16 jpv4 addr essesbitlength.
are 32 | IPv6, addresses ‘
IPV4 and IPV6? Draw and explain the IPy,
[pv4 addresses are binary IP yé odiceas are 128 bit length.
Q.8.(a) Difference between (8.5) represented in decimals. | represented in - are binary numbers
numbers
protocol packet format?
I _{pSec support is only optional. Inbuilt IPSec —
Ans. Fragmentatio ae
31 Fragmentation is done by sender
1619 nis done only by sender.
Q 4 82: and forwarding routers.
Packet :
IHL | Type of Service Total Length No packet flow identification. acks flow identifi
i ca tion
ion isi avail.i
within the IPv6 header ang oe
Flags Fragment Offset
Identification the Flow Label field. .
Protocol Header Checksum Checksum field is available in No checksum field in IPv6 header.
Time to Live
IPv4 header
Source IP Address Options fields are available in No opti
ption fields, but IPv6 Ex i
Destination IP Address IPv4 header. headers are available. a
Padding Address Resolution Protocol Address Resolution Protocol (ARP)
Options
(ARP) is available to map IPv4 is replaced with a function of
t, it examing addresses to MAC addresses. Neighbor Discovery Protocol (NDP).
hen a router receives a packe
Working of Ip: fragmentati onrW interfag
s and dete’ rmine s the outgoing interface to use. The Internet Group Management IGMP is replaced with Multicast
the destination addres th
size is bigger than the MTU, the router may fragment Protocol (IGMP) is used to manage Listener Discovery (MLD) messages.
has an MTU, If the packet multicast group membership.
packet.
of each segment is thy Broadcast messages are available. Broadcast message are not available.
into segments. The max size
The router divides the packet 60 bytes maxim um). The route Instead a link-local scope “All nodes”
(20 bytes minim um;
MTU minus the IP header size g following changes multicast IPv6 address (FF02::1) is
segme nt into its own packet , each fragment packet havin used for broadcast similar functionality.
puts each
segment size.
¢ The total length field is the Manual configuration (Static) of Auto-configuration of addresses is
last one, whichi
is set for all segment: s except the IPv4 addresses or DHCP -available.
¢ The more fragments (MF) flag
(Dynamic configuration) is
set to 0.—°*
in the
based on the offset of the segment required to configure IPv4
© The fragment offset field is si et,
d in units of eight- byte blocks. « The he addresses. :
easure
original data payload. This is m
bandwidth
checksum field is recomputed. Q.8. (b) Explain the maximum bit rate for a channel having
one of th
tha t a packet is a fragment if at least 3100Hz and S/N ratio of 20 dB.
Reassembly: A receiver knows i

following conditions is true: Ans. S/N = 20 dB

true for all fragments except the las te - 20 dB = 10log (S/N)
¢ The ‘more fragments” flag is set. (This is
field is nonzero. (This is true for all fragments except ti S/N = 100
° The “fragment offset’
first.) C = BW*log, (1+S/N)
The receiver identifies matching fragments using the
identification field. ™ = 3100*log, (1+100)
with the same iidentification H®
receiver will reassemble the data from fragments
the receiver
C = 3100log,, (101)Mog,,2)]
using both the fragment offset and the more fragments flag. When = 20, 643 BITS/SEC.
to 0), it can calcula’
the last fragment (which has the “more fragments’ flag set

—
offset by
length of the original data payload, by multiplying the last fragment’s
 >
Q.9. 6) Compare various frame retransmission techniques useq forg
FIRST TERM EXAMINATION [FEB. 2018]
Fey oO
:
correction in daca loink layer.t Request (ARQ): Automatic‘ Repeat request
fk Ae
(ARQ).
: SIXTH SEMESTER [B.TECH]
error control met
hod for data transmission
owledgthat
ment makes use of error-detectign nf
message s, and timeouts to ag o
tive ackn
is a mes' sage sent-by the receive COMPUTER NETWORK [ETCS-306]
reliable area
Se eeecke acknowledgment
data transmission. C accaraine? b: .

indicat
indicate that i ectly received
it has correctly a data
itter does not receive the acknowledgment before gime :: VA brs.
Pelee
Note: Attempt Q. No. 1 which is compulsory and any two other questions from
; MLM. : 50
mf
remaining
occursnee
Romar
timeout er eG
(7.e., within a yabie
eet
data fr ty \
after sending theecsia
amount of timecedianthoeman ions.

it cctee cunt the frame until it is either correctly receive: Pi ts Deyo a 1. (a) In the sliding window protocol , can the sender receive an ACK for
i
a predetermined num ber of retransmissions. k Pack NARGIS acket that falls outside its current window ? If yes, specify the scenario under
Three types of ARQ protocols are Stop-and-wait ARQ, Go-B: Q d Selec, hich this occurs as well. (2)
: y Ans. Outside can mean to either side of the window. The sender will
ARQARQ.
Repeat is appropriate if the communication channel has varying or ee not receive an
capaci ACK for a packet that is to the right of the window since the sender never sent such
a
Back as is the case on the Internet. However, ARQ requires the availa i‘ y of ah acket. The sender can potentially receive an ACK (duplicate) for packets outside of the
Seeeet results in possibly increased latency due to se eae aE an ola window to the left. This can happen when for example the original ACK got delayed
and
maintenance of buffers and timers for retransmissions, which in fa case oO} Net the sender retransmitted the packet on timeout.
congestion can put a strain on the server and overall network capacity. Q.1. (b) What is the difference between circuit
switching and Messa switching and packet
For example, ARQ is used on shortwave radio data links in the form of ARQE,| ge Switching.
combined with multiplexing as ARQ-M. Ans. Circuit-Switching, Message-Swit (2)
ching and Packet-Switching
Error-correcting code: An error-correcting code (ECC) or forward error corregi;, Circuit switching network establishes a
fixed bandwidth circuit (channel) between
FEC) code is a process of adding redundant data, or parity data, to a message, ay nodes before the users may communicate, as if the nodes were physical
that it can be recovered by a receiver even when a number of errors (up to the capab Ii with an electrical circuit. The bit delay is constant ly connected
during the connection, as opposed to
of the code being used) were introduced, either during the process of transmission, oy packet switching, where packet queues may cause
varying delay.
storage. Since the receiver does not have to ask the sender for retransmission oft Each circuit cannot be used by other
callers until the circuit is released and
data, a backchannel is not required in forward error correction, and it is therefore suit a new
for simplex communication such as broadcasting. Error-correcting codes are frequer
used in lower-layer communication, as well as for reliable storage in media such as ()
DVDs, hard disks, and RAM. \ Virtual circuit switching is a packet
switching technology that may emulat
Error-correcting codes are usually distinguished between convolutional codes Fe switching, in the sense that the e eaeuit
connection is established before
transferred, and that packets are any packets are
block codes: ‘ delivered in order.
* Convolutional codes are processed on a bit-by-bit basis. They are parti Message switching was the
precursor of packet switching,
routed in their entirety, one hop where messages were
suitable for implementation in hardware, and the Viterbi decoder allows opt at a time. It was first introduced
decoding. in 1961. Nowadays,
Fe
* Block codes are processed on a block-by-block basis. Early examples
of block
are repetition codes, Hamming codes and multidimensional
parity-check codes,
were followed by a number of efficient codes, Reed—Solomon codes being ted between nodes over data
the most no term “packets” refers to the links shared with other traffi
due to their current widespread use. Turbo codes and fact that the data stream c. The
low-density parity-check into packets of about 200 from your computer is
(LDPC) are relatively new constructions that can provide bytes (on average), which broken up
almost optimal effici |) Each packet contains a are the: n sent out onto the
“header” with information network.
Hybrid schemes: Hybrid ARQ is a combination of necess
ARQ and forward error corre¢t
There are two basic approaches:
fe
~ Messages are always transmitted with communications lines.
FEC parity data (and error-de' The
redundancy). A receiver decodes a message
using the parity information, and re { allowing for very efficient use of packets from many different sources can share a
retransmission using ARQ only if the B8enerally accepted th e fixed capacityi . With line >
parity data was not sufficient for onto the network on curre! nt technology, Packet
suce becomes overloaded, a firs
Piet-co me, s ar
decoding (identified through a failed
integrity check), packets are delayed s firs t-se rved basis. If the metwork
or discarded (“droppe
° Messages are transmitted Q.1. (c) Give two adv d”).
without parity data (only antages of using opti
information). If a receiver detec with error-d cal fibre cable com
ts an error, it requests FEC pared to Coaxial
transmitter using ARQ,
and uses it to reconstruct
information from! The c gah ett, and
ee
the original message, other electronici devi i
The latter approach ig in the form of signals ces transmiti the data
particularly att; ractii and using a transmis from o
rateless erasure code, ve on sion media
f - The tran
smission
ne
ime

ERT
ae *

RUT
 Network
ster, CO mptputer
Co ' : LP, University-(B Tec
Sixth Seme p U
h|-Akash Books 2018-3
2-2018 d and unguided, bi both data frame
into two types guide ‘ identification
.... identi
fot
purposes,
purpos es,
¢ rise io hh carries electromagnetic waves by male
ego i 3 S
nd am entally cat poe ACK frames are numbered
media cat n be fu
ica tio n whi c ysical conduct or. Guided medi
tornate I. A data
and gotten
ly 0 has I frame is acknowledged y an ACK| frame, indicating that
media is a wirel
ess commun
uirjriin:ng & a } phi
di media is classifj
a
re ceiver
alte” e tificat : data I 3 and is now expecti
aT nd wi thout
it req requ
Guide’ ye
the ion of data frames in case of duplicate te an
fair as medium s ©such as W res
sigtnatals article explains
to ran
tra smitit sig
nsm
re-optic cab le. The
m Lost Acknowledgment In this case, the data a oe
physical medium coa xia l cable and fib e
ca ble, i
aways twisted pair coaxial cable. etig, }
op tical fibre and ries ¢ ectromagn
difference between tion which car conductor, A
1s a wire Jess communica @ phy sic al
Unguided media uiring
and without req such as wires. Gui
ded
air as mediv im
by making use of medium to tr ans
mit signé als optic cable. Th receiver accepts th new copy and
a physical le and fib re- ¢ the appropriate ACK (assumin, ig the copy © arriv:
media need cable ble, coaxial cab es
on lost frame
ways twisted pair re and coaxial
cable. the new copy je wdasel a
classified in three between Opus al fib was aD ae ie ie 4 ecognizes Picate, acknowledges its
you the difference sues
whi ch tra nsm its the signal eceipt, then discards it and waits for the n
explains guided me dia t ext frame (see Figure)
Essentially, the
optical fibre isa as coaxial cable . Sender
(optical form). Receiver
device to another in the form of light :
one }
al form.
the signals in electric Fibre and Coaxial
Gaple
Between Optical cable
Key Differences in opti cal form while coaxial
ies the sign als ;
1. ‘Optical fibre carr ie
ricit y.
signal in the form of elect fibre and plastic. In contrast,
the co; Time out ars
is made of glas s Lost
2. Fibre optics cable l mesh braid. ‘
:
(copper), plastic and meta
is made up of metal wire e as it has higher noise imy First copy
3. The optical fibre is more
effic ient than coax ial cabl
i .—
coaxial cable.
4. Optical cable is co: stlier than cal fibre. As a
of bendi ng of the cable is negati ve in case ofan opti Time ; 'e
5. The effect
g.
the coaxial cable is unaffected by the bendin con Fig. ils Stop-and-w
idth and data rates. On the
: ait ARQ 7 lost A CK frame
6. The optical fibre provid es high bandw
ioral m 1 1 E
high bu aright
the coaxial cable are moderately
1 I I
bandwidth and data rates provi ded by
than optical cable.
corrupted in transit,
BS a NAK frame isi return
s installation of optical cable req numbered, tell the sender to retransmit he . NAK frames, which are not
7. Coaxial cable can be easily install ed wherea frame sent. NAK frames
extra effort and care.
g are not numbered because it tells the sen der ast
to retransmit
a small diameter. Conversely, aq send. Stop-and-wait ARQ requires that th the last frame
8. The optical fibre is lightweight and has ‘ e sender wait until it receives an
cable is heavier and has a large diameter. acknowledgment for the last frame transmi
OR
. tte before it transmits the next
mitted
one. When 7 the sending device receiv. es
to coaxial cable are: a NAK, it resends the frame
The advantage of using optical fiber cable compared transmitted after the last acknowledgme
than copper Wir nt >
Thinner: Optical fibers can be drawn to smaller diameters Q.1. (ec) Abinary signal is sent over
than coj a3 ee zs
10 is 20dB. Calculate the maximum achievable datae
Higher carrying capacity: Because optical fibers are thinner :
> signal-to-noise
copper wires.
more fibers can be bundled into a given-diameter cable than (2)
more phone lines to go over the same cable or more channels to come
thro 3 Ans. 3KHZ channel : ee
into your cable TV box.
Less signal degradation: The loss of signal in optical fiber is less than in co}
Light signals: Unlike electrical signals in copper wires, light signals fro
do not interfere with those of other fibers in the same cable. This means
coverations of TV reception.
; Q1. (d) Instop and wait ARQ, what happens if a negative ackn
is lost in transit? A
Ans. Biop and aay ARQ is a form of stop-and-wait flow control ext
retransmissiv. of data in case of lost or damaged frames, For retrans:
four oe are added to the basic flow control mechanism: 4 ‘a
i Br
mi ee nik 2.600% of the ines frame transmitted until
e sending devi:
dat peed
 twork ‘ :
. LP. University-[B.Tech]-Akash Books igo
gixth Semester Computer Ne oss
1x! » ? z 3
is (7)

= i Number ple links required for © ae Sl suite correlate to th oll management: determines when the node “has the right” to use the
Bite ers of the TCP/IP P! t 1 gical me! : Thi t;
” subnet, deciding
etwork peli tiie sate abioalc/ei tered eee Mais of the priority
ox ar ae! OSI) model has seven layers, This, ork conditions, of service,
p) map
(the wae whieh lee factors. It provides.
iaAns. tie Open Systems
: nning Wi the). ‘lowest’
Intereonnect The layers are hierarchy
in the stacked this dot
Jains them, begi |an ing: routes frames amon;
describes and exp lication.
<pighest’ (the @PP :
a houe
- BB ubnet a
3 :
O eae
traffic control: routers (network la yer intermediate
to the
and proceeding systems) can
ruct 2 sending station to “throttle back” its frame transmission when the router’
e Application s
* Presentation
n: if it determines that a downs itream
fragmentatio
e Session 5; ay. % i
« Transport ge oe unit (MTU) size is less than the frame size, a router eae
a nsmission < cuitses ia
Saal \ br transmission and re-assembly at the destination station. ee Pe
» Data Link + Logical-physical address mapping: translates logical addresses, or names, into
physical addresses. ’ '
* Physical “he physical layer, the Jowest layer of the OSI model, is concen,
Physical layer: The physi a y ethe unstructured raw bit stream over a phy, « Subnet usage accounting: has accounting functions to keep track of frames
fo rwarded by eubney intermediate systems, to produce billing iiformation.
with the transmission Se aceslietantl: mechanical, and functional interfaces,
medium. It describes re e ae the signals for all of the higher layers. It provides; Communications Subnet: The network layer software must build headers so that
physical medium, an woes . Be ple digital signal pattern (is and 0s) usedby! the network layer software residing in the subnet intermediate systems can recognize
. ing: modifies th sical medium, and to aid), them and use them to route data to the destination address.
aI TR the characteristics of the phy
PC tosaat : This layer relieves the upper layers of the need to know anythi bo:
See
and frame synchronization. It transmission and intermediate switching technologies used pane hele
inary : establishes, maintains
. t signal state represents a and terminates connections across the intervening
communicat ea
starts
. ms anos station knows when a “bit-time” facility (one or several intermediate systems in the communication subnet). od
a frame
« How the receiving station delimits possib ilitie s i In the e network
> layer and the layers
E below, f peer protocols exist
i betws ;
accommo: dating vario us
¢ Physical medium attachment, its immediate neighbor, but the neighbour may be a node through which asta rane
: not the destination station. The source and destination stati :
medium. the medium ?
be used to connect to
« Will an external transceiver (MAU) many intermediate systems. ; AE dad
and what is each pin used for?
¢ How many pins do the connectors have Transport layer: The 1 transport layer ensures that mess ages are deli i -
r the encoded bits will be tranm free, in sequence, and with no losses or duplications. It relieves the higher
« Transmission technique: determines whethe a
et
g. from any concern with the transfer of data between them and their peers
by baseband (digital) or broadband (analog) signalin
electrical or optical
* Physical medium transmission: transmits bits as The:size and complexity of a transport protocol de ‘pends on the of: : ice i
appropriate for the physical medium, and determines. get from the network layer. For a reliable network layer with virtual agence
*¢ What physical medium options can be used a minimal transport layer is required. If the network layer is unreliable and/or only eeprexts
datagrams, the transport protocol should include extensive error detection
¢ How many volts/db should be used to represent a given signal state, usi and recovery.
i The transport layer provides:
physical medium
Data link layer: The data link layer provides error-free transfer of da' ; ° Message segmentat
: ion: g Acce; pts a message from the (session)
i layer above it.
from one node to another over the physical layer, allowing layers above it to! Be ae a smaller units (if not already small enough) ae;passes a
virtually error-free transmission over the link. To do this, the data link layer pr BE ccerbles‘its tho
downaes
to the network layer.
ye} Th e transport layer at the destinatio
stinati n station i
* Link establishment and termination: establishes and terminates thel
j
between two nodes. ° Message
Me iwied ackn owledgment: 2 Provides
ae ere: : 5
reliable :
end-to-end message delivery with
¢ Frame traffic contro]: tells the transmitting node to “back-off” when
4
buffers are available. Bessize
° Messag
butts ares control
e traffic ie : Tells s th the transmitting
it i station
i to “back-of
) f’ when no
* Frame sequencing: transmits/receives frames sequentially.
* Frame acknowledgment: provides/expects frame acknowledgments. * Session multiplexing: Multiplexes several message streams, : or sessisions on to
recovers from errors: that occur in the physical la: yer by retransmitti
it i one logical li
frames and handling duplicate frame receipt, y nitting mae - Bloyer),logical
° | ink and keeps track of which
; .
messages belong to which 7
sessions (see session
t i delimiting: creates and recognizes frame boundaries.
: Typically, the transport layer can accept relatively large messages, but there are
rame error checking: checks received frames for integrity. i ct message size limits imposed by the network (or lower) layer. Consequently,
the
 F ‘ : ‘Ke LP. Univers;
6-2018 Sixth Semester, Computer Networ nversity-[B.Tech]- Akash Book
transport layer must break up the mess ages into smaller
units, or frames, Prepeng yvercome er ttenuation caused by free-spa 3
a header to each frame. Np » Joss. A series OF repeaters make possible hacks :
oe vaters remove the unwanted ic-field divergence
dj
The transport layer header information must then include Die ’ Ved i
no; extension ofa sign
»Such noise in
the ranspet a © other i the original digital signal, eyon j a
as message start and message end flags, to enable sion,
3 og transmis
to recognize message boundaries. In addition, if the masta 0 ea maint, d. With anal gnals are
sequence, the transport header must contain sequence ae enable the , y ise ag w
e right Orde J inately also amplif no
transport layer on the receiving end to get the pieces back togethe
before handing the received message up to the layer she < a
diss} pal
layers whose ol is bety, jog si6 ; nt
-to-E rers: Unlike the lower “subnet”
ee ies, the transport layer and the layers above are true “sou ar wpically ee oe pe 6,000 meter intervals - > digital sj
or end-to-end layers, and are not concerned with the details Of th,
to destination” 9, Ina wire conta tino system, s
underlying communications facility. Transport layer software (and software above if amplifier 4 aoe iat anaes ator, and two antennas. The ¢ an aradio Teceiver,
on the source station carries on a conversation with similar software on the destinatig, gnal oD a freque He) a mi na from the Teceived signal, This s Tansmitter produces a
station by using message headers and control messages. jg necessary Fada ae see transmitted signal from deca frequency offset,
Session Layer: The session layer allows session establishment between Process, f jsolator provi Protection in this respec Ing the receiver. The
jocate : He
running on different stations. It provides:
* Session establishment, maintenance and termination: Allows two applicatia a “a me ions over q
processes on different machines to establish, use and terminate a connection, called a SeSsiq be ‘ in satellite wireless, a repeater (more ite
¢ Session support: Performs the functions that allow these processes to communigy, yplink signals and retransmits them, often on diffe y called a trans Pon
feations. : rent frequencie der) receives
over the network, performing security, name recognition, logging, and so on. 8, to
a reveater fg ss destination
Presentation layer: The presentation layer formats the data to be presented), 4. Ina cellular telephone system,
the application layer. It can be viewed as the translator for the network. This layer my geographi c area that collectively serve a et eof a grou; . f
stemuser, SF P of transceivers in a
translate data from a format used by the application layer into a common format at th
sending station, then translate the common format to a format known to the applicatig,
layer at the receiving station. a

The presentation layer provides: i

‘ j : !
* Character code translation: For example, ASCII to EBCDIC. } requires careful attention to ensure that internal circuit 2 Sign
* Data conversion: Bit order, CR-CR/LF, integer-floating point, and so on,
* Data compression: Reduces the number of bits that need to be transmitteq,) extend signals in the radio frequency ran
the network.
¢ Data encryption: Encrypt data for security purposes. For example, Passwoy and retransmit signals from and to a number Pate of ub repeaters,
encryption. ; 7. : A bus repeater
oS sear s : a links one computer bus to i
a bus in anoth 3
Application layer: The application layer serves as the window for user. essentially chaining one computer to another. PE: €r computer chassis,
application processes to access network services. This layer
contains a variety) _ (ii) Gateway: A gateway is a network that connects two networks
commonl
pe y needed function
ag s: ache 3 _ | different protocols together. While a brid ge 1s node
used
using
to join two similar types of
Meat cee | agateway is used to join two dissimil networks
ar networks. ~
A gateway device provides comm unic
ation to a remote network
system that is out of hounds for the or an autonomous
* Remote printer access host netw
e ork nodes. Gateways serve as
| and exit point ofa network; all data the entry
- * Inter-process communication routed inward or outward must
and communicate with the gateway first pass through
¢ Network management in order to use routing paths. Gener
configured to work as a gateway ally, a router is
de vice in computer networks.
* Directory services A network gateway joins :
* Electronic messaging (such as mail) t wo networks so the devices on one
E
a communicate with the devi ces on network can
another network. - Without gateways, you couldn’
3
* Network virtual terminals ie F : » YO uli
i ple to access the internet, communicate and send data Back at POA
Q.2. (c) Briefly discuss the role of following devices in context of networkin) vanaieen
©etwork
™Plemented
gat, completelyy} in software :
(i) Repeater (ii) Gateway ire, hardware, ora combination of both. Because a
i Ach as mecca by definition appears at the edge of a network, related capabilities
Ans. (i) Repeater: 1. In digital communication systems, a repeater is a dev f ; and proxy servers tend to be integrated with it.
i Ree of Gateways for Homes and Small
receives a digital signal on an electromagnetic or optical transmission med Businesses
regenerates the signal along the next leg of the medium. In electromagne ;‘ ichever type of network gateway you use in your home or small business, the -
 _—
8-2018 Sixth Semester, Computer Network
function is the same. It connects your local area network (LAN) and all the dey;
to the internet and from there to wherever the devices want to go. Types 9 Q.3- (©) What is CSMA?How
gateways in use include: f nett p- Persis
tent Cs
¢ On home networks and in small businesses, a broadband router typically ee ersistent ? ier S d Mult; MA is diff
erent fr
ans; Carmen vense “iple Suchs ag(C a :; om non
ed on shared network topologies Acces
as the network gateway. It connects the devices in your home or small businegg 4.,,'"%
internet. A gateway is the most important feature of a router. Routers are the mt isan,
ust
vices attached to the network cable listen (nes to co val fo (2)
common type of gateways.
Sat pannel is in use, devices wait before tr, access method
¢ In some cases, such as in a residence that uses dial-up internet access, the ga; ey 4 s e eee Nse) hefo
itting. MA (Multip), ae the network.
devices can connect to and share th «re transmitt i
is a router at the internet service provider's location. This has become increasingly },! mitting. If the
common as dial-up access declines in popularity. % Ww as the network when it is clear.
ea © same network. Al] eae
* Some small businesses configure a computer to serve as the gateway ta thy ‘aveaudicates that
equal accegg
internet, rather than use a router. This method requires two network adapterg 26
connected to the local network and one connected to the internet. ne
Gateways as Protocol Converters: Gateways are network protocol conve
Often the two networks that a gateway joins use different base protocols. The Batey,,
facilitates compatibility between the two protocols. Depending on the types of Protogy
they support, network gateways can operate at any level of the OSI model.
Q.3. (a) An 8 bit byte binary value 10101111 is to be encoded usin: € AN eve,
parity Hamming code. What is the binary value after encoding ?
Ans. The given 8 bit byte with a binary value of 10101111 yields
(8) Sharing tim
a parity value y e with CSMA met
101001001111. Lv ped to overcome hod.
the chanceofs col the
This is how you solve it:
lision So as to imp inAl 0
principle of ‘carrier rove thepat minimize
sense’. The statio a ican A protoc
1. Mark all bit positions that are powers of two as parity bits. n Senses the carrie ol h
8, 16, 32, 64,etc.)
(positions 1, a
state of chann
r or Se fore po ei
Even though devices attempt to Spree: e it a
2. All other bit positions are for the data to be encoded. (positions Sense whether ne aaeerit ile
3, 5, 6, ut chance that two stations will attem busy.
10, 11, 12, 13; 14, 15, 17 etc). sae tarmegers pt to access itj al perWor k
the transmission time between one end
of the
: is in use, there is a 5 good
—— it — O10 —4 411 eee On large networks,
ale ee access the cable even though anothe
3. Each parity bit calculates the parity for some of the bits in the code r has already ieee fot one
rei e chances of collision still
The position of the parity bit determines the sequence of bits i fs exist because
that} transmitted by one station takes Some time to reachof ae ecuei n meee re
alternately checks and skips. other stations may sense the channel me
Position 1: check 1 bit, skip 1 bit, check 1 bit, skip 1 bit, etc. in the collisio n. i it thes . : me,
“2 boidle and transmit their frames. This results
(2/3,,054, 9st 3,5155-5)) There Are Three Differen
t Type of CSMA Protocol
Position 2: Check 2 bits, skip 2 bits, check 2 bits, skip 2
bits, ete. () I-persis tent COMA s
(2, 3, 6, 7, 10, 11, 14, 15, ...) (ti) Non- Persistent CSM
Position 4: Check 4 bits, skip 4 bits, check 4 bits, skip
A
4 bits, etc. (iit) p-persistent CSMA
(4, 5, 6, 7, 12, 18, 14, 15, 20, 21, 22; 23...)
Position 8: Check 8 bits, skip 8 bits, check 8 bits, skip 8 bits, etc.
(8-15, 24-31. 40
— 47, ...) a
Position 16: Check 16 bits, skip 16 bits, check 16 bits,
skip 16 bits, ete.
(16— 31, 48 - 63, 80 - 95, ...) ;
non: Persisiont
Position 32: Check 32 bits, skip 32 bits, check 32 : :
bits, skip 32 bits Ty f CSMA
(32 — 63, 96 — 127, 160 - 191, ...) ete. @ Lpersistent CSMA
26 * In this method Oe ;
channel to check whet! i
Station
ether that lwants
the channe is idleto orbusy.
transmit datae sor
° Ifthe channe’ 1 is busy, the statio 3 oUsly. pce
n waits until it becomes idle.
Doing the calculations: Position 1 yields a value
of 3, which isoad and therefore! mith en the station detects an idle-channel, it immediately transmits the frame
Positions 2, 4 and 8 equal 4,2and4r
espectively. So they each get a value of 0. Fil _ Probability 1. Hence it is called I-persistent
CSMA.
them in gets you: F fing * This method has the hi :
cha ! ighest chance of collision because two or more stations may
Q.3. (b) How throughput is impr nnel to be idle at the same time and transmit their frames.
oved in slotted ALOHA over pure/ ;
Ans. Refer Q.2. (c) of First Term When the collision occurs, the stations wait a random amount of time
Examination 2017. h
| , allover a and start
gain. e Zi

neers:
 he

er Network a University-[B.Tech|-Akash Books

Sixth Semester, Comput 2018=it
10-2018
suppose, j j (ii) p-persistent ae ;
affects this protocol, = Let us ‘ dis used when ch
Drawback Se time greatlyly allec Just ane
ae has time slots Such that the ti
e The propa gatio n delay ot a es aé data - This spate ace ion delay time e time
aximum Propagat slot duration
, station 2 alsotyeteerea : a
pee its transmissionI sign allNon ae equalWhtoenev ole
°F er a station becomes ready to send,
the sear station ! al ]has not smi result incalls! , it Senses the
j
If the ssio n, This will
senses the channel. begin its tr | channel
to be idle and will
sense the channel transmit
Sense & olf channel is busy,| station waits until next slot
with a probability
Continuously sense pannel is idle, it transmits : D.
0 Ife
re Bu!sy 4 n then wai
th statio
Idle F the pro! bability Glq =/—p, P, the
9 With
aits sce the beginning of the next
Station can transmit Ese slot.
Busy channel idle channel f
Ifthe ne xt slot is also idle, , iitt either
ei transmits‘ or waits again with
BT CSMA Coa
|-persistent : ‘|
probabilities p
statio: ns b | 0
eis zero, collision ; will still occur.j If two : i ; : :
if propagation delay tim wait unti ;
‘en mission, bothi statiaSons will fae and 9:
o This process is a repeated till either frame
has been sites
readyRat in an anad e of thi ind station's trans
the middl
:
begin their transmission
exaelly i ; a or another station 4
ends and : then eeboth will mitti ng.
Hi
firs + stati on j|pos begud trans
trans: missi on of
collis ion. :
simultaneously. This will
also result in
se of the transmissio n by another station, 6the station acts as thoug ughh a
the cha o Inca
Gi) Non-pPersistent CSMA a frame and a
it finds that
t
collisi
Pa on has c u
occur r e d re
rand
and it waitsa rando m amoun t of time and starts agai

UR
fr
d interval of time |
nsmi b a : ; gain.
station wants to tra
« Inthis scheme, ifa
on is transm itting)
then it will wait for fixe issfree_
is busy (some other stati and if the chan nel
channel
checks the status of the
e After this time, it again i
it will tran smit . nne l.
ses the cha
« Astation that has a frame to send sen
it sends immediately.
e Ifthe channel is idle, senses the
a random amount of time and then
¢ If the channel is busy, it waits . - Continuoush

WW)
channel again. y chann el;
inuo usly sens e the
statio: n does not cont
e In non-persistent CSMA the smission.
it detects the end 0: f previous tran
the purpose of capturing it when fea
Advantage of non-persiste nt
stati ons wait a rand om amoun tof
sion because the
* It reduces the chance of colli of time andy
stations will wait for same amount
time. It is unlikely that two or more i
retransm it at the same time.
nt
Disadvantage of non-persiste
the channel remains idle when
° It reduces the efficiency of network because
due to the fact that the stations waita
may be stations with frames to send. This is (a
random amount of time after the collision.
Non-Persistent ~

Station can transmit

Sense and transmit 7

‘Mes: . .
Sense Sense message is 10011010?express answer
Wer | rer pis. Generated polynomial x3 +22
Given message, 10011010 se
Idle cha _ ORC calculation:
Busy channel me)
 —
12-2018 Sixth Semester, Computer Network
11111

4101)10011010 END TERM EXAMINATION

11014
1001 SIXTH SEMESTER [
1101
4000 COMPUTER NETWORK [B.TEC
TCS. 206
1104
101
4011 por Attempt five question
vote tion inin all allinclud
including Q.No:1 Which is oo M.M.:75
110 4 peach unit. : ™Mpulsory select one question
4100
110 fn qi Attempt the following questions;.
110 1

<= reine wie disadvantage of wing layered pegs

What are two reasons for ys}
1? What the ong
CRC generated Ie Ans. A layered protocol architecture provides 4 Conceptual
At receiver side, data received is 10011011. 2
the complex task of exchanging information between rene heamework for dividing
41111 « Each protocol layer has a narrowly defined reaoeine into simpler tasks,
a) 1)10011041
.+ Aprotocol layer provides
Consequently :
Senge a the
standard interface to.
deel th a; ‘ 3
11011 ae next higher protocol layer.

infrastructure. nderlying physical network

ra101 | e Benefit: The same user-level] (application) pro,
1000 communication networks. Program can be used over very diverse
1101 ¢ Example: The same WWW browser can be used wh
1011 internet via a LAN or a dial-up line. en you are connected to the
1101 Some of the disadvantages of using layered protocols are Ae todas
1101 lower performance. dancy and overall

sel
ON Re tpl aoe everything from remote and wired to connectionless and
0000 association situated
; application correspondence, with each ri dleheg
; . Tinger and shriek of BGP,
Sena ait MPLS, multicast steering and so on would be a gigantic endeavor and potentially
The remainder obtained is 000, it means, there is no error. unmanageable in this solid across the board convention. I could envision specialty utilize
Q.4. (b) How is selective repeat better than Go-bacKN. Explain ? (§)| situations where most extreme execution is im perative most importantly
Ans. Go-Back-N ARQ is a more efficient use of a connection than Stop-and-wait where an
exceptionally basic framework might be wanted.
:
ARQ, since unlike waiting for an acknowledgement for each packet, the connection: Q.1. (b) Compare and contrast CSMA/CD with CSMA/CA.
(5)
still being utilized as packets are being sent. In other words, during the time that would Ans. CSMA/CD (Carrier Sense multiple Access Collision Detection): This
otherwise be spent waiting, more packets are being sent. However, this method aly multiple access method is used in wired networks since it is possible to detect collision
results in sending frames multiple times - if any frame was lost er damaged, or theACk and then proceed with so used in LANs and WANs. ; :
acknowledging them was lost or damaged, then that frame and all following framesin This is used by IEEE 802.3 standard Ethernet networks in which each node monitors
the window (even if they were received without error) will be re-sent. To avoid this| - the traffic in the line and if no traffic is available then a particular node can transmit.
Selective Repeat ARQ can be used. But at the same time if two are trying to transmit then it is known as a collision. This
Go-Back-N ARQ situation is sensed by all the nodes in the given network. After that the stations which
1. Retransmission begins with the last unacknowledged frame event! had the collision will try to send data again after some random time interval which are
subsequent frames have arrived correctly duplicate frames are discardet vary for each node. If again a collision takes place the random time taken is increased
and waits again. This is the procedure used in CSMA/CD networks
2. Go-back-n ARQ -- Receiver must get Frames in correct and the method does
order not have any deterministic capability.
Selective Repeat ARQ 4 a
CSMA/CA (Carrier Sense multiple Access Collision Avoidance): This isthe
3. Only the unacknowledged frame is retransmitted.
4. It may be (slightly) more efficient than Go-back-n multiple access scheme used in layer 2 access methodin which the following method is
ARQ, but also much 'sed when nodes are trying to transmit simultaneously in a shared network. Fee
complicated. : ee node which wishes to transmit first has to listen to the medium for a pre determined
5. Selective repeat ARQ -- correctly-received
out-of-order Frames are 8 : rid to assess the channel state. If the channel is idle them the nodeis capable of
Receiver until they can be re-assembled into correct Transmitting. Else the channel is said to be busy and node has to wait until the channe
order
_ Somtoes
idle mode.

aaa
14-2018 Sixth Semester, Computer Network D LP. University
-1B Techy ‘Al
Aka: Sh B
This is implemented in IEEE 802.11 wireless LANs and other Wireless é approaches can be used to prevent congestig ‘Oks
and this is preferred since the wireless networks cannot detect Collision Work, spyer ceiver should send acknowledgement a Te a
ghe re 2018_15
transmitting like wired networks. So the implementation of CSMA/CA will tine ile ement for a single packet. The Teceivey sho
1petkets rath dement
the packet dropping in wireless networks. ; Prov, gckn0 wieto sent a packeta i
or a timer xpi
expires ld sen,
-+ pas setts Pah : ida aq
er ndi
‘aN seng:
CSMA/CD Vs CSMA/CA jfit # Admission Policy: In admission Policy g iscdia "ovledemers ie
1. CSMA/CD is used in wired LANs and CSMA/CA used in wireless LANs and of
types of wireless networks.
1 ation Switches
pita ina ai Ree
: first check
( the hae se j
x hey
2. CSMA/CD is standardized in IEEE 802.3 and CSMA/CA is standardized inty flo jewestion
pefore inMET ate routes Shonen
the network, uld deny 8establishing
4 chance op, « Wireme,
Congestion (to prevent
«- @Detwork
802.11. Ry ompeg
eventt
r SS eee
the above policies are adopted to Prevent congestion
6 8 avirtual Network there is: 9
\ 5
3. CSMA/CD will not take steps to prevent transmission collision un
place whilst CSMA/CA will take actions not to take place any collision si nce
til ibis ta j Connection
the lati york.
wor! ‘ : fore ith
j ;
has no means of knowing whether a collision has taken place.
P loaed Loop Congestion Control “Pens in the
Q.1. (c) What is the relationship between SONET and SDH ? (5 Closed loop congestion
control technique is
Ans. Synchronous Optic Network (SONET) and Synchronous Digital Hierarg, afer it happens. Several tech Used to treat
niques are used by
(SDH) are standards used in fiber optic networks that power large telephone and 1, Backpressure: Backpres different .T alleviate
Inte; y sure ig q technique in Rae Som
networks. SONET networks are deployed in North America; SDH networks are deplo receiving packet from upstream node. This may eaac
everywhere else. The data communications industry uses the concept Ye ein ch a co Bested node ot,
of "backbon i Pore congested and rejects receiving data from above ia
od Sto}
refer to a large network capable of carrying heavy loads of traffic. SONET tonode congestion cont
and SDH Rt. rol technique that Prop
optic networks, although expensive, are ideal backbone networks, agate in th seen
offering hi &h speci, flow. The backpressure techniqu “Pressure ig anod
and reliability. has information of its abov
e can be applied only to ae posite directi
e upstream node. ts
SONET adheres to standards T1.105 and T1.106 developed by ANSI +
or theAmeri In above diagram the 3rd node
National Standards Institute. SDH is an ITU or Intern is con; ested and Stops receiv;
ational Telecommunicatig,
Union standard developed to allow it to work with
SON KT line rates. It is a flexit),
a
owing down of the
e)
le Output outya ket 28a result
standard that connects to several interfaces and band: the source to
widths and has bandwidth slow eee =
demand capability.
Q.1. (d) What is the difference between open-loop congestion
contd
closed -loop congestion control?
Ans. Open Loop Congestion Control
6) at each ofits output lines. whenever th
| which is set by the administrator,
Open loop congestion control policies are applied to prevent congestion the
before | giving it a feedback to reduce the t,
happens. The congestion control is handled either by the source
or the destinai tion, | packets has traveled are not
Policies adopted by open loop congestion control - warned about congestion,
3. Implicit Signaling:
1. Retransmission Policy: It is the policy in which retransmi In implicit signaling,
ssion of the packeh the congested nodes and the source. there is no communication
are taken care. If the sender feels that a sent packet is lost The Source guesses that betw,
or corrupted, the pack network. For example when there is co stio .
needs to be retransmitted. This transmission may increase the sender sends se veral packe
congestion in the net for a while, one a ssumptio ts and there 1 is noackno w!
To prevent congestion, retransmission timers must be designed
n is that there i ae
to prevent congesti S a congestion. . i senent
and also able to optimize efficiency. 4. Explicit Signaling: In explicit
: exp] icitly sends signaling, ifa node experiences ree
a‘ packe t to the source or destination k can
2. Window Policy: The type of window at the sender side difference between to inform about congestion. The
may also affe ch oke packet and explicit signal
congestion. Several packets in the Go-back-n window are resent, the packets that ing is that the sig is include
nal
although some carry data rather than creating d in
may be received successfully at the receiver side. This
du’ plication may incre packet technique different packet as in case of choke
congestion in the network and making it worse,
Therefore, Selective repeat window should be ad opted as it sends the speci pale Signaling can occur in either forward or backward direction.
that may have been lost. conge stion, The donne: In forward signaling signal is sent inthe direction : of the
3. Di varding Policy: A good discarding policy adopted Policiee pene warned about congestion. The reciever in this case adopt
by the routersii the
routers may prevent congestion and at the same time
partially discards the cc ther congestion, oes : Ba
Of the eo: ae o Rignaling: In backward signaling signalis senti
the opposite
in direction
wi . B
or less sensitive package and also able to maintain
the quality of a message.
In case of audio file transmission, routers can discard less
Q.1.() Are} © source is warned about congestion and it needs to slow down. ei
sensitive | ‘both UDP and IP unreliable to the same degree? Why
prevent congestion and also maintain the ‘why not?
quality of the audio file.
4, Acknowledgment Policy: Since acknowl
in network, the
edgement are also the Pp
acknowledgment policy imposed by the receiver may
a
thom boca oese layer 4 protocol and IP is layer
mS.
aie
3 of TCP/IP model. You cannot bike
ie
o
also affect hey serve r different purposes. And anyways,
protocols
 Sixth Semester, Computer Network
7 LP. Univer
sity -(B techy Aka
sh Books
16-2018
that is unreliable, So __pacult to install. Attenuat
it's the medium through which data transmission occurs, no iJ] be attenuat ion & Dispers::
transmits its data over this unreliable medium and forgets about it. On the othe, dy ig pa to be a S ed and dispersed, whan ieee
10n di
sey IsHigher Th
TCP also transmits data over unreliable
medium, but it makes sure data Teaches n
othe stat vel Thigh e
an Co; ean oe ota die
omer
side by making use of Acknowledgements.
etc.) , chunks it(ifreq : atl a cae like fiber cable
.
1 ee by as much ag 60% e
ne Be
IP encapsulates the data sent by above layer (TCP, UDP,
adds its own header and sends it to Layer 2. So, if data sent by IP
is not receive’)
a
need eate s ene: Peiuae opel
n technolo cables. Owever, Opticalfb
ate ot

you will have to send it again. But, ifyou are

.
Joop, 2” eae uae
destination and you are using TCP, tical network: ca
su ch
hg (pat as * (ber
UDP, you just send it once. Spe sion, eqsoaimepinedl Is ia ling subscrib an . 7 ces se
When IP sends a packet, it receives a confirmat
ion for each packet and,if neeg special equi en Requir er d end sae renin) ea
not get confirmat io, transmission, ed: fe ens eth
rebroadcast the packet. This makes it reliable. UDP does (optical time-dom
ee duipment is needed, y, rate : a
supa ge ea
you. This makes it i t
broadcast each packet only once and assumes that it get to Sy equipment such : as ain rePfflectometry) ig pow required emXpensiy e,8
: ie 5
or rebroadcast. Meter areeneedeq
r ttPe sent Opticalotest
reliable then IP because it does not check for problems to properly provide optesting
tical profoboptical
es and beet n ae n
of fiber optics over copper as a transmj, slT a endpoints
Q.2. (a) What are the advantag es Q.2. (b) What is the remainder obtained bya:
optics over copper ? (6.35 Vidi 7
medium? Is there any downside of using fiber polynomial x° +1? Bx $54 Lby the generator
Ans. Advantages of Fiber Optic Transmission—
Optical fibers have largely replaced copper wire communications
in core netwo,
Ans.
1O
GG) =
eed
3
(6.25)
ay
in the developed world, because of its advantages over electrical transmission, Here
es of fiber optic transmiss ion. Since degree of G(z)isr-=3, append 3 as fies ‘0 th the fo end of M(x),
the main advantag Frame = 10100
00
Extremely High Bandwidth: No other cable-based data transmission Mediyy _ Generator =
transi
offers the bandwidth that fiber does. The volume of data that fiber optic cables Message after 3 zero bits are appended: ; A ‘ iG Goce
transmigg,
per unit time is far great than copper cables. Longer Distance: in fiber optic . : 00
can},
optical cables are capable of providing low power loss, which enables signals 1001 10100001000
transmitted to a longer distance than copper cables. Resistance to Electroma
Interference: in practical cable deployment, it’s inevitable to meet environmeutsjj,
gnes,
voor
0110 i
power substations, heating, ventilating and other industrial sources of interfere
0000 :
However, fiber has a very low rate of bit error (10 EXP-13), as a result of fiber being, 1100
ly
nojy
resistant to electromagnetic interference. Fiber optic transmission is virtual
free. Low Security Risk: the growth of the fiber optic communication market
is main 1oo1
the alternatiy 1010
driver by increasing awareness about data security concerns and use of
raw material. Data or signals are transmitted via light in fiber optic
transmissin
in" toh,
1001
Therefore there is no way to detect the data being transmitted by “listening 014111
securi
electromagnetic energy "leaking" through the cable, which ensures the absolute 0000
of information. Small Size: fiber optic cable has a very small diameter. For instang| 1110
is small
the cable diameter of a single OM3 multimode fiber is about 2mm, which 1001
than that of coaxial copper cable. Small size saves more space in fiber optic transmissin 1110.
Light Weight: fiber optic cables are made of glass or plastic, and they are thin 1001
than copper cables. These make them lighter and easy to install. Easy to Accommodst| 1110
Increasing Bandwidth: with the use of fiber optic cable, new equipment can be addedli 1001
existing cable infrastructure. Because optical cable can provide vastly expanded capaai 111
over the originally laid cable. And WDM (wavelength division multiplexing) technolog
met The remainder obtained by dividing x’ + x° + 1 by generator polynemialx3 + 1is 111.
including CWDM and DWDM, enables fiber cables the ability to accommodate
bandwidth. / ree (a) What are the disadvantages of of using small, fixed length cells in
ag ? : ; (6.25)
Disadvantages of Fiber Optic Transmission
Though fiber optic transmission brings lots of convenience, its disadvantages di he pr ATM is not contention-based or shared like FDDI, ethernet or token ring. A full
cannot be :znored. My ; ps is continuously available to every endstation. ATM transmits fixed-length 53
Fragility: Usually optical fiber cables are made of glass, which lends to the; pice cells. It consists of a header field (5 bytes) and information field of 48 bytes. The
more fragile than electrical wires. In addition, glass can be affected by various ch eader field contains routing info and the information field carries the service data.
including hydrogen gas (a problem in underwater cables), making them need more Small fixed length cells facilitate efficient multiplexing and a traffic prioritization
when deployed under ground. Difficult to Install: it's not easy to splice fiber optic Scheme ensures delay-sensitive traffic receives special consideration. Multiplexing
And if you bend them too much, they will break. And fiber cable is highly susce pide
becoming cut or damaged during installation or construction activities.
All thet
 Le, University_[p Te
18-2018 Sixth Semester, Computer Network . chl-Akash
Books ;
cular ATM link at any given 5
one service to share a parti ; ; 018-19
techniques allow more than | oe
virtual ae VC iden, i
Services are organized into virtual channels(VC) and | Hightora Bee PPP
ATM can be switched at both theit
the service and a VP identifies a collection of VC's. bs ata Link ;
and VC level. oneace . d yer Pr Po int-to-Poi;nt Py, eee
is done by conhecti n ' ane os
ATM uses cell switches, not bridges or routers. Switchingmakes it more efficient a
nted protoco]
in hardware which Bie : / B,vte Oriented
.
port to another directly. All this is done Only synchronous media Protocol
This is because bridges
faster. Bridges and routers also introduce latency/delays. ; synchronous a8 well as
introduy,
routers normally process large/variable length packets. These large packets Ihedia
| ynentication No pee of ete
noticeable application delay to a user. Ovideg authentication
are added, availa, authentication
In Ehternet or FDDI rings, media is shared. As more stations Does not offer dynamic D
segment the fp panic addressing
bandwidth to each station decreases. When this happens, we normally incre addressing, oe addressing is
However, these now
network into smaller peices by adding bridges and routers. lem ented in Point-to-point and malaaee ed.
must cross more devices between endstations. Additionally, agg; Only point-to-point
latency since packets
complexity, configurations,
and additional
more routers/bridges means adding more subnets opt ibility with . Can not be operated se configurations.
more bandwidth by adding more dey;
traditonal networks scales poorly. As we add
ther protocols non-Cisco devices, Interoperable With
like bridges and routers, we suffer from higher latency, worse performance end to eng a ii Non-Cisco devices also.
0 2
and added complexity to a network. tion of HDLC:Detelceeee
are ny Eg ignite
: ATM is scalable and inherently exhibits low latentcy. ATM networks HD: Slated hte dee ee is a WAN protocol miiiae.
between endstation,
contention-based. There are built on point-to-point connections
and switches. Each connection runs at a full guaranteed 155Mbps even through a switd, oe Fa REG Fe cee ae layer. The encapsulation ofthe datas ee
Stations are added by plugging in to an unused port or additional port modules. Aly oe the Synohuonoae oer zwe LC is the Predecessor of the HDLC which
good ATM switches can tie switches/modules together to act like a bigger switch. Aly, a redeveloped be Need cabal Protocol. Both SDLC and HDLC
standards. te @ ANSI and 180 for the acceptance
ice 0 technology isee ve rapidly.
dropping a eS, Priceaot an
en : psthe international
dapter card is about
adap HDLC protocol foll 7
ee
the cost of an Ethernet NIC card in 1983. : j fhe data transparen
hieving Here therbit-orienteg
follows:
protocol cy. concept
_
bit oriented approach arateand uses bit stuffing for
i signifies that the single bit is
use of small, fixed-size cells. First, the uset{icod to present the controll informnntiGes :
There are several advantages to the
‘ aoe shy eh is ; The fr of HDLC contains the
it waits less ifi/iidress, control, data, checksum and flag fields ae ae
small cells may reduce queuing delay for a high-priority cell, since : : a u l t encapsulation protocol in
gained access to a resource (suchas ihe (i
arrives slightly behind a lower-priority cell that has
be switched more efficiently
the transmitter). Second, it appears that fixed-size cells can
i t
which is important for the very high data rates of ATM.
Q.3. (b) Compare and contrast HDLC with PPP. Which one is byte} Frame format for the bit-oriented protocols
(6.25)
oriented;which one is bit oriented ?
Ans. The major difference between HDLC and PPP is that HDLC is the bit orientel
protocol, while PPP is the character-oriented protocol. The HDLC and PPP are the crucil
data link layer protocols used in WAN (wide area network) where the HDLC can alsolt|
implemented with PPP for the efficient results. : :
Frame format for HDLC protocol —
HDLC describes the encapsulation technique employed on the data in tt
synchronous serial data link. On the other hand, the PPP protocol deals with tt} * Address field — It is used to i rminal. shat
l field — as eile dee ae i
encapsulation of the data transported in the point-to-point links and it could)fin R* Contro
nowld aeiie a bits in the control field is intended for the Bequence number
synchronous or asynchronous.

Fe ae . ae field — This field is used to hold the information.

,_ Checksum field -In this field, the bits are reserved for the performing the cyclic
a Comparicon Ciett Mundancy code, i 5 é
2. Definition
Bee Commands and Requests
3, Key Differences
4, Conclusion f ine Heo uses a group of commands and responses for its working. There are three
mes information, supervisory and unnumbered. — :
20-2018 Sixth Semester, Computer Network
ey University-(B
+ Tech
Information transfer for B
in a sequential manner, which containin the information
me) - It trafield
A
mat the(I-Fra
infor’ nspsorts the mab etg4 jpistere nces Between HDLC a “dere
; etwe
oe 20
a
* Supervisory format (S-Frame) - The supervisory frameg Cong, ¢ jbecause
HDLC it 168 eee
managerial functions such as acknowledgement, ae transfer gta, Us Ya can be sent over Protocol while
the dial-up ppp isline
mo demi ;
"e-oriente
caecae sl i
a
in this are RECETYy iN va!
and error recovery. The commands and requests included ae 2
READY, REJECT, etcetera. - gynchronous miodif can be ee
RECEIVE NOT

* Onnumbered format (U-Frame) - It basically ees category Ee

functions. There several commands and requests fall under extends the data lica
such , 3 ‘ ous media, teaen ae
asynchronous agree
Kinkandauthentication Ontrast, PPP can Work with
TEST, FRAME REJECT, REQUEST DISCONNECT,
Definition of PPP: PPP (Point-to-Point Protocol) is' ice
also a WAN Protoed iy 7 ‘ be ths cake GHD
pP can dynamically assign and frees up the IP sate Teas it ig Provided in ppp
4 0st 88 according
are several enhancements made in the PPP protocol after HDLC. Priory, to the,use. As
. . .
: means that it can be used with two differs s o 5) thy
protocol is not proprietary, which ‘ater limitation be
thieoper
ab oftwHDLC this elimina
ee e no n- 8CO devices Faeces in; HDLC
: .
devices without committingwa changes over the forma’ t of the
nt
pact data. Al Aj] of t weve"? il ity n cj ted fr om the Ppp stecaye LC j P
collaboratively treated as single, independent IP network ot achievable,
which is having its qa (a) Compare and contrast the Go-Back-NARQ Protocol
format, hardware addressing method, and data link protocol. A point-to-point Cong pat ARQ- w; sele ctive.
with
is obtained without assigning multiple IP addresses to the tangible wires,
andiP Ans. Refer to Q.no. 4(b) First Term Examation 2018.
needs the IP network number. lainwhycollisivag ‘ S
, : 4, (b) Explai y 10n is an issue in a r,
There are several features of PPP, which are discussed below. b
Econ trolled access or channeliazing rentocs ea .
¢ To clearly identify the start and end of the frame, the framing method, J ccess protocol but
the asynchronous data. It is also beneficial in the detection of the errors. (6.25)
i
+ A link control protocol is used for enabling the network lines, testi,
ion and none is assigned the control over an

i astation to transmit.i
Transmission is random among
+ issioni : xf uled time
¢ It can select the NCP (Network Control Protocol) for each supported Netvoif,ihods are called random the stations
access. Also, no rules specify which eaten 4 ae pee
Frame format for the PPP
fations
a compete with one another to access the medium.
Bytes 1 i 1 lor2
:
Variable 2or4 fealso ecalled contention methods, Pr ‘otocols in this cate That is why these
; methods
gory : ALO) HA, CSMA , CSMA/
Hag Address Control Protocol Payload Checks Channelization is a multiple-access
method in which the available bandw
is shared in time, frequency, or throu idth ofa
gh code, between different stations.
Frame format for PPP protocol fihis catego ry: FDMA, TDMA, and CDMA Protocols
The PPP frame contains two flag fields, a protocol filed to determine the Inrandom access methods, there is no access contro
l (as there is in controlled access
packet residing in the payload, and a payload field which can variate. However lpe hods) and there is no predefined chann
el (as in channelization). Each station can
of the fields are the same as the HDLC protocol. fansmit when it desires. This liberty may create
collision.
Working of the PPP protocol _ak.
| Inrandom access method.
, there is no control over which station holds the link, any
*¢ The User first calls the internet service provider's router to establish theq#@tion can randomly
use the link, which may even cause collision. In controlled access
connection; the request goes through the modem. ttocol, j there is a primary station that controls
the link, and decides which station
* After the generation of the reply by the router's modem, the physical isthe link at a given time,
thus reducing chances of collision
is established. Inarandom access method, the whole available bandwid
th beldngs to the station
* User device sends a set of LCP packets in the payload field of one fat wins the contention; the other stations need to wait. In a channeli
zation method,
frames sequence wise. ‘ vailable bandwidth is divided between the stations. Ifa station does not have
data
© The PPP parameters are chosen according to the packets and thei nd, the allocated channel remains idle.
* After selecting PPP parameters, the NCP packets are deliv Ih Controlled access protocol, the whole available bandwidth belongs to the primary
configuration of the network layer,
f : ; (the sation with control to the link). In channelization protocol, available
¢ Then the IP addresses are d: ynamically assigned with
the help. ft qvidth 's divided between stations. If a station does not have data to send the
newly attached devices for the dur ation of its login by P ated channel remains idle. We do not need a multiple access method in
the provider. this case.
* Now the device is an internet host and is able to send local loop provides a dedicated point-to-point connection to the telephone office.
and re oo) Explain the similarities and differences between bridge and Fy
¢ At, the NCP ends the network layer connec
tion and spares
 ne, University-[B.Tech} Aka
22-2018 Sixth Semester, Computer Network 8)
ulti ort Bridge is employ
Router is that Bridge iat k ed when
Ans. The basic difference between Bridge and
n ™™ ple is containing the physical addr
device mainly operating at the data link layer of the OSI model with filteni
forwarding capabilities. Bay]
While Router is an internetworking device operating at the network layey ra
forwards Packs ttt oe ddresses.
OSI model. A router is attached to the two or more networks and
one networ k to another . fy ee Definition of Router: A router can be de i
Difference Between Bridge and Router re oF hardware, using metrics to decid la; #
tor © the best path to use ieza hanism,be it
The basic difference between Bridge and Router is that Bridge 18 a network a uter is a device that
filtering ang foryy Og ne
mainly operating at the data link layer of the OSI model with routes or forward dat
a from 0 Me netw,
capabilities. my ait IP address. When Data packet is receiveg from the router,
While Router is an internetworking device operating at the network laye, ther’ packet is; meant fo> Asie terother
ta's IP a ddress and; determines ife the Packet based . thon
r i inspects
t for its own e
meives the data, or if the data is for another n
OSI model. A router is attached to the two or more networks and forwards Packets Cormier
ia ee hetw
ieaienn
ither network. It means that routers only
one network to another.
the intended user. nation computer ig
Content: Bridge Vs Router
1. Comparison Chart Routers have ae 3 A es ey addresses and contain
Software that ena
2. Definition them to deerme i ae 3 = e several possible paths Betweari those bles
addresses is the
pest for the particu Heda
3. Key Differences break up broadcast and collision dom ains. It means b
4. Conclusion esRouters
ed Oe ouggh the
router, by y de default. ‘ It also Provides
i eeoeae
connections for sia r
theme
Router Key Differences Between Bridge and
Basis for comparison Bridge Router
1, Bridge functions at Data link layer while rout
Operates on Data link layer ‘ Network layer er Operates at the Network lay
er
Store and Forward Frame Packet ofthe2, OSIThe model.
bridge can relay frame from one segment to anot
Bridges are protocol independent Routers do not provid. her whereas Router store
Transparency | and forward packets.
station and transparent to the end transparency. :
i 3. Routers are not transparent to the end stations
stations. . i
transparent to the end stations and does not rely on the
Reads MAC address of the device IP address of the deyiy a ne a.
4, Inabridge, frames are forwarded on the basis of the
Works on Single broadcast domain More than one broadeay MAC address of the frame
domain As against, Router checks logical address (i.e., IP address) of the packets.
5. The router can work on more than broadcast domain while Bridge can work on a
Definition of Bridge: A bridge is a networking device for connecting two segme single broadcast domain.
of a network and transmitting data between them. Bridge operate in both the physi Q.5. (b) What is quality of services and what are general techniques to
and the data link layers of OSI model. There is a compulsion of using identical proton | improve quality of service ? (6.28)
for the segments to communicate. The primary use of a bridge is to send, filter, orf
Ans. QoS is an overall performance measure of the computer network.
any arriving frame depending on the MAC address of that specific frame.
Important flow characteristics of the QoS are given below:
The frames can be transferred between two originally separate LANs by abri
However, bridges incorporate logic that permits them to retain the traffic for 1. Reliability: If a packet gets lost or acknowledgement is not received (at sender),
segment separately. When a frame enters in a bridge interface, it checks the de the re-transmission of data will be needed. This decreases the reliability.
MAC address and forwards the late copy to the segment to which the address belo The importance of the reliability can differ according to the application.
Bridges are also apparent to the end stations. For example: ps e NS
Bridges are hardware based, and it uses specialized hardware called/ se
E- mail and file transfer need to have a reliable transmission as compared to that
(Application specific integrated circuit) chip to maintain MAC address table and tot ofan audio conferencing. ee:
filtering decision. Bridged network fragments collision domains, but do not frs
broadcast domains, instead forward all broadcasts. d
2. Delay: Delay of a message from source to destination is a very important
Types of bridges:
characteristic. However, delay can be tolerated differently by the different applications.
i
For example: . o
* Simple Bridges are the predecessor of its other types and the least expensiv?
of bridge. A simple bridge connects two segments and contains a table that lists The time delay cannot be tolerated in audio conferencing (needs a minimum time
addresses of all the stations included in each of them. What makes it primiti - delay), while the time delay in the e-mail or file transfer has less importance.
these addresses must be entered manually, e
 vy
IE University-{B,Tech_ Aan ‘
ork
4—2018 Sixth Semester, Computer Netw Coke
si Je: -mail or file transfer, 2018-95
ay.
1 in the packet del
3. Jitter: The jitter is the variatior er, 9 jot ox awith Integrated Services.
larg e, the n it is called as high jitt n the
If the difference between delays is pd! is problems with the Integratedservices
e ;
I], it is kno wn as low jitter.
n delays is sma are;
contrary, if the difference betwee i
phe gbility
Example: 10, 11, 12. Here a i) ted Services, it is necessary for ea
Casel: If 3 packets are sent at times
0, 1,2 and received at conversation, , ch router kee:
oni c ip ie not always possible due to
table for the teleph Srowing network, “©P information of each
delay is same for all packets and it is accep
at 31, 34, 39, so the delay is differ ”- on
..e- Type Limitati
Case2: If3 packets 0, 1, 2 are sent and received
the telephonig (i) >*
delay is no t acceptable for
for all packets. In this case, the time ated services model provides only two types of Seryi
conversation.
grein ; ps } ces, Suaranteed ang
h. opload:
need t he different bandwidt Services (DS or Diffserv):
4. Bandwidth: Different applications f pifferentiated
as computer networking model, which is de Sign
For example: ed to achiev
an ¢. e the Scalab
in comparison to that of sending
oD ang the network traffic. ility
Video conferencing needs more bandwidth designed for .
mail. jg a class based QoS model specially
Engineering Task Paes} é
Integrated Services and Differentiated Service G was designed by IETF (Internet

These two models are designed to provide

Quality of Service (QoS) in the network.
ated Services: the problems
handle
1. Integrated Services( IntServ) problems ofunitIntegrated
she solutions to handlemainthe processing Services are exn|s: below:
designed for IP. can be moved from te explained
Integrated service is flow-based QoS model and pility: The
ii) the
SealaLa a acl : S sea Ey The router does not ate ae the
a flow in the network, from source to
In integrated services, user needs to create the applications (or the hosts) define the store the
routers (every router in the system implements ation aboutthe tlows and
destination and needs to inform all ng the packet s. type of,Services
requirement. E want every time while sendi
IntServ) of the resource
services works. Service Type Limitation: The routers, route the packets on the basis of class
Following are the steps to understand how integrated
Protocol (RSVP): An IPis connectionless, datagram, services define in the packet and not by the flow. This method is applied by defining
(i) Resource Reservation applications.
packet-switching protocol. To implement a flow-based
model, a signaling protocol is used dasses based on the requirement of the
mechanism to make reservation (every g.6. (a) An organization is granted the block 211.17,
to run over IP, which provides the signaling : 180.0/24. te
to make reservation), this protocol is called as RSVP. |, ,istrator wants to create 32 subnets.
applications need assurance
resource needs to define the }
(ii) Flow Specification: While making reservation, (i) Find the subnet mask
.
(ii) Find the no. of addresses in each subnet
flow specification has two parts:
flow specification. The
resourtes that the flow needs to reserve, sses in subnet 1 -
(a) Resource specification: It defines the (ii) Find the first and last addre
For example: Buffer, bandwidth, etc.
(iv) Find the first and last addresses in subnet 32.
categorization of the flow.
(b) Traffic specification: It difines the traffic Ans, 211.17.180.0
(iii) Admit or deny j
Class C Address
the router decides to admit
After receiving the flow specification from an application, (a) Subnet Mask 255-255-255-10
the service and the decision can be taken based on the previous commitments of
or deny . W1111111 . 011111111 . 011111111 . 00000000
the router and current availability of the resource.
define Integrated Services } (0) No. of address
Classification of services: The two classes of services to = g32n
are: n= 24
that the packets arrive
(a) Guaranteed Service Class: This service guarantees
Here

the traffic flow maintains the traffic 982-24 — 98-256 (is the no. of addresses)
within a specific delivery time and not discarded, if
specification boundary. (0) 11111111 , 011111411 . 011111111 . 011111 000
needs a guaranty of
This type of service is designed for real time traffic, which
minimum end to end delay.
For example: Audio conferencing.
is designed for the
(b) Controlled Load Service Class: This type of service
to overload network a0
’ applications, which can accept some delays, but are sensitive fe
to the possibility to lose packets,
26-2018 Sixth Semester, Computer Network
= LP} University_1
B meq
hAkash Book
ep
Subnet gimilarly, @ device's IP lay s
er y eceives dat:
1. First address A d, so they end up at the ¢ Bramg,
ee See 2018_
pul pultiplexing and De-multiploni. aa on
211.17.180.0/24 211,17.18.255/24 subnet the devat@8tams Need to hy :
(d) 211.17.180.0/24 amhe question is: How do we de-multipley Sing Ports,
32 subnets ‘hat TeCeiveg them oo
Q.6. (b) Explain distance vector and link state routing protocols, different application processes on the same noses Of
What pear the IP a node How doeg thet
IP data
the merits and demerits of the both protocols ? 8consider ae that need to
ee hich process a i thelaue 08 if : ayer know which ae host
Ans. Refer Q.1. (h) of End Term Exam. 2017. (P.
iNo.- 13-2017.) ") Thetagram.
first par
Thisi
Q.7. (a)In case where reliability is not of wer
i
field carries a Hescode
in the
thatProto
; field
out included
primary importance,UDp Woy
make a good transport protocol. Give examples of specific cases. Es most end-user applications use TCP weupu in; pe
e
6 geldin a received datagram tells IP to Pass data to aiethe Protoco} the Reader of each
‘Ansport ieee nt
Ans. The User Datagram Protocol (UDP) is a connectionless, unrelia the data,
ble trang, But both TCP and UDP are used by ma
protocol. It performs very limited error checking. If UDP is so powerless » Why wou rt i process to send the NYdais
applicat; er TCP or
process want to use it? UDP is a very simple protocol using a minimum Overheag
ypPmust figure out which oo 8 at once, This as 7
«2 dressing element is necessary. Thig transport Jay ‘aces
process wants to send a small message and does not care much about reliability, P Possible, ang TCP oe
it i Aport can also be called a TSAP (Transport Sernnress is called aa additional
use UDP. Sending a small message by using UDP takes
much less interaction betine can also be called a NSAP (Network Service Accegg Point) i
the sender and receiver than using TCP. Example Reliability is not of primary impo, Point), An TP ad dr
® layer addressing is accompanied using TCP an d UDP », Y Concept; TCPAp
in applications such as echo, daytime, Users, TFTP (trivial file transfer Protocol) nad
tag particular IP device identifies a Particular application Be ach port number withing
SNMP (simple network management protocol).
3. Source Port and Destination Port Numbers: cess,
So here is the solution to the problem. Part) Reliability is the not at all th ety In both UDPcE and TCP messa Bes, two addreggssin;
€ Primi
Simple Network Mange and a Destination Port. These are analogous toi the Rela
concern in many applications like Bootstrap Protocol,
Protocol , Trivial File Transfer Protocol But in case eereliably destination address at the IP level, but at a higher
Peet for a Source Port
of
nee custom made softw:
plays very important role in client-server . na level of deisnoee address and
a pplications are help in buildi inati ddress at the IP], eve i s
reliability and provide a low over head service. Part2) Both UDP and |p a
ng y | address and Ce oes a identif‘ y the source e the source
connectionless and unreliable protocols. The differ
ence b/w them is that in Case of Upp}
scoop epi aoe ar a Destination Port identify then. ie
it calculates the check sum for the entire datag Process on 4 UDP port b ° estination Process
ram but the scenario is different in cage on the destinationsatan
of IP it only calculates the checksum of the IP header 3 TCP an Port numbers are 16-bits in length.
. Part3) There is no need of Py; Somerset f Port Numb :
addresses to be unique universally and is used with se ers for Data ort eye atsmission is
;
the IP address . summarize, here's the basics of how tran Sport-layer and Reception:
Q.7. (b) Explain the IPV4 protocol 3
with header format. Compare Pr in TCP and UDP: addreggj :
“sing (port addressing)
ee works
it with PVs, (a) Sending Datagrams: An application
gs ecifie ives
(65)| it wishes to use for the communication.
Ans. Refer Q.8. (a) of End Term Examation 2017. Theseps ase es ae
depending on which transport layer protocol the application one
Q8. (a) What are the different elements of transport is using. When TCP or UDP
layer protocols? (| ?#8¢s data to IP, IP indicates the Protocol type (TCP or
Ans. Elements of transport protocols (etp) ‘IP datagram. The source and destination UDP) in the Protocol field of the
port numbers are encapsulated as
° ‘Transport <> Data Link. TCP or UDP message, Within the IP datagram’s data area, part of the
© Addressing. (b) Receiving Datagrams: The IP software
d receives the datagram, inspects the
¢ Establishing a connection. | Protocol field and decides to which
protocol the datagram belongs (TCP or
* Releasing a connection. : UDP receives the datagram and passes its UDP).TCPor —
contents to the appropriate process based
on the destination port number.
* Flow control and buffering.
* ae FLOW CONTROL and BUFFERING
Multiplexing. : . | 1.Buffering .
* Crash recovery.
In detail- " (a) Sender:
i :
The transport service is implemented
1. In case of an unreliable network service the sending
by a transport protocol. ° 8 as it might need
transport entity must buffer
1. Addressing in TCP and UDP zl TEDU' to retransmit them.
Pees ;
case of a reliable network service the sender may/ma ;
A typical host has many application
processes running on it. These procestt y not buffer outgoing
generate data that ! : (by. R ‘
is sent to either TCP or UDP, which in turn passes it, : 3
transmission. This multiplexed stream of. to IP ft eceiver;
destinations. datagrams is sent out by the IP layer to ve ; ™ a Ifall TPDU's are of same size, there will be a pool of identically sized ss (one
; Per buffer), ; SANa ts, i
 De, University_
(B.Tech
28-2018 Sixth Semester, Computer Network mbers are looked at by Tp } Sh Books
ort pu aA a >and the p
ii. If there is a wide variation in the TPDU sizes, variable sized buffer are useg P ymbers are Ps Indepenceny c Cette Dat numb Port numbe TS b
n 2018-99
iii. Another option is to dedicate a single large buffer per connection, ‘
2. Dynamic Buffer Management :
Ua value
pimum aa foraethis alfieldbytes
is g. of the UD fae
ader and the 4 YUDP. The TCP port
pl one
Checksum: aon c
In cases where traffic patterns change, the sender and receiver need to dynamica}, pals is computed as the te. @Psulateg data. Th
adjust their buffer allocations. Y
complement
id the data,sum ol a pseudo header of informati
padded as needed with zero byes from : i Compl,
Working: the Phe Ment of the One!
(a) Sender requests receiver to allocate certain number of buffers to it. bytes: If the checksum is set to Zero, then lisa a ‘the eng to T, Ses
5 2:
(b) The receiver depending on its load allocates as many buffers as possible,
(c) When the sender sends a TPDU it decrements its allocation.
ake the checksum optional
i
to allow im june 8 disabled gre etiple of
‘omputationall overhead Ethie computed foe ementations 4 : a esigners chee
(d) When the sender finishes its allocated quota it stops sending.
jo OF FFF: “Um 8 zero, then this sate With little
(e) The receiver piggybacks acknowledgement and further buffer allocations on the Must be get
Q.9. Discuss the header format of TCp.
reverse traffic. There might be a case where there are sufficient buffers at the Sender
and receiver but the subnet is congested. Therefore, along with buffering, flow contro] Ans. . TCP Header: TCP is theie toe
pri ‘
transport Protocol useq t
; full-duplex connections. The most co (12.5)
should also concentrate on the carrying capacity of the subnet. on u z 0 ‘ %
i an IP datagram. Although IPis
encapsulated in ee TCP is to exchange pale
3.Multiplexing
TCP is typically implemented on hosts only to BeHes on end-to-end
both hosts an es
(a) Upward Multiplexing: If only one network address is available on a host all
The unit of transfer between the TCP software Sate Te iable
the transport connections on that machine use that single address. When a TPDU comes
Segments are exchanged to establish
in, upward multiplexing is used to decide which process to give it to. eonneecee CP segment,
acknowledgemen ts, advertize window sizes and acs eine transfer
(b) Downward Multiplexing: Assume that a subnet uses virtual circuits internal} ‘Ctlons,
data, send
and a maximum data rate is imposed on each one. If a user needs more bandwidth than
what a single virtual circuit can provide, downward multiplexing can be used to open <—— teen
multiple network connection and distribute the traffic among them. 3 —__—_____
: TCP segment
Q.8. (b) Discuss the header panel of UDP. (6.5) haa
Ans. UDP header: UDP is a simple, datagram-oriented, transport layer protoco]: TCP header TCP data (optional)
each output operation by a process produces exactly one UDP datagram, which causeg
one IP datagram to be sent. The encapsulation of a UDP datagram as an IP datagram TCP encapsulation
looks like this: :
Each TCP segment, encapsulated in an IP datagram,
<< IP dttagran has a TCP header - 20 byte:
long, unless options are present. Let's look at the standard header
—_———_
<_< UDP Datagram —-_____________> format in more eee
We will come back to the practical use of this format using a simple case study,
after
[ iP header [UDP header | UDP data arg | handling the process of connecting and disconnecting two TCP applications as well as
the data flow control mechanisms in TCP.
UDP encapsulation

UDP Header Format -TCP header format

Each UDP message is called a user datagram. Conceptually, a user datagram ze : 50332 bits
consists of two parts: a UDP Header and a UDP data area. The header is divided in four
16-bit fields as shown: ; source port ‘| destination port
UDP header format sequence number
acknowledgement number
“ 32 bits
Hien | reserved |r|2|s|5|: window
Vv

source port destination port ay checksum urgent pointer

[options }
length checksum
Source and Destination Port: The port numbers identify the sending process and Source and Destination Port Number: Identification
rien)
application. Along with the source and destination TP addresses in eae
the receiving process. TCP and UDP use the destination port number to demultiplex
the connection as a socket.
incoming data from IP. Since IP has already demultiplexed the incoming IP datagra! ; ;
to either TCP or UDP (based on the protocol value in the IP header), this means the P
 30-2018 Sixth Semester, Computer Network
7
Sequence Number: The sequence
Ifthe SYN bit is set, the sequen number of the first data byte in this
ce number is the initial sequen
data byte is initial sequence number ce number ang (ey
+ 1.
Acknowledgement
Number
Ne fnyTy
Ifthe ACK bit is set, this field contains the value of the
sender of the Segment is expectin next Sequence p
g to receive. Once a connection
always Sent, ig establishe dl te
Hlen: The number of 32-bit 1g i)
words in the TCP header. This indica
der is always a multiple of 32 bitstesWhere t},
begins. The length of the TCP hea
Flags: There are six flags in the TCP head , © dat,
Same time. er. One or more can be turned op
URG The URGENT POINTER field contains At the
valid data
ACK The ackowledgement number is
valid
PSH The receiver should pass this data
to the application as soon AS Posgi
RST Reset the con nection
p)
SYN Synchronize sequence numbers to initiate a :
connection,
FIN The sender is finished sending
data.
Window: This is the number of
bytes, starting with the one Specified
acknowledgment number field, that by ¢,
the receiver is willing to accept. This
field, limiting the window
to 65535 bytes.
ig a 16i
Checksum: This covers both the header
and the data. It is calculated by Prepending
a pseudo-header to the TCP segment,
this consists of three 32 bit words which
the source and destination IP addre
sses,
contain
a byte set to 0, a byte set to 6 (the
number for TCP in an IP data Protocg]
gram header) and'the segment
bit one's complement sum of length (in words), The 16.
the header is calculated (i.e., the entir
considered a sequence of 16-bit words). e pseudo-header jg
The 16-bit one's complement of this sum
in the checksum field. This is a mandator is stored
y field that must be calculated and store
the sender, and then verified by the receiver. d by
Urgent Pointer: The urgent pointer is valid
only if the URG flag is set. This pointer
is a positive offset that must be added
to the sequence number field of the segment
yield the sequence number of the last byte to
of urgent data. TCP's urgent mode is a way
for the sender to transmit emergency data
to the other end. This feature is rarely used,
 Ans. Refer Q.No. 1(b), Fir
st Term Examination 201
} ee 8.
1. (b) Ho w
We nee
mady to en sengd 2 aces aei s -
channel with a bandwidth
Ans. Nyquist bit rate for noi
seless channel:
Bitrate = 9* Bandwidth =
# log, (A)
265 kbps = 2* (20 KHz)
* log, (A)
265000 = 2* (20.000)
Es * log, (A)
log, (4) = 6.628
a
AeeeN) = 26.625 =
Q.1. (c) Justify the purpose of claddingin an 987 levels:
optical fiber? Discuss its density
lative to the core.
Ans. An optical fiber has two layers, the inner (2)
layer and the outer layer. The inner
yer is called core. The data pass through a core.
This core is made of dense of glass or
astic. The outer layer of a fiber is called cladding. It is also
s glass or plastic is less dense. Cladding is used made of glass or plastic.
in optical fiber for prevent any
fraction while passing data. The function of cladding is to occur full inter
nal reflection
optical fiber. That is why a cladding’s density is lower
then core.

-~beceptance © Cladding
( Angle
\

Acceptance Cone
Q.1, (d). Give the relationship between propagation time and Leases
eed? (
Ans. Propagation speed is the amount of time it takes for one particular signal to
from one point to another, Transmission Rate is the total amount of data that can be
nt from one place to another in a given period of time. .

d
2-2019 2019-3
Sixth Semester, Computer Networks Mee University-[B.Tech]}-Akash Books
a
Q.1.
er manne the probability of successful transmission of dat, a ] has a separate Presentation] 4. TCP/IP does not have a separ
ate
= um efficiency of slotted ALOHA over pure ALOHA? Da, Cees ft0s! mote Session layer. Presentation layer or Session layer.
™S- Refer Q.1. (e) End Term Examination 2019. 5. Transport Layer is both Connection
Q.2. (a) For ‘n’ (2) or Layer is Cor nnection Oriented|
port
devices in a network, what is the number of Oriented and Connection less.
Ports
A:
i Tnpu O'tpy,
for each device in a mesh, ring, bus and star topology»
ata 6. Network Layer is Connection less.
as- Refer Q.2. (a) First Term Examination 2018. @ |A. Network Layer isboth
. Q.2. = (b) Com Pare the OSI reference model and TCP/IP Mode} oriented and Connection less.
le) Whi) : h| 7. TCP/IP model is, in a way implementation
Scussing layered architecture. oslisa reference model around whic
Ans. Diagra mmatic y it is of the OSI model.
Comparison between OSI Reference Mode} and 79 6) ft the networks are built. Generall
Reference Model ‘ Pp used as @ guidance tool.
provides 8. The Network layer in TCP/IP model
TCP/IP Model 8 Network layer of OSI model
OSI Model provides connectionless service.
“poth connection oriented and
Application Layer connectionless service.
.
ng 9. TCP/IP model does not fit any protocol
9, OSI model has a problem of fitti
the protocols into the model.
Presentation Layer = Application Layer - model 10. In TCP/IP replacing protocol is not easy.
10. Protocols are hidden in OSI
and are easily replaced as the
technology changes.
faces 11. In TCP/IP, services, interfaces. and
Session Layer 5S a 11. OSI model defines services, inter
protocols are not clearly separated. It is
and protocols very clearly and makes
also protocol dependent. —
clear distinction between them.
It is protocol independent.
“Transport Layer 12. It has 4 layers
12. It has 7 layers
Exposed Station problem
Q.2. (c) Differentiate between Hidden Station and
(3)
Ess in wireless LAN protocol. :
Network Layers = 2)
ess LANs is the fact that, in
Ans. A significant difference between wired and wirel
ot b assumed. This
general a fully connected topology between the WLAN nodes cann
.
problem gives rise to ‘hidden’ and ‘exposed’ station problems
‘Data Link Layer Hidden Terminal:
hes
¢ The transmission range of A reaches B but not C. Similarly, the range of C reac
Bbut not A. Also the range of B reaches bothA and C.
E PhysicalLayer “| ° Now, the node A starts to send something to B and C doesn’t receive this
transmission.
* Now C also wants to send data to B and senses the carrier. As it senses it to be
OSI(Open System Interconnection) TCP/IP(Transmission Control
Protocol/Internet Protocol) free, it also starts sending to B.
e Hidden terminal problem occurs when two nodes that aré outside each other’s
1. OSI is a generic, protocol independent 1. TCP/IP model is based on standard
| range performs simultaneous transmission to a node that is within the range of each of
standard, acting as a communication protocols around which the Internet has}
gateway between the network and end developed. It is a communication protocd, | them resulting in a collision. %
user. which allows connection of hosts overa * That means the data from both parties A and C will be lost during the collision.
network. ¢ Hidden nodes mean increased probability of collision at receiver end.
2 In OSI model the transport layer 2. In TCP/IP model the transport layer * One solution to avoid this is to have the channel sensing range much greater than
guarantees the delivery of packets. does not guarantees delivery of packets
the receiving range. Another solution is to use the Multiple Access with Collision
Still the TCP/IP model is more reliab
Avoidance (MACA).
3. Follows vertical approach. 3. Follows horizontal approach.
 puter Ne tworks
4-2019 Sixth Semester, Com
IP. University-[B.Tech}-Akash Books 2019-5
Exposed Terminal:
>. Maximum through put of pure ALOHA= 184% = 0.184
B node is c urrently seriding some data toy,
¢ Imagine a situation wherein the
data to some a since N station shares a rate of 100 kbps.
* Now the other node C which i s right now free want to send
which is outside the range of A and B. ) Hence usable channel rate =
* Now before starting transmissio! nit senses the carrier
and realizes that the 0.184 x 100 = 184 Kbps
Arig,
is busy (due to interference of B’s signal). e bits per sec. outputted by
Bach station outputs 500 bit frame per 5000 ms henc
* Hence, the C node postpones the transmission to D until it detects the Medium, each station.
be idle.
¢ However such a wait was un-necess: ary as A was outside the interference
Tange op SOS
5sec
er Goeealsec:
a weak enough to be unable to penetrate into C
* Also a collision at B will be
ofa node tha Therefore N station give an output of 100 bps ona channel having a rate of 18.4
« Exposed terminal problem occurs when the node is within the range {bPs. hence
is transmitting and it cannot be transmitted to any node. N= (18.4 x 10)/100
* Exposed node means denied channel access unnecessarily which ultimately esl, = 184 Stations.
Q.4. (a) For a generator polynomial of x? + 1, what is the CRC value if
the
in under-utilization of bandwidth resources.
messa ge is x7 + 2° + 1? Express answer as a bit sequence with
no spaces. (5)
¢ It also results in wastage of time-resource.
many Ans. Refer Q.2. (b) End Term Examination 2018
Q.3. (a) 8-bit messages are transmitted using a Hamming Code. How Q.4. (b) How is selective repeat better than Go-Back N, Explain
with help of
single ()
check bits are needed to ensure that the receiver can detect and correct an example?
bit error? Show the bit pattern transmitted for the message 10011101. Assume Ans. Refer Q.4. (b) First Term Examination 2018.
that even parity is used in the Hamming Code. ,
Ans. Abyie of data is 10011101 create the code word, leaving spaces for Parity bits,
ete 00 ee
* Position 1 checks bits 1, 3, 5, 7,9, 11

Q_0_000_0101
* Odd parity so set position 1 to 1
Pal Or Sabato
* Position 2 checks bits 2, 3, 6,7, 10, 11.
1Q0_000_ 1001
* Odd parity so set position
2 to 1
hab sO ibs kato al
¢ Position 4 checks bits 4,5,6,7,12

111QQ00@_110@
Even parity, so set position 4 to 0.
1110001_1101.
* Position 8 checks bits 8,9,10,11,12

1110001Q0000
Odd parity, so set position
8 to 1..
111000111101
-, Code word: 111000111101
Q.3. (b) A group of N stations share 100 Kbps slotted ALOHA chann

has not been sent. What is the required value of N?

 TT
LP. Universi
ty {B Techy
8
END p) Differentiate between PAN,
TERM EXAMINATI ON Qe f cal Area ; 3 201
[MAy. 20 19) yoga
Network (LAN): LAN Stands for} ne
a local area. This usually
SIXTH SEMESTER [B.TECH] yon? es.
includes ast oeWAN,DeHOR. oy oi
(5)
COMPUTER NETWORKS [ETCS_g9 ae Network (WAN): WAN, in cont 8 aNd they, VTS, ag
Time : $ hrs The name is exactly what it sounds ig Sever MY Calo
Note: * Attem,
aittempt aues Q.no. 1TRG
any five questions including 6) yer . Beyond that, the definition is less 8i
which ig compulsory, MOM, Mang
.», eying [eid multiple buildingsTheon™most clear, Distances * overs
a corporate an areaais :
Q.1. Attempt any five erent countries. popular orWANcollegs
is thecampusta
sane links
ae Satellite fom anetn,ork
Q.1. (a) Differentiate of the following:
between Intranet, internet and
Internet? ; po Be
p S It’sAre actually a collection
Network (PAN): of other
PAN networks, induding
Stands for Personal ou r ‘oread
Ans. Many networks exist LANS andthisWang
article,
People connected to one networkin the world, often with different hardware (6) ey what it sounds like: a
different One. This often want to communicate with people network Covering a very g
desire requires connecting d 0 are. bee pest known wireless PAN network ate tha again,
*ncompatible networks, sometimes together different and technology ig Bluestone Y smal}
by using machines called ood
connection and provide the gateways ie *equently omoe vnartphones
wired PAN is USB. You
mightnay
necessary translation, both in as components In a network, think of your Wireless headset,
A collection of interconnected terms of hardware but they are definitely tan eet me
networks is called an internetwork i the Dice Many peripheral
_ Itis simply any network or just ines ee, aes as a PAN technology, devices aoe ctually computers
made up of multiple smaller since Wi-Fi is algo used over a theirin own rightcue
internetworking Protocols. networks usin : i smal] makers
Internet (big “‘T’), nor does
An internet (little “”) isn’t » jetropolitan Area Network
it necessarily use TCP/IP as
necessarily cone
. located in the same metro
(MAN) ‘Ametropoitan area network (MAN) connect
There are isolated corporate its internet workin *0 the jt id this kind of network area. One of the most common
ways for organizations
internets. is
eee neape apie: peat a microwave to use microwave transmission technology, Youmight
with a capital “I,” refers to antenna onan)
©t and continues toda the network that began its nee Nea extended high in the
life as ty, Es and sound pS air, beaming
directly or indirectly Ears as, roughly, the confederati ar cen
ee cee
Seamer US. backbones, Seen up sere work, | sing fiber-optic cable, ee
quite a few different on ane another carrier. Laying . :
networks cable themselves isis quite
qui expensive,i
‘ nal 1. (c) Differentiate between Adaptive
and non-adaptive routing? (5)
Itmeans a ie 1. Adaptive Routing Algorithm: These algorithms change their routing
Specific worldwide
internet that is wide
ly used
sions to reflect changes in the topology and in traffic as well. ieee et necsnnting
ect from adjacent routers or from all routers. The optimization parameters
| are the distance, number of hops and estimated transit time. This can be further
dassified as follows: : oo
(a) Centralized: In this type some central eee re ae ae
information about the network topology, about the sia es sae eects that
then transmits this information to the respective rou — i eaae taut te
oly one node is required to keep the information. declan
central node goes down the entire network is cos, 1.€. ae aun peck
(b) Isolated: In this method the node desitehie know about the status of a
In its simplest form, information from other nodes. The sending node
on the network having
an Intranet can be set
up ona networked PG — ay be send through a conges
access via the Intranet without.any wrticular link, The disadvantage is that the packe! of algorithm for routing are:
network to the Internet. Toute resulting in a delay. Some examples of this re carn ofitas fast asitan,
with a few PCs and a few printers
all networked * Hot Potato: When a packet comes to a meee regard to where that ee
by putting it on the shortest output queue-witho iting with the hotpee ane
‘ariation of this algorithm
is tocombine See into account both the sta!
When a packet arrives, the routing algorithm oi
U:Ninet\index htm ofthe links and the queue lengths. » ating tables atcache as
From then onwards they would * Backward Learning: In this method the ro = smplement back :
4 : tS had
they would get around the Internet, navigate around the Intranet in the same byinformation from the incoming packets. One way 10 *
mates. way
 4
LP. University-[B.Tech]-Akash Books 2019-9
tworks
Computer Ne
gixth ySemestes
pix
together with q h 0) hy A,B,C,D and BE, in which class A,B and
C
8-2019 n each packet, ilable are dive eae the five classes used and
tity of the sour ce node
de rece ives & pack et ina Particul,
1 ey sess are frequently used because class D is for Multicast and is rarely
2
to include the iden
each hop. Whe n 2 3 5 to reach it from the source tet ing ad! Bis reserved and is not currently used.
that is incremented on than the curren
t It, lass
Snes is bet ter On
E t coun = eine e : i . ClassA
t stor
revious value of ho ks?future ine. 31
e iS betteh eal ont ae bit# 01 E
Se is done but ifoi current valu e to uae recat
T
n the best aa the nod es hav Orget the an
The problem with this is that whe ce 4 | ie
a particular node. Hen Te
second best roxte te over agai ae Network sear
ically and start all an MOst\NUMbOl a
informations period ouring
Number
24 Bytes 5
receives informatithe on from HS neighbady Class B 8 Bytes
(c) Distributed: In this the node d packet. The dis , Node 31
sion about which way to sen paket cometh
15 16
and then takes the deci O 2

se
sen ds the bit#

oe
ati on and
interval it receives inform
that ifin between the the , B
may be delayed.
changes then the packet r
algorithms do not base theiroy
2. Non-Adaptive Routing Algorithm: These current traffic and topology, Instage ¢— Network Number——> <——Network Number——>
16 Bytes
estimates of the 16 Bytes
decisions on measurements and Class C
ae is computed in advance, offie

Se |
one node to the 31
the route to be taken in going from 23 24
known 88 stati O 3
the network is booted. This is also
bit#
and downloaded to the routers when ce
routing. This can be further classif ied as:
which every incoming packet ;
(a) Flooding: Flooding adapts the technique in arrive 4 Network NUMbe (oceans
Host —>
which it d. One problem a 3 24 Bytes Number
sent on every outgoing line except the one on 8 Bytes
may receive
of this a node adapte ge ie ‘
method is that packets may goin a loop. As a result ques d to oye, Vera]
which is undesirable. Some techni
copies of a particular packet s to a particular class that’s
why they are classful
; a Each of the IP address belong
these problems are as follows: classless
addre ssing syste m did not have any name,but when
r. When a addresses. Barlier this
* Sequence Numbers: Every packet is given a sequence numbe existe nce then it is named as Classful addre
ssing
r. If the node ie came into
receives the packet it sees its source addréss and sequence numbe addressing system d the flexibility
not transmit the packet ia disad vanta ge of classf ul addressing js that it limite
that it has sent the same packet earlier then it will system. The main any device.One of the major
that can be assigned to
just discard it.
. seul and number of addresses it does not send subnet infor
mation but it
ul addre ssing is that
¢ Hop Count: Every packet has a hop count associated with it.
This jg disadvantage of classf
netw ork addre ss.
decremented(or incremented) by one by each node which sees it. When the ho will send the complete
Cae
becomes zero(or a maximum possible value) the packet is dropped. Classless Addressing:
the network and host
« Spanning Tree: The packet is sent onl y on those links ble number of bits for
- .
i that lead to the destinatitio1 Classless adi dressing uses a varia
n-
See a a spanni ng tree routed at the source. This avoids loops in onan
| portions of the address. zeroes,
ace ° ly TEED all the intermediate nodes have knowledge of the network topology, ; bit stream of ones and
s the IP address as a 32
Classless addressing treat: portions can fall anywhere
between bit
nee eee a ae sees for general kinds of applications. But in cases where high i en netwo rk and host
reat tions, flooding is of greathelp. j where the boundary
betwe ss Inter-
cis Ss ai ase an as in in militar
mili y applica known as CIDR(Classle
ssing sys tem is also
of 0 and bit 31.Classless addre a Way to allocate and speci
fy the Internet
SPsRan ad as Aah Vhs :-In eid this ok hs packet isi sent by the node to one of ess addre ssing is of
Domain Routing).Classl with the original system
is highly robust. When the network is high routing more flexibly than
seer ccron es algtxt addre sses used in inter-domain (Clas sless Inter net Doma in Routing),
gorithm2 has the propert y of of makini g excellent use of al
perty address classes. CIDR er of significant bits
Internet Protocol (IP) y by base address and numb
rarily-sized subnets solel
routes. It is usually implemented by sending the packet onto the
least queued ig 4 es arbit na :
L 20 defin .
QL (d) What is difference between Classful IP addressing and Classless? | in the address. ALO HA? (5)
and PURE
maximum number of subnet in CLASSA with 9 ut of slottedALOHA
addressing? What are i
throughp
255.255.192.0. Q.1. (e) Calculate the
A LOHA channel: Let Gbe
the average number of fram
es
Ans. Classfull Addressing: Ans. Throughput of pure tran smis sion time. Then it can be proved
em during one frame The
generated by the syst pure ALOHA is S = G x eS.
essful transmissions for e is generated
that the average no of succ ie. if one - half a fram
18 0.184, for G = 1/2.
maximum throughput S,,,. sion time, then 18.4 pe rcent of these frames reac
h their
during one one frame transmis
 7
ks LP. University-[B.Tech]-Akash Books 2019-11
tere omputer Networ
Sixth Sem able 4
ause : the vulner isa demultiplexer. It separates the composite si 3
10-2019
Ra e xpected result bec tes onl y ong BS ie
ifa station gen era
inert jrawback of this multiplexing technique is aol WG cate ia
This 8 * Therefore,
destination successfully: igsion “re destinat jon successfully, Ate,
tra nsm l time-: Ks same time, there is a possibility of crosstalk. In brief, FDM divides erodes
times the frame ee oo
will eo
time, the frame Fd provide frequencies for the users. It does not divide time among users.
Re chacerabie ‘Ais S= nae (v2)
The throughput
for put : e ALO H
= 84w {DM: In TDM, the users can get complete channel bandwidth to send signals but
OF umber of collisions th; at can be 8 fixed time slot. It scales the time among the users. Assume that there are three
The maximum throughput Spas
erformance- The n
py a huge margin. And hence, the peepee
fo ser aS ul, u2 and u3 and the fixed time slot is t0. First, u1 will get the entire frequency
Sac
Slotted aloha gives better re teal
Collisions may only
aloha is a Place With nog te pandwidth for tO Ce When ul transmit data, the other users cannot transmit data.
take place in slotted oha. Collisi tantial es
comp ared to ae ne theless, this is a subs Tedlutin, , that time slot is over, u2 can transmit data for a t0 time. When u2 transmit data
become much better :
at the same t ;
cic , he other users cannot transmit data. Then, u3 transmits data for t0 time.
nS es ae iba
k
that are ready to spea
out a :
Throughput of slott is 9 = ne daring In brief, TDM divides time not the bandwidth among users. They can only transmit
for slott ed ALO HA
si : Re x nals within the available time slot. As only, one signal transmits at a time,
the
transmis sion s
words , ifaf
:
ram® ta estitination
is gener a
d
her sesstalk in TDM is minimum.
quel
0.368, : when G=1. Inot nati on succe
ssfully, This regu
frames reach their
time, then 36.8 percent of these rable time is equal to the frame transmission time WDM: Fiber Optic communication use WDM. The concept of WDM relates to Physics.
se the vulne
expec ted becau the fy; ‘ame Wil] : separates into individual colored light
can be
only one frame 1n this vulnerable-time, When a white light beam goes through a prism, it
Therefore, if a station generates y the prism. Each light beam has different wavelengths. This scenario works
reach its destination
successfully. beams b;
AL OHAis S=G xe“
combine back to generate a white light
The throughput for slotted
vice versa as well. The individual color beams
Snax 0.368 when G= 1 beam.
The maximum throughput that can possibly Le
The number of collisions : FDM vs TDM vs WDM
Slotted aloha gives better performance. the performance
in slotted aloha is reduced by a huge margin. And hence, FDM is a transmission TDM is a transmission |WDM isa transmission
take place place with nodes technique that modulates
Collisions may only take |technique that allows
become much better compared to Pure Aloha. But nevertheless, this is a substantia technique in which multiple
numerous data streams, optical
to speak at the same time. |multiple users to send
that are ready data signals are combined
signals over a common carrier signals of varying
reduction. for simultaneous transmi- into a single light
rate and bit interval in terms of data ssion via a shared channel by allocating fixed | wavelengths
Q.1. (f) Define the terms bit rate, baud beams through a single optical
(5) _ |communication medium. |time slot for each user.
transmission. Tr. Z
communication.
: Ans. Bit rate and Baud rate, these two terms are often used in data Functionality
Bit rate is simply the number of bits (i.e., 0’s and 1’s) transmitted
in per unit time, :
es multiple light
While Baud rate is the number of signal units transmitted per unit
time thatis needed allocates a fixed time WDM combin
FDM divides the bandwidth |TDM
; : for each user to send: beams from several channels
to represent those bits. into smaller frequency ranges |slot
ne them to a single
The crucial difference between bit rate and baud rate that one change of state can and transmitter transmit |signals through a common and combi
and sends through
l. User gets the entire light beam
transfer one bit, or slightly more or less than one bit that relies on the modulation data simultaneously through |channe la fiber optic strand similar
within that
ae used. Hence, the given equation defines the relation between the two: acommon channel within |bandwidth
Bit rate = baud rate x the number of bit per baud ae time slot. {Eoie) Mesa
their frequency range.
whee fa computer efficiency, the bit rate is the more important where W? Stands for
ene pe aa ong it takes to process each piece of information. But when Wea
FDM stands for Frequency |TDM stands for Time
ed about how that data is moved from one place to another we mp Division Multiplexing.
on the baud rate. The fewer signal i Division Multiplexing. é
Type of Signals
bandwidth needed to isa: sige | Be Dlote aad ae WDM uses optical signals.
|TDM uses digital and
Q.1. (g) Differentiate betw een FDM, TDM
é ee ; 5 , and WDM. : FDM uses analog signals. ~
analog signals. (6.5)
es.
st all the network topologi
FDM, the Bppiieceuiadin i
; : : tee
or a frequency range,ed Inamon,
2 chen f ae Q.2. (a) Compare and contra s con nec in
ted a net work
to the arrangementof computer
oe oe ee aon Therefore,
Ans. Network topology refers e, opti cal fibe r etc. Topology generally
_ through some physical [ medium s uch as cabl s of network topologies are as
e of the network. The various type
ane One ee ingthey
end,have
all their
hetsichale ae
own pathiwidae 8or frequency ae
ra
data. On ra the shap ee
: 2
. Afterward, the signal travels ae
Hhergh eeae
e channel.a oa
On follows:
12-2019 Sixth Semester, Computer Networks
7 LP. University-{B.Tech|-Akash Books 9019-13

1. Hierarchical topology 9. Bus topology 3, Star to tages of Star Topology

? : ;
‘
5, Mesh topology 6 Hybria gaDo { Cost 0 er.
pallat) is high '
topology
4. Ring Topology
1. BUS
‘ 2 pxpensive ee
Bus Topol 2 all the nodes depend
4,j “3i » bub fails then the whole networkrk i is stopped because
‘ai
Feature of "ed
turesste Topology
[Lit tenmeretts daa only in ; one direction,
cabl
single cable
stan toto aa single itchi
pe @.2. (b) What isi packet switching technique? Why is circuit switching better
a y device
2. Every device is connected (6)
vantages of Bus Topology wot telephone network?
1.It is cost effective ans. Refer to Q.1. (b) First Term Examination 2018.
: ( = required is least compared to other network topology. Q.3. (a) An organization is granted the block 16.0.0.0/8, The administrator
Pe oe janis to create 500 fixed length subnets, Find the following: (6)
(ji) Subnet Mask
Disady antages of Bus Topology
il Number
(ii) of addresses in br ief subnet,
1. Cables fails then whole network fails.
9 (iii) The first and last address in first subnet.
2. If network traffic is heavy or nodes are more the pe rforman Ce of the Networ, Ans. (i) 255.255.128.0
decreases
‘ (ii) 245 — 2 = 82766
3. Jt is slower than the ring topology.
(iii) First valid address - 16,0.0.1
RING Topology
Last valid address - 16.0.127,254
Features of Ring Topology. ee dards? Explain Syadtpicant
bee ans eins a pep eh!
' 1.The transmission is unidirectional, but it can baieae bidirectio Ethernet, Past Ethernet, Gigabit Ethernet
for High speed LAN. (6.5)
called Dual Ring To nal by having
connections between each Network Node, it is is used over ‘thick’ coaxial cable (Normally yellow in
papi pology, Ans. Traditionally, Ethernet
oe signal is baseband
CO
manner that is bit by bit. Data transmitte colour) called 10Baseb (the ‘10’ denotes 10Mbps, base means that the
Dae Bale beet ect aos f ma can transmit at one
mE . s ce : e network, till the destination node, , | \. takes the whole bandwidth of the cable (so that only one device
enone maximum length). The minimum
time on the same cable), and the ‘5’ denotes 500m
5 ing Topology
2 1 eee network is not affected by high traffic or by addin ir ee ee Topology’. Stations attach to it by
& More nodes, 48 The cable is run in one long length forming a ‘Bus
nly the nodes having tokens can transmit data. N-type connec tions or a transce iver which is literally screwed into the cable
way of inline connection
ie 15-pin AUT (Attachment Unit Interface)
2. Cheap to install and expand (by way of a ‘Vampire Tap’) providing a a drop lead connec tion
a DB-15 connec tor) for
Disadvantages of Ring Topology (also known as a DIX connector or ohm
of 60m length) to the station . The segments are terminated with 50
1. Troubleshooting is difficult in ring topology. (maximum
should be ground ed at one end only.
resistors and the shield as normal
— ing or deleting
2.Adding eleti the computers disturbs formats and CSMA/CD technology
the network activit; Fast Ethernet uses the same frame delay for the : signal across; the
3,Failure of one computer disturbs the . The difference 6 is that the maxim um
10Mbps Ethernet the
whole network, A of 51.2 microseconds. This comes from
STAR segment is now 5.12 microseconds instead microsec onds and that the slot
(time to transmit one bit) is 0.01
Topology fact that the bit time
Features of Star Topology Gap (IPG) for 802.3u is 0.96
bit times, The Inter-Packet 4
time for a frame is 512 for 10Mb ps Ethernet.
1. Every node has its own dedicated 9.6 micr osec onds
connection to the hub microseconds as opposed to dard s and is an exte nsion to
the newer stan
2.Hub acts asa repeater for data th t is the most popular of and it uses the a
flow, Nast data speed
ol he 100 donotes L00Mbps -
3.Can be used with twisted d pair,paj Optica) sont uolng O8MAICD. 3 and 6 for rece ive) and must only *
Fi { and 2 for transmit, oan
two pairs as 10Basel for it to be bet
Advantages of Star Topology installations with provision
on Category 6 UTP cable
Laat ae ted Pair-Physic
‘1.Fast performance with fi ew nodeg layer boing based on tho Twis ual data
; Nhe oppor phys ical
by ANS I X8T 9.5 committee, The act
b Bales tomas: and low network traffic,, oloped
(1P-PMD) dov ase'l.
Dependent 4 times that of 1OB
thr oug hpu t inc reases by between 8 to
8. Basy to setup and modify,
 i... Ae

14-2019 gixth Semester, Computer Networks LP. University-(B Te

‘ ae -Tech]~Akash Books 2019-15
h the functional principles of Gigabit Ethernet are the , =
cket leaks at a constant ‘
Although the tun’
format, the Physical
ing format, t! 3 ag
ee 1 te tant ; rate ; rate, meaning the network interf ;
eaenaenet ie, CSMACD andslotthetime.Framing
The standard Ethernet
; Slot tj Works, t Oe jets at a con
eriace transmits
aevery different. One One difference
at isaethe
aa over 100m of en ting F , pursty a 18 ae to : uniform traffic by the leaky bucket
‘ CSMA ICD half-duplex mode is not long ona
Carrier Extension is use. t0 guarantee 8 612-00 Per, » In practice the bucket is a finite queue that outputs at a finite site
Q.4.(a) Given the network topology below, use the Dijkstra’, algo gis arrangement can be simulated in the operating system Cae
ne
aiielts ofOe a afinite queue.
of this algorithm‘ thiis easy and. -Cnsists
compute the shortest
: +
path from 1 to all other nodes. Make sure to show ett
tep. e i yer ware:ra Implementation
pa cket arrives, , if there re 1si room in
a foe then the packet is discarded. © queue it is queued up and if there is
ine gr ES Ce 8) } 40109
Ans. ;

1 2 3 : : ‘i A 9 |
© 2 2 oo (oe) 0 oO) (oo) 0 cs Ss i
2 00 Cd
co
00 eS ‘gon ei re Sk J

3) 10
10 10 ©) *” St 20 Be po bursty “
(epee 2147 © 13. a
7 x a 13 %
es ca 16 14 @) 13
= - 16 14 a
= a 16.) ©

— - or a _- Za jon c 3

Shortest path: Fig. (a) Leaky bucket (b) Leaky bucket implementation
Q.5. (a) Draw and explain packet format of Transmission control protocol.
Shortest Path Weight Explain various steps that are followed in releasing aTCP connection? (8.5)
1to2 1-2 3 : Ans. The basic steps followed by TCP to establish the connection are as follows :
103 1-2-5-3 8 Release can be asymmetric(one sided, where data may loss ) or symmetric (both
|. \Afo4 1-2-4 10. | sides agreed) Steps for symmetric Release:
1t05 1-2-5 6 : * Step 1: Client end system sends TCP FIN control segment to server
1to6 1-2-5-3-6 16 + Step 2: Server receives FIN, replies with ACK. Closes connection, sends FIN.
14 ° Step 3: Client receives FIN, replies with ACK. :
1to7 tea
1-2-4-8 12 Enters “timed wait” - will respond with ACK to received FINs ©
1to8
1to9 1-2-4-9 13 ° Step 4: server, receives ACK. Connection closed.

3 @4. (b) Explain leaky i bucket algorithm. What kind of problemsueare address : (cient
; server
eter bucket algorithm? (4) Close-——~ IN

whoever may
ee tet Algorithm:
Ee te pes Consider
water pouring a Bucket
into the
with a small hole a the bi
bucket, the rate at which water comes a
be a ee aus i cae This scenario is depicted in figure (a), Onee the
i aoe ae nal water entering it spills over the sides and is lost (ie. it
e eae put stream through the hole underneath), ;
timed wait

he same idea of leaky bucket ca :

n be applied to packets, as shown in i
Conceptually each network
me inter face contains a leaky bucket. And the followit
are performed:
Closed
n the host has to send a packet, Symmetric Release
the packet is thrown into the
b'
 . ;

16-2019 Seah
eet seinbutse Computer Nevycr TP University-(B.Tech]-Akash Books 2019-17
eived, and tkhe e t transmitter
i willil usually
us also need to imple i
i
Connection releases: a the condition where the receiver does not respond. ees a1 ence tonecover
; a Hoey Under ae al pe trai tae issi ‘ :
the
sender will receive an ACK for the data and then
Host 1 ss [Send DR OR ss commence transmission ol the next data block. For a long delay link, the sender may
SIGE OR i+ start timer ro ~| | pave to Ve appreciable time for this response. While it is waiting the sender is said
estattimer SS | | to be in the “idle” state and is unable to send further data.
Send DR oR Le Leta OR | Go-Back-N ARQ 18 a more efficient use of a connection than Stop-and-wait
oR psa timer | eee ars van time ARQ
aed since unlike waiting for an acknowledgement for each packet, the
3 ee | peing utilized as packets are being sent. In other words, during theconnection is still
time that would
connection ” | oe otherwise be spent waiting, miSte packets are being sent. However, this method also
SendACK - __ACK | results in sending frames multiple times-if any frame was lost or damaged,
a | send ACK | ae acknowledging them was lost or damaged, the if that frame and or the ACK
a Release, | | the window (even if they were received without
all following frames in
error) will be re-sent. To avoid this
i (Timeout Selective Repeat ARQ can be used. 3
|| a | Tele;
tconn oat
t
Q.6. (b) Explain
.
OSI reference model in detail? Explain which layer is
1S ~ aie Sa responsible for data and flow control? — (65)
(a) b : o =
(6) Ans. In the OSI model, control is passed from one layer to the next, starting at the
Host 1 Host 2
“seoR] Host 1__
[Send DR, _OR Host 2 application layer (Layer 7) in one station, and proceeding to the bottom layer, over the
channel to the next station and back up the hierarchy. The OSI model
| SendDR — —9R St bpetattmey (+ start timer
takes the task of
ad inter-networking and divides that up into what is referred to
7 Start timer as a vertical stack that
ma fone fbhi ola eaatee --4, Send DR consists of the following 7 layers. ;
Mie =< bo je ee Start timer
|| SDR
send DR jee De fensh
ae cts oa
lan iee :: Thele 7 Layers yers ofof OSI

| Ss
|
f start timer?>———__—")_

|
ees
eo See) 7
i
|
[See
§

+start timer
(Time

i
— once
ad eA

sqrt
:
— 2es

User
=
A
.

jeonnection f : x Application (l 7
j i } e a
ie vee j i $ { $ Presentation (Layer 6}
| Send ACK }-—_ ACK <u} Release | (N Timeouts) (Timeout)

| =
| || ;é release2 |

|
releaseSis . Session (Layer 5)

(ete
| connection |
| {connection| iconnectcn
.

: Network (Layer 3)
Four protoco] scenarios for releasing a connection. (a) The normal case of a three-
way handshake. (b) final ACK lost. (c) Response lost. (d) Response lost and subsequent

7
|
DRs (Disconnection Requests) lost. 6 Physical (Layer 4)
Q5. (b) Compare Link State routing and Distance Vector Routing using
suitable diagram. 4)
Ans. Refer to Q.1.(h) End Term Examination 2017. cme
PHYSICa! Lik
Q6.(a) Explain advantage of GO-Back-N ARQ protocol over stop-and-™
ARQ? Application (Layer 7): OSI Model, Layer 7, supports application and end-user
processes. Communication partners are identified, quality of service is identified, user
authentication:and privacy are considered, and any constraints on data syntax are
identified. Everything at this layer is application-specific. This layer provides application
Services for file transfers, e-mail, and other network software services. Telnet and FTP
receives each PDU and sends an Acknowledgement (ACK) PDU if a data PDUis
are applications that exist entirely in the application lev el. Tiered application —
correctly, anda Negative Acknowledgement (NACK) PDU if the data was not: architectures are part of this layer. —
In practice, the receiver may not be able to reliably identify whether a PDU i
_ | |
|
2

Computer ie Networks |

Semester,
|
Sixth nce from ie Akash Book
provides i
ndepende . ty -[B.Tech]-
18-2019 is laye r nl re
t
LP. Universi
. Th
o epricato 7
eats
t a Gre 7 ed parity bit to
ng an extra bit, call
va
aS
from applicati |
yPtion) by
e
encryp ~ dating by ad di d
rity or odd in case of od
gBo e . . a k do ne
: e parity chec is
so(e
r n
parity Check Th ber of 1s either even'in case of even pa
anreSe
se nt e =:
: oe_
A
eel Re the form, wth & ye data to make , a nu
m
> om co
oes and eocr be eae the ‘
parity
8 e
sys
c o
ee dom 110nisl yor fbility problems. It ce toiagl d adds th the
r of 1s in it an ad s
a T iding fr
oa pa ti te rm ts the numbe
network, pr
wae! ea
ov
s
i
rm} whi nd er co un
plishes, Manage and te
— e se
is layelay r esta ng a frame, th
y This
5): Th hile creatiing way . If the
e 0.
: parity bit valu is
ver ; ll ow
es
s, act “ODnegti, pi ti n th e fo
u m b e r of 1s is even then
coordinate Rtg ty : If a n
pari parity bit value is 1.
eventhen
be? rt’ ee
udediealnogueslaye termina: + In case3 isof odd
to
nch
iex : an
ge s,q applications at each na Tt deal number of 1 lue is 0. Ifa number
= ve a
conversation
s, an
ection coordina
tion. een the a number of 1s is odd then parity bit va
fodd pari ty: Ifet
4, Provi des trans Incase “ a : i
with session and conn y Model, Layer
e - end-to-end transfey ota ‘
Transport (Laye r e aren ne vr a is even
s the number of 1s in it. TDIn case of even parit
y
2 ms, r hosts,
syste SeOverY ang to . oe receiving a frame, , the receiver count otherwise, it is rejected. A simil ar
Sere isures : aes gata transfer.le es ; of 1s is even, the frame is accepted,
check, if the count
control. switching and routing technologies le is adopted for odd parity
check.
Network (Layer 3): Layer 8 provides CTeating
rtual circuits, for trans mitting data from node to nods Rout; le bit error detection only.
The parity check is suitable for sing
!
essi ng, i
logical paths, ee . his layer, aS well as
of this ‘8y™
addr
= internetworking et : s error detectio scheme, the fo
n llowing procedure is applie
d
cing. Checksum In thi
packet sequen ets are encoded segments.
OSI Model, Layer 2 data pack
trolAt and
toon con2):
i , congessti
¥ fixed sized frames or
* Data is divided into
an

handling ’

nt ie decodeg ee oe ot It
ta Link (Layer wledge and manageme
,?

we Vs an
.
e
°

es tra nsmission protocol knome synchronization. The datia

a e The sender ms the ea h the ames.
i e It fe ui ecksum and sends i ong wit
w control and fra then complem ents the sutomget the ¢
errors ini the physical layer, flo
(MAC) layer and the is t withthe checksum usi ng l’s
ded into two sub layers: The Media Access Control e The receiver adds the
incoming segments along
is divi computer on the eae n complements it.
MAC ub utes cone i
how a to get the sum and the
Link Control (LLC) layer. The Controls Ro complement arithmetic accepted; otherwise, they
are
om e e it. The
If the resu lt is zero , the received frames are
ee Pe = eee —:
e
: eys the bit str; eam - electrical j
aeLayer 1 conv discarded. i z lves
ee e
PhysicalSie (Layer 1): OSI Mode l, : Im) tmp, ancy Check (CRC) invo
electrical and
ca"
y Che ck (CRC): Cyclic Redund n by
. ugh the network at the Cyc lic Red und anc d divisor agr eed upo
light or radio signal — thro a
car en Ha e level, I being sent by a predetermine
inary division of the data bits
g data on
s of sending and receivin
Sie the hardware mean and ATM ar ding defining | The divisor is gen era ted usin g pol ynomials.
cables, card s and physical aspects. Fast Ethernet, RS232, i © protocols with the communicating system. t by the divisor. It
ry division of the data segmen
physical layer components. * Here, the sender performs bina of the data segm ent. This makes .
and flow control. called CRC bits to the end
Layer 4 is responsible for data then appends the remainder
divisible by the divisor.
Q.7. (a) Explain CSMA/CD and CSMA /CA rand the resulting data unit exactly
om access protocostamat ing data unit by the divisor.
If there is no remainder,
Ans. Refer to Q.1. (b) End Term Exami natio n 2018. e The receiver divides the incom Other wise, it is understood that
data unit is ass ume d to be correct ani dis accepted.
the
mechanism is therefore rejected.
r correction and detection the data is corrupted and
ane faorre varlons erro (6)
s: Error correction techniqu i
es find out the exact
Error Correction Technique
waysin two
Ans. Error control can be done ‘ r locations. There are two
i
n involves checkin g whether any error has occury red of bi corrupted and as well as thei
« Error detection “Error ; detectio at: oe a that have been
not mate
bits and the type of error does
or oe number of error receiver detects an error
n (Retransmission- Ifthe
¢ Error correction “ ioni so e Backward Error Correctio frame. Itis arelatively
ae al at the exact number ef ansmit the
ngis not expensive
retr
a it requests the sender to
iently used only where retransmitti
er e e in the incoming frame,
bits that has been
For both error detection
and
Tr
the sender needs to send am!
5 simple technique. Butit can be effic
for retransmission is low relative
to the requirements of
additional bits along with the data oe ks based as in fiber optics and the time .
bits. If . The rece iver perf orms necessayychec
s, itremv | . the application
upon the additional redundant Ifit finds that the data is free from error e error in the incoming
the message to the upper layers. n- Ifthe receiver detects som
e Forward Error Correctio
f
: before passing
tl he redundant bits
that generate s the actu al frame. This saves
; ‘ de rrecting code
Erro re are three main techniques for frame, it executes error-co le in real-time systems.
However,
Checksum The
ieee eck,o e and Cyclic Redundancy Check (CRC). ansmission. It is inevitab
kK ®
in ee bandwidth required for retr
to be retran smi tted.
errors, the frames need
if there are too many
 20-2019
Sixth Semester, Com
puter Networks
The four main erro
r correction codes
are
* Hamming Cod
es
* Reed ~ Solomo n Code
* Binary Convol
uti on Code
Q-8.(a) How does AT . Low-Density Par
M differ from Fram ity-Check Cx
Service classe
s in detail? e Relay? Ligt 8nd
defing ; =
Ans. Asynch ronous Transfer compression. Class B traf
that uses a cel] bas Mode (ATM): ATMi he 4: fic is real-time VBR
ed sa network Witchi critical, such as interact , (RE-VBR), where
of fixed size cells of methodology to quantize data. AT ng ive video conferenc end-to-end delay
(15) ing. Class C 18 non is
53 bytes. An ATM M data traffic, where delay
Payload. This sma cell con tains a § byte hea Communicatig “Olay, is not so critical, such
ag
-real time (VBR-N
RT)
ller size, fixed-len der and 48} ” COnsigty mail messages, video playback tra
Video data as the gth cells are goo ining tapes and
delay is minimized d for ‘ransmitt
ing voisse? video
. 8ofATY Class D: Class D
ATM is a connec traffic is split into
tion oriented
protocol and the MARE ang _ Unspecified Bit two classes: Availa
established bet
refore a Virtua Rate (UBR). The ble Bit Rate
ween Sending
and receiving l circuit ch | more tolerant of se classes are
for bursty LAN
(ABR) and
two points whe points. Itestabl delays and cel] loss traffic and dat
n the data tra ishes a fixe . UBRis a “best
nsfer Starts. route Ould be | bit rate or traf effort” Service that
a that ig
Another import fic Parameters
ant aspect of devised as a and has no qual does not specify
multiplexing ATM is its asy twee, way to make ity of service
Therefore cell nchrono us use of excess guarantees,
unlike in conventio s are transmitt operatio
n in ti loss and the
discard ofwhol bandwidth, Originally
ed only whe: Nn UBR is Subjec
nal time division e
multiplexing wh, ere data is available to differs in that e Packets. AB t to increased
transferred if the it is a managed R, like UBR, is cel]
re data is not Synchroniz3 atio Service, based als o a best effort
available to be
sent. n byteg 8 low cell loss, on minimum service, but
= cel] rate (MCR
) and witha

e
i

Fe

sf
eg

5
fo
6

gg
¢
£
|
ATM provid £2
information
es 2 good qual
ity of Service one of the £0 up to 19
Such as data, in networks wh ISDN
D.
Voice, and are =
types can Pas Supported. With er e different t
s through a ATM, each of
Single networ
k connection ‘YPes
thes ie informationof
Frame Rela .
y: Frame Te
layisa Packet :
Switching te
chnolo gy fo
r connecti
ng ne twork

f - All the Points

frame is made which need _* Easy growth to
of two parts 4 port. Every to connect
which can be Port has a un
ique Address,A higher bit rates which
called as ‘actua Maintenance Pro enhances
the D channe itecture j am e as defined l dat a’ an d the ‘frame
cess, the administrati
l) Which has for LAP-D (L relay * Itis capable on and
over Virtual Co a Variable len ink Access Pr of transporting :
nnections, gth for info ocedures on broadband Sig :
rmation field. *Itis multiven nals.
These fram dor and suppor
Frame relay es are sent ts different Ope
can “reate mu ° It provides rators.
without having ltiple redund ; network transport
multiple phys an t connection interactive multim Services on LA
s among va N Such as vid
means to buff ica) links. Since frame rela
and provides rious edia, eo conferenci
interconne er Speed Vari
ations, it ha
y ig not specific for routers, * Optical fibre ba ng, and
ct medium
between diff s the Possibilit media, ndwidth can be inc
reased without limit
The ATM Fo erent types y to create a * Switching Protec in SDH.
rum hag defi of network good tion to traffic is offere
i of Service (Qos n, Po in ts with differ d by rings.
Quality ent speeds. ° SDH allows qui
) levels: Yer servicece clas ck recovery from fai
c ] , each with stscalable
_ Class: ses, eac! lure.
Class A or Consta PDH: In PDH, dig
nt Bit Rate (C ital
stream of bit
s at 4 steady BR ), tra ffi ‘re derived from differ multiplexer’s inputs (bit streams) are of Sa
bandwidth rate, g uch as c is characterized by ent clocks from different me bit rate and
oscillators. Each
traffic that
ighighly
TDM m ul
tiplexer traffic, a continuous tolerance of few clo
ck Periods, Hence it is will differ within
Sensitive to delay and Class Atraffic is
low- Bit Interl
called Plesiochr
onous,
intolerant to cell loss eaving is used in PDH to com
, bine digital signals,

—
 works
22-2019 Sixth Semester, Computer Net

544M) 6.342 Mbps 44.736 Mbps oat Mbps

64xbps (O50) OS-2 os-3
; DS-1 c
400.355,
$ ee
2 Primary {I Ds.g Mors
MUX —j mi2 V
channels | — xe Ss 23
eae ee x7 =)
Iya)

X28

= x3 |__399.264 mine,

a=}
PDH Level Hierarc
in Am

of PDH:
Following are the limitations
H
° In PDH, different frame is used for transmission and in data layer, €Nce
complex.
multiplexing and de-multiplexing is very
be de-multipleted
* Accessing lower tributary requires the whole system to
limited in bandwith,
° The maximum capacity for PDH is 566 Mbps, which is
® Tolerance is allowed in bit rates.
¢ PDH allows only Point-to-Point configuration.
; Hes
* PDH does not support Hub.
+ Every manufacturer has its own standards; PDH also has different multiplexing
ing networks.together Bie
hierarchies making it difficult to integrate interconnect