‘Watt Moret Futons Module 03: Data Link Layer Q. Framing Methods ‘©The DLL akan packets om the Network Layer and converte thm i ames, © Faced San Fm Franes ar of fued sae Eg: ATM vie Aes ‘Network © Varlale Siz Framing: Hee the sz fe tare is variable tus we need ‘a retrod to spect the sta and end of a frame ‘eshniues of Vaable Size Framing: ‘Character Count 1 Tat ayn an gray en a ee ia intr Catecen a o ppt esr ; 2. Tascane sore yn ee bates Chr cit a rata bi ine stciponear dean ste mete Sirecin tnd lve hdottuntomrtary tn 2 Rowisaue saunas saooung bed. Hannon arc "destination or receiver might Jose synchroriza 4 ee ee ee ie ofnentrame‘dvtnaton renter mht lowe aynchvoruation 4 The dstneon or recetver mgt alo be laa feat rary begining ote tame ‘Was Moment Tuition [Jelba2]—[ ellis] Jill 1. Character suing i also known #8 bye sling of characer-orentedKeming and i sre as iat of stung Bu byte suing actualy operates on bytes ‘ereas Dt sling operates on bts Inbye slung, » speci! bye tals baslealy known as ESC (Escape Characer) that has» predefined paler i generaly added tothe daa secon othe data slveam or frame whan there is message or character al haste same pater ‘8 at of fg Oe {ut the recover removes ths ESC and keeps the data par thal causes some probems or sues In simple weds, we can say tal character sting is an ‘ddiion of 1 adatonal byte there x presence of ESC or Nagin tet ‘Wilast Moment Titans ovsorattaattastiits 10010 ovsorssstOrtts1O1811 1010010 ovsoutsagttyit4141410010‘hat Nat a predetned pater ik general added tothe data section of he data ‘Srear or Fame when tere i mesnage or character Mal has the same pati ma that of tag byte [Bute recetver removes his ESC fred haope the date part thal causes some [problema oF lanues, In simple words, we can say tha character stuting Is an ‘®8a%on of 1 adctiona bye there Is presence of ESC oF fag in text ‘Rage ang nig OrvouTatdttttt11114110010 orsordys1011111011111010010 orsorssytittttt111110010 1 stung salto know as bt-oneted raming or b-orented approach. ni stung, exo bis ore being adced by network roocal designers to dala streams 2. tis general inserbon or adn of extra sino a ransssin ui or macsaye lobe bansmited a 2 simple way to provide and giv sanaing information ond dato he reeiver ado vol 0 gnre the appearance of Unintended or umecessary contol sequences 3. lis ape of pretoeal management simply perormed to break up abi pater that ess in ranemission logo ou of synchronization 4. suing la very exten pr of he vasmssien roost work re communication potco| tis 80 requires nUSB. ce Ph or cs wien: Np Ee ee for rietwork in which encoding on physical medium incudes some sort of redundancy use of more than one draphical or visual stucure to simply encode of epresen! one varabe of ala. 1. Encoding violations 2 method thal i used only Q Sliding Window Protocols elective Repeat “+ For noisy inks, there is another mechanism that does 10 just one trae 1s damaged; ony the damaged frame is resent yend N frames whenVWhtaat Moma For noisy inks, there is another mechanism thal docs not resend N frames wien just one frame is damaged: only he damaged frame Is rosent ‘» This mechaniam i called Selective Repeat ARQ {itis more efficient for noisy links, but the processing at tho recalver Is mare ‘Windows: : Sis, 1 The Selective Repeat Protocol lo uses wo windows: a aed wow and 8 teceive window Soe 2 However there are dflerences between the windows In this protocol end the’ — cones in Go-Back-N ‘Seng Window. First. the size ofthe send window Is much smaller; itis 24m-1, Second, the receive window is the same size asthe send window. ‘The send window maximum size can be 2'm-1 For example, fm = 4, the sequence numbers go from 00 15, but the size of the vwandow i Just 6 (itis 15 in the Go-Back-N Protocol) ‘WiLost Nomen 4. The smatier window size means less efficiency in fling the pipe, but the fact that there are fewer duplicate frames cen compensate fortis, 6 we discussed for Go-Back-N.‘WiLast Moment Tutor ‘The smaller window size means less efficiency in filing the pipe, but the fact that there are fewer duplicate frames can compensate for this. ‘The protocol uses the same variables as we discussed for Go-Back-N =} ‘The ectve window in Selective Repeats tly ciferent fom the on in Go-BackN First, the size of the receive window is the same ashe ize ofthe send window @m-1) $e Slectve Repeat Protocol lows 8s many fries as thes fhe eco window to ative out of order and be kept url theresa set of n-oder frames to be delivered tothe network layer. = * ‘Because the sizes ofthe send window and receive window are the same, all the frames in the send fame can ative out of cer and be sere unt they can be 1. inthis case when @ damaged frame artves the receiver simply discards al the ‘subsequent frames. 2. ican transfer more than one frame al a ime thus itis faster than the 1-it siding window protocolWorksng pee ‘© Sender sends N frames and wails for an acknowledgement, if tram Is in leror, packets received after Rth frame wil be discarded and it starsresencing from R” to N° Fame. £9: Assume window size= 4 frames," re ‘Suppose the sender sends fames 1103. ha aN {thas to now wat or the acknowledgement befef i'Gan proceed. AS each successive 2 acknowedgement is received lt ‘ana the sender can send the nent ames. BS Suppose an acknowledgement ofthe 0 frameis los hen rs al {he frames after the lost frame (ie. 1,2, scones Max Senders Window Size = 2k~ 1 eaton ‘ ES ‘© Suppose that we keep window size = 2k ~~ © Assume k=3, herefore 2k=8 © Altime t1 sends frames 010 7 t0 © B receives each of them in arder and at Ue 2 sends acknowledgement — forthe most recent [Link] 7. ‘Suppose that Bis [Link] ost Now 8 doesnot know thatthe acknowledgement it seri lst. ‘does not receive the acknowledgement and tes out, On timing outit resends te fames 0107 0 8 Butts 6 duplicate ofthe frames that were already received by 8 ‘Witast Moment Tu ‘© This problem occurs because 2 consecullve windows contain the same {rame numbers, Reducing the frame size by 1 solves this problem, © Hence maximum sender window size 2k ~ 1 ‘¢ Max Receivers Window size #1 ‘© Reason The frames are always received in order. ‘©The frames must be received inorder,‘Aces nol recetve the acknowledgement and times out On tring out it tremens the frames 0 to 710 B. [Butt a duptcae of te amen that wore siredy receted by B ‘WiLast Moment Tut ‘© Tris problem ocaurs becaute 2 consecutive windows contain the same frame numbers. Reducing the frame size by 1 solves this problem ‘© Hence maximum sender window size 2k 1 ‘+ Max Receivers Window size ‘© Reoson. The frames are always receive inorder ‘+The frames must be received in order. Q Channet Allocation Problem 11 Channel allocation isa process in which a single channel is divided and altted to multiple users in order to carry user specific tasks ‘The user's quantity may vary everytime the process takes place. If there are N umber of users andthe channel is dvided into N equaksized sub chennels, Each users assigned one portion. It tee numberof users ore small and don't vary a times, then Frequency Divison Auatplexing can be used as itis a simple and efficient channel bandwith ‘locating technique. : ‘Channe! allocation problem can be solved by two schemes: Stabs Chane! ‘Alocation in LANs and MANS, and Dynamic Channel Allocation. 1 in AN 41 is the classical or tradtional approach of allocating a single channel among ‘maltpie competing users using Frequency Division Mutiplexing (FDM). it oere are N users, the frequency channel is divided into N equal sized portions (eandwicth), each user being assigned one portion ‘Since each user has a private frequency band, th itis not efficient to divide Into fixed number of chunks, is no intererence between 2 Qynainic Channel Atocation, Possible assumptions indude 1 Station Model ron hasumes that exch of N stations independently produce frames, The jrobebay of producing & packet in the Inert IDI where | the con‘WiLast Moment Tit ‘2. Dynamic Channel Allsation, Possible assumptions Incide + Staton Mode! ‘8 Assumes tha ench of N slations independent produce frames, The ‘probably of producing @ packet in the interval I where |i the constant ‘anival ale of new frames. ‘Single Channel Assumption ‘8 In ths allocation al stavons are equivalent and can send and receive on that channe! CCotison Assumption ‘8 Iftwo ames overiapinime-wse, ten thats 3 colision. Any cosion Is ‘anerror, and both frames must be reransmited, Colisions are only possible eros. ‘Time can be divided ito Siotied or Continuous ‘Stations can sense a channel busy before they ty i © Carrier Sense Multiple Access/Collision Detection (CSMA-CD) 1. The CSMAICD(Camie Sense Multiple Access/Colision Detection) specications have been standardized by IEEE 802.3 standard 2. tis 8 very widely used MAC protocot Mesia Access Control “The problem in CSMAis that a tansmiting sation continues fo transmit its frame ‘even though a colision occurs ‘The channel time is unnecessary wasted due to this. In CSMACD, ita station receives other ansmissions when iis transmiting then a eolision can be ‘celecied 95 s00n as t occurs and the transmission ime con be saved ‘is soon asa colision is detected, the transmitting staions release ajam signal. ‘Ihe jam signal wil alert the ther statons, The stabons then are not supposed to transi immediately after the colision has occured. ‘Conenvise there is @ possibilty thatthe same frames would colide again. ‘ater some “back off delay tie the stations wil retry the transmission. f again the colision takes place then the backoff ime is increased progressively. 1 \whtast Moment Tutrae “Dack off delay time the stations will tr citdonutespucstente bes ofune titer oe ‘The station that has a ready frame sets the parameter() to Zero ‘Then it senses the line using one ofthe pecsistent strategies then sends the frame. tere s no colisin fr a period corresponding to one ‘complete frame, then the transmission is successhul ‘Otherwise the station sends the jam signa to inform the other stations about the collision “The station then increments the back off time(t) and waits fr a random back of time and sends the frame again If the back of has reached its iit then the station aborts the transmission (CSMAVCD is used forthe wacitonal Ethemet ‘wast Moment Tul Q. Cyclic Redundancy Check & Checksum 4. Songer kind of ertor-detecting code i in widespread use atthe irk layer: the CRC (Cyclic Redundancy Check), also known a a polynomial code 2. Polynomial codes are based upon treating bi stings as representations of polynomials with coefcens of and 1 only. Aki fame i regarded asthe Poetcient ist fora polynomial wih k terms, ranging from x" 10 x‘ime and sends the frame again. & Ie back off has reached Hs imi then the station aborts the transmission, 7, CSMA(CO is used for the traditional Ethernet. ‘Wilast Moment Q. Cyclic Redundancy Check & Checksum 1. Stronger kind of ertor-detecting code i n widespread use atthe lik layer: the GRC (Cycke Redundancy Check, also known as a polynomial code Polynomial codes are based upon treating bit strings as representations of olynomiols with coefficients of 0 and 1 ony. A k-bit frame is regarded as the ‘coeficient list for a polynomial with k terms, ranging from x" to x* For example, 11000 has 6 bits and thus represents @ six-lerm polynomial with Coeficients 1, 1,0, 0, 0, and 1: + txt+0x'+Or'+0x's1x When the polynomial code method is emplayed. the sender and receiver must ‘29ree upon a generator polynomial, G(x), in advance. Both the high and low order bits of te generator must be 1 ‘To compute the CRC for some frame with m bits corresponding to the polynomial M(x), the frame must be longer than the generator polynomial ‘The idea is to append a CRC tothe end ofthe frame in such a way thal the polynomial represented by the checksummed frame Is divisible by G(x) When the receiver gets the checksummed frame, it ines dividing f by G(s) if {here Is a remainder, there has been a transmission error. ‘The slgoritim for computing the CRC is as follows Let rbe the degree of G(x) Append r zero bits to the low-order end of the frame s0 it now contains m + r bits and corresponds tothe polynomial xt MQ) Divide the bit string corresponcing to G(x) ino the bi string coresponcing 0.x" M(x), using modulo 2 divisions ‘Subtract the remainder (which is always of fewer bats) rom the bit sting comesponding to xr M(x) using module 2 subtraction The result isthe check summed trame tobe transmdted. Call its ‘polynomial T(x) sroronsiyy too 110000111 0— Quote (roan svay)Dat Weare fat ‘Emtec fora fre 101011111 using fe generator Gon # +1 sersyey ° t 1 ° ' (Oo Rerence temenmectome 11 01 01111 10.01 Oo Frame wa ts toc apanc ns rare recs 4 The secone lund of error-detectng code, tne checksum. 1 cosety related to groups of panty bis The word “checksum” i olen used to mean 3 gio of Sloe bis sssocated with a message. regardess of how ae calculated, A QiOuP fof panty bls 1s one example of a checksum vicwaves tere ere other, stronger checksums based on a runang sum of fhe foaw bis of he message The checksum usualy placed at tne end of ie massage a8 We compiement ofthe sum function WiLast Mast sing he ert receneed coxa evots may be Gotordod by ones dk be eso, 0 81 Ps ER. a chs tHe Sen te VO eset ceca send on nat tus ech # ws fw srg ‘oes op ach LT a 1 easinine ot cron 4 pnts jas he WF pac ‘ated vio VOM mer ecu Hoy Md‘Lt Moment Tait pe entre recenved codeword Bom "pero no error has been ‘Tie way errors ny be Gatco By SATIN Wt tre ret comes ot 10 D8 ata tte anc checker