CHAPTER 2 Inrecuction, functions. Design tesues: Services Correction, Paty Bits, Hamming Wait, Sliding Window Protocol WAN and Slotted ALOHA, CSMA, WOMA, ¢. Ethemet IEEE 802.3, IEEE 802.11 aft aming ARO strategies: Control Protocols: Unrest DLC. MAC Sub ayer: Multiple Access Ps Binary Exponential Backof algoiten nes n, IEEE 802.15 and IEEE 802.16 Standards. Introduction 24.4. Functionality of Data-lnk Lay Data Link Layer Design iss 224 _ Diference between Frame and Packer Services Provided to the Netw er. 2311 Types of Services provided by Data Link Lays, Framing in Data Link Layer. erent nats eed of raming 7 SA AURRTOW MSTA Different Techniques of Framing UO. “What are the diferent techniques of framing ? Explain any twa 251 Character Count 252 Flag Byte with Character Stutting £3 Starting and Ending Flags, with Bt Stuffing 254 — Encoding Violations. Error Control re sola he purpose of enor contol rom DLL perspective, [SMR TSM ITRTASTTITEY Flow Control GQ. Explain the purpose of low contro from DLL Summary for Data Link Layer Error Detection and Correction 294 Types of Enos. 292 Redundancy Cee eee spective. Uo, 2.9.3(A) Error Detection 2.9:3(B) Error Correction, Peay nae Linear Block Codes : Hamming Code, Hamming Distance, Pa 2.10.1 Hamming Weight of a Code Word {W(x)] 2.10.2 Hamming Distance.an Computer Networks. 104 Hamming Codes: 4 Seoutty (SPPU-Sem.5:CONE) m) Minimum Hamming Distance 2.10.2( Hamming Codes. les on Harring ang Code Solved Examp Advantages and Disadvant Party Check: ait a4 va. 2122 2123 Checksum va. ua. 2134 2132 ‘Simple Parity check: Disadvantages of Simple Parity Check Longtud inal Redundancy Check (LAC) a age er Parity Check Vertical Redundancy Check (VRC) or Explain the VRC with example. Advantage and Disadvantage of VAC... clic Codes. 2.11.3(A) Drawbacks of 2D Parity Check 4 Cyclic Redundancy Check (CRC) 2.12.1(A) CRC Generator 2.12.1(8) CRC Checker. x 2.12.1(C) CRC is Appended at Trailer Not at Header: Why does the data link protocol always put the CRC in a trailer rat! [(SPPU - Q. 3(e), May 17, 4 Marks) Advantages and Disadvantages of Cyclic Codes Solved Problem. Data Link Layer. her than in a header ? Explain Checksum with example, (SUTRAS MIEN SIRO ETS Explain how a strong Generator Polynomial is formed, Give the Ale : (One's Complement. Intemet Checksum Elementary Data Link protocols, 2144 Types of Data Link Protocols... Flow Control Protocols. 2.18.1 Stop-and-Wait va. 2152 ua, Noiseless channels, 2464 2162 2163 Noley channels, 2174 2172 2173 Justity that the stop and wait protoco EON 245.4(A) Advantages and : Sing Wind aM20°S and Disadvantages of Stop-and. wat Unrestricted Simplex Protoco} Stop and Wait Protoco| 998 and Dis, Bnd Disadvantages of top-end. Wa Prot (oco} Simplex Protocol Piggybacking Stop and Wait Igorithm for computing the checksum. snot good for networking communication.222 223 226 225 226 (SPPU-New Syllabus w.ef academic year 2021-22) (P5-34) repeat ARO in dota (SELUSCERTE NAST ALINE Ua. Compare data flow control techniques, (SEENON RNAI va. ppose you have to develop an ¢ avery protocol for link that is example, unreliable and i ve, which ofthe following protocol would you to choose? Justify your answer, () Stop and v t secve repeat (i) GO BACK N (SEVERE NSTSRTNAT 2 HDLC (High-Level Data Link Control Ua. Explain bit oriented protocol for communication over po fer Modes. Framing in HDLC Explain control field of HDLC w.rt Lframe, S-frame and Usrame. Point to Point Protocol (PPP). 9.1 Services Provided by PPP 9.2 PPP Frame Format. Ua. _ Explain PPP frame format. 2.183 Transition Phases of PPP. Explain in detail,working of PPP with state transition diagram ? Mutiplexing Link Control Protocol (LCP), Media Access Control Sub Layer. Channel Allecation 221.1 Static Channel Allocation 2212 Dynamic Channel Allocation 221.3 Difference between Static and Dynamic Channel Allocation ALOHA (Additive Links on-line hawaii area) 222.4 Pure ALOHA 2222 Siotted ALOHA UQ. Prove that the slotted ALOHA performs better than Pure ALOHA. (SEUUECOMEIUMIENEE MONIES)... 2-4 fo point and multipoint link 2202(A) Advantages and Disadvantages of Slotted ALOHA. at 2.20.2(8) _ Important Points to Remember ‘ 2223 Difference between Pure and Slotted ALOHA 2-4 Ua. Differentiate between Aloha and Siotted Aloha. Dat cma. 2.42 Wavelength Division Multiple Access (WDMA). CSMAICA, Explain working of CSMA/CA with the help of flow diagram, (SGU NM KASLETESS) 2-44 va. Ua. Draw flowchart of CSMA/CA. (SAUER MEE APMC) CSMA/CD UQ. Write a short note on CSMA/CD. (STURT MI ATER RENEE UO. Explain CSMA/CD in detall ? (SLUCMTEM ENE ACA) 2264 Working of CSMA/CD 226,1(A) Collisions. 46 2.26.1(B) Late Collision. 2-46 Uarecn neo Pubicatons.AData Link Laye Bray Export rs iss Ethernet, Sai Networks. ca en Ethemet and LAN. t Network: Oe cnetetween ret and LAN ssp and Frame Fora ao neh MAC 802.3 frame format. EGGTUETCREIC SPEAR va, p30 Fast Eheret sy 30.1 Advantages of Fast Ethemet. Data Transmission Speed. Gigabit Ethemet 31.1 Gigabit Operation 2. Gigabit Ethemet Implementation 2313 Gigabit Ethemet Applicaton, 2s2 (EEE 802.11 ees 2301 Architecture of IEEE 802.11... LE AGRI, form on 2822 IEEE 602.11 Standards. : : mall 4 4 | 2.323 (EEE 602.11 Frame Format 20. ©epain 802.11 Wireless frame format ? (SUERTE MTR 6 Marks) : 2 2324 M 2 Transmission Techniques used by 802.11 4 ua. What are the transmission techniques used by 802.11 to send a MAC frame from one station to another? Explain two of them, = 2 IEEE 802.112 Standards and Frame Format ‘ aoe 2 2381 IEEE 202.114 Standards... Bare te : 2 UQ. Explain Frame format of 802.114 eee 2382 IEEE 602.114 Frame Format 236 IEE E 802.11b Standards and Frame Format . 2344 IEEE 602.11b Standards... ae s te 2 ( 2242 IEEE 602.11b Frame Format... eieteteeanenese Bt IEEE 802,119 Standards ... IEEE 802.11n Standards... ee n oa 2.6 237 IEEE 602.15 Standards... if P . 2374 IEEE 602.15.4 Frame Format. z a a 238 IEEE 602.16 Standards , Pisce : 228.1 IEEE 802.16 Protocol Stack... va, White short note on 802.16 Protocol Stack, USPPU - a, 6 (a) ETT 73 - Mf - Q. 6(d), 2- 2382 IEEE 802.16 Frame Format re anos 2 va. Draw and explain fra 239 Solved University Examples ee ome wee oe . Chapter Ends. " &) Tech-Neo P, Ublications..4 SACHIN SHAH VentuData Link Layer...Page no. ( Computer Networks p> 2.1 INTRODUCTION sts which are directly connected in some sense, This direct conn same link. The work of data fink layer tends Systems on broadcast network are said to be Data lnk layer has two sub-la Logical Link Control: It deals with protocols, ol, and error control a fi) Media Access Control: It deals with actual control of media 53 2.4.4. Functionality of Data-link Layer er does many tasks on behalf of upper layer. These are ) Addressing @)_ Synchronization Flow Control JSF Multi-Access em into Frames. Then, it sends each frame bit-by Framing : Data-link layer takes packets from Network Layer and encapsulates t bles them nto fram fer picks up signals from hardware and ass biton the hardware. At receiver’ end, data link I Addressing : Data-link layer provides layer-2 hardware addressing mechanism. Hardware address is ass ink. Its encoded into hardware at the time of manufacturing 3) Synchronization : When data frames are sent on the link, both machi Error Control : Sometimes signals may have encountered problem in transition and the bits are and attempted to recover actual data bits. It also provides error reporting mechanism to the sender. xd or capacity. Data-link layer ensures flow control that enables both yes must be synchronized in order to transfe flipped. These errors are detected (5) Flow Control : Stations on same link may have different spee machine to exchange data on same speed. (6) Multi-Access : When host on the shared link tries to transfer the data, it ha fh as CSMAJCD to equip capability of accessing a shared media among multiple Systems. probability of collision, Data-link lay mechanism s Wi 2.2 DATA LINK LAYER DESIGN ISSUES bits over communication channels, It has a numbe: i) Dea 1) The data link layer uses the services of the physical layer to send and rece defined service interface to the network layer Ling with transmission errors, functions, including : (i)Providing a wel Gil) Regulating the flow of data so that slow receivers are n Sending machine ot swamped by fast senders. ) To accomplish these goals, the data link Receiving ma layer takes the packets it gets from the seinork ayer and encapsulates thes ilo frames for ans 2) Fact fame consins «frame header, © pala felt for holding the puke, and frame trailer, as illustrated in Fig, 2.2.1, Frame management forms the heart of wha the data link layer does. ewig Payette) Tater + Frames and Packets 34) &) Tech-Neo Publications...A SACHIN SHAH Venture (SPPU-New Syllabus w.ef academic year 2021-22) (PSFrame isthe data link Bese : )) 2.3 _ SERVICES PROVIDED TO THE _ © NETWORK LAYER _ funtion of the data lnk layer isto provide ence ae © eamtering data frm the network layer on the source chine tothe network layer onthe destination machine are mactine i an ently, call ita process. It take from network ayer So that it canbe transmit to 0 the On the destination machine they are handed over to thé network layer as shown in Fig. 23.1(a). ‘The actual iansmisson follows the path of Fig. 23.1(), but itis easier so tank in tems of two. data ink layer’ processes communicating using a data link protocol, Host? Host Host2 ta 4 a gf 3 2 4 2 ee Actual z datapath “) a {Ft-23.1; Virtual and Actual Communication é Acual Communication ; tp hig Communication, physical Calm 18 preset though which Dan ‘Link Layer simply ansmis data frames. The ao Pata link tayer -> Physieg, Tayer on sending machine, then J, ,pivieal media and after tay 1, Physical Layer -> Data ink layer -> Network Layer on receiving machine, eee ae ol data layer protocol "| Source and destination MAC address. apsulated within [ Seament is encapsulat | Packet isthe network layer prot | source and destination IP address, Packet is encapsulated within afr ere ae layer offers three types of services a AY, Unacknowiedged connectionless service Y_Acknowledged connectionless service Acknowledged connectionless service > 9 “Onacknowledgedconnectionten sere 1; Here, the data link layer of the sending machine independent frames to the data link layer receiving machine, ihe receiving machine does not acknowledge rece the frame, se 3. No logical connection machines, is set up between the 4 Error and data loss is not handled inthis service *- Application : This is applicable voice communications, in Ethemet services Acknowledged connectiontess service Ir Hete, no logical connection is set Up between the b machines, but each frame ‘acknowledged by the desti 2. If the source does not Within a stipulated time, Application; services, Sent by the source machine ination machine on receiving the acknowledgm then it resends the frame This is used in Wirt (EE 802@ Set up of connection : A logical path is set up machines Buffers and counters are initialized to keep track of fram Sending frames : The frames are transmitted, ii) Release connection : The connection is released batffers and other resour re released, Application : It is app v2.4 FRAMING IN DATA LINK LAYER IQ. What is need of framing ? EEC EO Ire: ‘The data link layer, needs to pack bits into frames, so that, each frameis distinguishable from another. The Data Link prepares a packet for transport across the local media by encapsulating it with a header and a trailer to create a frame S) Framing is function of Data Link Layer that is used to to destination or separate message from source or send receiver or simply from all other messages to all other destinations just by adding sender address and destination address. or receiver address is simply used to to go and sender or 25° The destination represent where message or packet used. to source address is simply help recipient to nowledge receipt. ink layer that is points, It umes are generally data unit of data transmitted or transferred among various network p includes complete and full addressing, protocols that are essential, and information under control. Physical layers only just accept and transfer stream ore it is up without any regard to meaning or structure. Therefo to data link layer to simply develop and recognize frame boundaries, ) This can be achieved by attaching special types of bit Patterns to start and end of the frame, If all of these bit patterns might accidentally occur in data, special care is PU-New Syllabus w.e f academic year 2021-22) (P5-34) ©) Beamin, I thee el oes i should be framed int discernible Blocks of informati ith vl 2.5 DIFFERENT TECHNIQUES OF 7 SANG Character Count Flag Byte with Character Stuffing Starting and Endin with Bit Stuffing Encoding Violat 2.5.1 Character Count This method is rarely used and is generally d to count total number of characters that are present in fra be done by using field in header. Character to Count m ensures data link layer at the receiver or dest a total number of characters that follow, and al frame ends. Fig. 2.5.1(a) for four small example fr sizes 5, 5,8, and 8 bytes, respectively Disadvantage : There is disadva g this method i.c., if any how character count is disturbe distorted by an error occurring during transmission, the destination or receiver might lose synchronization. ‘The destination oF receiver might also be not able to, local identify beginning of next frame Example : if the byte count of Sin the second frame of F 2.5.1(b) becomes a 7 due to a single bit change, the destination will- get out of synchronization. It wtil then be unable to locate the correct start ofthe next fram Also if the checksum Is incorrect, the destination knows that the frame is bad, it til has no way of telling wh next frame st Sending a frame back to the source asking for retransmission does not help either, since the destination doe not Know how many bytes to skip over to get to the stat of he retransmission, For this reason, the byte count meth rarely used by itself [Breen Neo Publestons SACHIN SHAR2314 ———— Frame 3 8 bytes 12 3] 4]s[elel7 —— Nowa byte 2.2.5.2 Flag Byte with Character Stuffing Seema eaten Char fing i also known as byte stuffing or character-orie ‘sctually operates on bytes whereas bit stuffing operates on bits, ve the problem of resynchronization after an error by havi both the starting and ending delimit called a flag byte, is used as Two consecutive flag bytes indicate the end of one frame and t Just search fortwo flag bytes to find the end of the current fr, What if flag Byte occur in Data ? Problem with this technique : What ifthe flag byte oceu ‘ze being transmitted. This situation would interfere wi ‘count ((889Fig.2.5.1(a) : Byte Stream Without error (b) Byte Stream with errors SECs nted framing and is same as that of bit stuffing but bytes ing each frame start and end with special bytes. Often the fet. Shown in Fig. 2.5.2(a) as FLAG. we start of the next. Thus, ifthe receiver ever loses synchro rame and the start of the next frame. the data, est ith the framing, ig flag byte can be distin guished from one in the data by the absence or 8 before giving the data to receiving end removes the escape § What if ESC byte occur in Middle of Data? Payload field After stung URE ay Esc Eso}| Es Esc pecially when binary data such Presence of an escape byte befo layer. This technique is cl so | [FLAG] [Te 7 EAT ese) ree EF]225.3. Starting and Ending Flags, with Bie Seuffing Y popular HDLC digh Control protocol. Bach trame begins cvds with a spedal Bit pattern, CLUIIIO cr nent bit the out i cape byte Is stuffed into the outgoing charcicr seen fag byte in the data. It also ensure that help the pi B (Universal Serial Bus) uses bit stuffing for this atomatically destuffs (ie, deletes) pletely transparent tothe is bit stuffing. data contain the flag pattern, O1111110, this f tld as 011111010 but stored in t ™ Shown in Fig. 2 the boundary between two frames can be casily recognized by the flag pat n. Thus, if the receive they can on! 111111111110010 014111010010 t ‘stuffed bits’ onna14 (0) 049044441411111111110010 Fig. 25.3: (a) Original Data (b) Data On Li (6) Data after Destuffing that the length ith both bit and byte stuffing a side eff 8 frame now depends on the contents of the di 100 bytes: carries, For Tnstance, if there are no Mag Byle 100 bytes might be carried in a frame of rough (SPPU-New Syllabus w.e academic year 2021-22) (P5-34) q ti | 2.5.4 Encoding Violations t re 1 Inelud a Example : In the 4/5 gnal bits to ensun This means that 16 out of th d end of frame fl i viola delimit fam 4) In these technique they are reserved he start and end of frames and then data, Many data link p tse nethods for safe (8) A common pattern used for Ethernet and 802 frame begin with a well-defined patter ca preamble ist teader thas used to locate the en ofthe YA2.6 ERROR CONTROL EEE UQ. Explain the purpose of error control. fron perspective EQUECSTOM INET (1). Using above technique, the peoblen : frame 1d, Another probl 2) In case of unack ection! sender Just kept outputting frames ir Bu for it is an issue for reliable, connection-orient To ensure the rellable deliver some feedback to sender vi . 4) Typically, in this prot hi pecial control frames telling positive or _negati\ Y WY ‘ works 8 S00 computer NON ive tek pres 2 pe s-Comp) nity (SPRUSEN. 7 Sy west frame has arte SY sang ns ‘ is Lost Completely OF Acknow! ae a noise ne to lost completely (e.g. in 3 ‘rt rect at al Alife Ss 4, the sender will not know how tn -s system will get hanged faulty communication 1 Probien mi Tn above bh the tot with any reason ie functioning, Irs socted by introducing timers into the data < sender transmits frame, it generally also © sation is sett expire after an interval long enough for the reach the destination, be processed there, and have ledgement propagate back to Normally, the frame will be comet! meat will get back before the timer runs out, in wich case the timer will be cancelled the acknowledgement is los, the timer 8 the Sender about a problem. Only solution transit the frame agai, 2 Problem : However, when frames may be transmitted multiple times the receiver will accept the same frame two ‘mare times and pass it to the network layer more than FZ Setution : To prevent this from happening, it is generally Sarto assign sequence numbers to outgoing frames, so that Se ceiver can Gistnguish retransmissions from originals ————__ Y 27 FLOW CONTROL SO Bolan the pupose of tow _ PeCiVe control from DLL + Another important ‘esto transmit frames, thea This sta design issue when a sender faster than the receiver can accept on Can occur When th (ie Poet opr ade av ene ©) Satin Tac wi shore session ny Me commoaly used ieee Feedback based flow contol, i) Rate-based flow contol In feedback based Information tothe 4 at, Data Link Layer...P. G20 (a4, Penta ENEMES ATE EEN ALDEN compte and higher layers ao me eate-based ow control, the protocol ha te Dy 2.8 SUMMARY FOR DATA LINK Lay, eZ) ata-link layers the second layer after physical : two hosts or nodes. Sub-layers and their functions are a data link layer is divided into two sub-laye below. The (2) Logical Link Control Sub-layer (LLC) (1) Provides the logic for the data link , Thus it contol synchronization, flow control, and error checking function he data link layer. 2) Functions are ()_ Error Recovery, It performs the flow control operations. (ii), User addressing. (11) Media Access Control Sub-layer (MAC) (1) tis the second sub-layer of data-link layer. It controls flow and multiplexing for transmission _m ‘Transmission of data packets js controlled by this lay layer is-responsible-for-sending-the-data-over_the-netw interface card. (Functions are (“To perform the control of access to media. (iW performs the unique. addressing to stations dis 2M connected to LAN, (ii) Detection of errors. Design issues with data link layer areas follows (0 Services provided to the networ layer acts as The briniple service transfers data ffom merck on Sending machine to the n ‘Machine, This. (Dynamic Link Lit letwork layer on destina ransfer also. tal brary). kes place via DU w of ‘locks called frames to The starting and ending of ea ‘Wentiied So that the frame can te destination machine OW control is at the MUSE NOL send. SADECILY OF destin destination machine, frame should be ‘ecognized by the dW) Fy (W) Flow controt ; Fi done t pai " to prevent the No it Teeelver-end-"The source: machi a frames at a tate fas ation machine to Of data frame accept them:ie) Brror contrat - — _ Pat eek PLE® ERROR DETECTION AND ea eas a CORRECTION © error | 01) A conaition when the receiver's information does not mat hinectre and are mot aware of actual hardware data |. Shuttle reeves processing. Hi per layers expect errortree | ©) That means a0 bit may change 1 1 or = Most of the applications would not function as exp tion a expec «Applications such as voice and video may ‘not be that eid and with some errs they may stil funtion wel W3ZG.A_ Types of Errors 7 (GQ) stand explain with example ditterent types of errors There may be thee types of errors 2. Multiple bits error 3, Borst Single bit eror ZT. Single bit error Sent Received TDDET) => GET ET 8 frame, there is only one bt, anywhere though, vin s eorupt error (a) Single bit error 2. Multiple Sent Received Th bh => Hoo sé with more than one bitin corupted slate A. Burst error {b) Muttipe Bit error Sent Received 7PEb bE => Carr Frame coniains more than one consecutive bit corrupted (c) Burst eror (ieaFig. 2.9.1 : Types of Error To send some extra bits with our data to detect or correct the errors, These redundant by the ceiver. Error control mechanism may involve two possible Ways A], Error correction + To check exact number of bits that are corrupted and thet I Error detection ; To check if any error has oceurred or not. W (SPPU-New Syllabus w.e academic year 2021-22) (P5:34) [ab rech.nieo rubles \s.A SACHIN SHAH Ventureorrect getect and £0 mit 4 data without the redundant error aon a od, th som ata fae in is mi, I hie da ae Resch us aes hat the data it finds ha fr fom is 1g code that generates th dunn is nin endant ecti yl frame efore passing the message eaken er hers tothe uppe Ea Retransmission iver det o resend the message. tects the occurrence of an error and asks >s 2.9.3. Error Detection and Correction JQ)” Wite Shor note on ror Correction and Detection 23 2.9.3(A) Error Detection Si Ars BOLE * Ifthe following two conditions are satisfied then receiver can detect a change in the original code word. * The receiver has alist of valid code words * The original code word has changed to invalid ones, * Fig. 292 shows the process of error detection. Sender a code word ing data word as input and generator rules and procedure of encoding Receiver Data word Entract (Correct incorrect), Generator Discard nBils Word, then it lse discarded, a) Data Link Layer. an detect only single erro, r ‘This type of coding ca nore errors undetected. some popular techniques for error detection are ‘Simple Parity check. (b) Two-dimensional Parity check (©) Checksum. (a) Cyclic redundancy check 2.9.3(B) Error Correction SEED IETA In error detection, the receiver needs to know only 1 received code word is invalid; but in error correctio, receiver needs to find the original code w need more redundant bits for error correction, Fig. 2.9.3 shows the process of error correction. It is similar to detection only checker correct the en, instead of only detecting and discarding. Send Receiver Encoder Decoder k Bits [Data word Data word] «Bits Exact (Correct incorrect) Generator Correct EEE) (1343)Fig. 2.93 : Process of Ei 1 correction Be Dl 2.10 LINEAR BLOCK CODES : HAMMING 210.1 Hamming Weight of a CODE, HAMMING DISTANCE, PARITY CHECK CODE Code Word Iwo] Definition : The y, defined as the number of non ze Hamming weight of a code vec between that code word lammi weight of @ code word x ro elements in the code wor tor (code word) is the distan : ‘and an all zero code vectér. (a cat having al elements equal to zeroy, as Example Tx=1101010 then the aes then the Hamming weight As, the number of. ner of none; is4, “ro elements in the above code wot!Compuiter Networks & Security (SPPU-Sem 5. ome Data Link Layer. (2-13) Ye 2.10.2 Hamming Distance os Consider two code vectors (or code Words) having the gan © O1OH 11110) «3,4 (0101.11 110) number of elements + ‘The smallest hamming data Definition : The “Hamming dstance® between the two eo words is defined as the number of differences between tne oe of and y is denoted as xy gig Die foe deleton Walon tS code wo Code word2 = 11110 erroneous cade word may correspond to anoth bits 2, 4 and 5 are different from each ot word and errors cannot be detect Hamming distance is 3. + The error detection and correction capa echnigue depend on the minimum distance d 2S 2.10.2(A) Minimum Hamming Distance inthe Table 2:10.1 Meaning : It is the smallest hammi Table 2.10.1 : Role of d,,, for detection dist possible pairs and correction of errors is used to define the minimum hamming distance in a 2) dan. p to's” errors per word [a up rs per word ind this val , We find the hamming distances between all words and select the smalllest one. Conect up to "errors and. detec st ix Hamming distances are given below find minimum th example. 1 m>=3 is defin by the following equations: k The family of (n, k) hamming codes for an in Block length or code word length:n=2"-1 ; No. of message bits or data words =m=1 orn-m No. of parity bits: (nk) = m 2. That is the minimum number of parity bits is 3. Therefore, minimum hamming distance d,..=3. The code rate =k/n If m>> 1 then code rate r=1 and k = 4. This is the hamming code C (7, 4) with For example, if m=3 then n ling code structure ied between the data bits. * Itisemor detecting and correcting code. The parity / redundant bits are inse s used commonly but the concept can be extended to any number of bits | p4 | D3 2 | PL 7 | po | ps Pan] > fe Ep AE * The 7-bit hamming ec Fig. 2.10-1(a) : 7-bit hamn DIS Dis | D13 | D12 | Dil | DIO | DI PB Di [era | i Fig. 2.10,4(b) + 15-bit hamming code © (15, 11) t rech-Neo Publications...A SACHIN SHAH (SPPU-New Syllabus w.e f academic year 2021-22) (5-34 Da reen-neo rutData Link Layer... Comp) NY computer Networks & SeCUnty (spPu:Sem.5- 0 De Data bits; P= Parity its inserted at enc 34.2" bit where, n= O41 2, pa js at 2° = 4 and PB is a2 Parity bits ae Frnus plist 2c. IM bit P2 isat2 EP 7 bit hamming code Aaa ¢ transmitted. Therefore code word pa sume that 4 data bits are to be transmit re code a Dr] ses _ [fees Da AEE | 1) beeo | Ds=1 [BA] D3=1 20s EES) E®Section of parity bits This adjusted to 0 of 1 as to establish even parity over bits. ‘© ForP] Section of party bits are 1,3,5,7 ie. P1, D3, D5, D7 ‘© ForP2 Section of party bits are 2,3,6,7 ie, P2, D3, D6, D7 ‘© ForP4 Section of party bits are 4,5,6,7 ie, P4, DS, D6, D7 2 2.10.4 Solved Examples on Hamming Codes Ex. 2.10.1 : A data bit word 1011 is to be transmilted, Construct even parity 7- bit hamming code. © son. : > Step 1: The codeword format Given date bit = 1011 os=0 [p= [Ml] bs=1 [TA Step 2: Decide P1 23 > Step 3: Decide P2 ; — For P2, sections tobe considered are 2,3, 6,7 Here, we have to set P2 = 0 as 3, 6, Iz = 101 in order to have the even parity D6=0 | Ds Pa Pa | ps Y Step 4 : Decide P For P4, sections tobe considered are 4,5, 6,7 Here, we have 10 set Pa = 085, 6 D7 Thus, the code word which is rans Ot inorder to have the even parity D6=0 | ps=4 mitted to th I Teceiver = 101019) Fx.2.902: 47 bit hamm he corr ham ning code is, ved as 1110101, wh © Soin, Hamming code = 1 0101 ecb ae » Step 15 Check bis 4,567 Thus set P4, i ici for even parity ‘0 be odd party, Hence there is an error (SPPU-New Sylabus we academic year 29, (P5-34) yllabus Mic year 202 22) (P5.-3, ) Tech- N-N€0 Publications A SACHIN SHAH Vert”Data Link Layer...Page no. (2-15) piiter Networks & Security (SPPU-Sem.5.Comp) “Therefore, 6™Dit in the received code word is incorrect Thus, Invert and correc it by changing 00 1 Or 1 10 0 I Dizi | nse oGncaciviy [ost [SES] os a oe [os [oie [o13 [oe [Lon [oie | os [EB oy Los [os [ER] o> ERE EF Section of parity bits Ex 2103 EET ECR OMICS Justify Hamming code is error Detection and Correction Code, 21 Soin. Hamming codes are basically linear block codes. tis an error correcting code. The parity bits are inserted in between the as ae PL_| 7bit hamming code bit | bit | svie | vie | abi bit Where D-data bits and P- parity bits i coded to get the data back. * The hamming coded datas then transmited. At the receiver is # Thebits (1,35,7)(2,3.6,1)(1,3,5.7)(2,3.6.7) and (4,5,6,7)(4.5,6.7) ae checked for even parity or odd parity + Ifallthe4-bit groups mentioned above possess the even parity (or od parity) then the received code word is correet but ifthe Parity is not matching then error exis. * Such error can be located by forming a thre bit number out of three parity checks, seme, © First, to detect whether there are any errors in this received Hamming Code : Error Detection hamming code * The first step is to identity the bt postion of the data & all | Step 1 : For checking party bit PI, use check one and skip skip P2 the bit postions which are powers of 2 are marked as parity bits (eg. 1,2, 4, 8, et), The received hamming code of 7 bits P2 1 {7 | 6 | ps [ps 0 ‘one method, which means, stating from P1 and the take D3 then skip PS then take DS, and then skip D6 and take D7, this way we will have the following bit, DI pi | (SPPU-New Syllabus wef academic year 2021-22) (P5-34) fi Wa rech:teoPubications. SACHIN SHAH Venturee willl ‘xd 80 We = Aswan of pa for P2, we will on e checking for P2s We ee ek fr P2 ba WHE RE TT ye us th saree and £70 Dea checkin ft goats But ee pou nt be un is 2 out from paw we me ove th nee Sera Stealing for Me wil D sep 3:Cec fo eck ethod, which will give us the < check four and skip four method, Dr faowing data bis. But remember since we are checking f Mowe have seed our count fom PAPI & P2 should not be considera). 7] bs | ps | ps {i fo As we can observe that the number of 1's are odd then we wil wit the valve of P = 1. This means there is an error. xt number of 1's are even, then we = 0, Tis means there isno error. So, fom the above parity analysis, Pl & P4 are not equal to soweccan clearly say thatthe received hamming code has errors, Hamming Code : Error Correction Since received code has an error, It must be corrected. To rect the eros, use the following steps: Now the enor word E willbe: [pi] re Tai [1 To Now we have to determine the decimal value of this error MomiAOl whichis 522 #1 421+ 04 20 #1 = 5) We gE =, which tates that the exor is in the fifth data O% 70 comect it, just inven the fifth data bit So the comect data wil be: bs {05 [sos [pa [pi] ee Data Link Layer. Page foi = Posi or Parity/Data Bit_| Pi | P2 [es tet) es Bit Stream ijotot ty | ‘The Actual bit transmitted is then : O101 | Hamming Code has used Odd Parity Pi calculated using positions (1.35.7) = 101 P2 calculated using positions (2,3,6,7) = doo, P4 calculated using positions (4,5,6,7) = 1101 17th itis inverted then bit stream will be-> 1091 PI calculated using positions (1, 3, 5,7) = 1010 parity, so errr i detected). P2 calculated using positon 7) = 0000 (it’s not odd parity, so erro is detected). Py. tions (4, 5, 6, 7) = 1100 (it’s not odd parity, using po detected) So we discover that Parity at position 1,2 and 4 are income By adding this number 1+2+4=7, and that bit position 7, > location of the bad bit. By inverting the last bit from 0 to |. corrected at receiver's end. 3 2.10.5 Advantages and Disadvantages of a Hamming Code 'S Advantages 1. Hamming code method is effective on networks wice data streams are given for the single-bit errors. ( Hamming code not only provides the detection of abi but also helps you to indent bit containing error can be corrected. 3._The ease of use of hamming codes makes’ it bea & Suitable for use in computer memory and singlee correction: 1. Single-bit error detection and Correction code. Howe This can cause the ode algorithm can solve only Vemma S?PU-0. 405) pee ers A ao Ste 1001101 5s hosed a ing an hamming ‘code, Show transmitted, $y re 7 He ‘verted during ‘Yansmission, ‘show fee ies ms Someta the receiver's ng Pm elected ang Pang, Site bit seam is 109) 39) ‘arty positions are 2d te ain 2A, 80 ve mak tng a ing bits wil bed, bits as P lata bits, HF arity bits (SPUN (ew Syllabus ef academic year 2994 22) (P5-34) 2 20 Simple Parity check YQ 0 ) Sea extra bit is sent ‘long with the original bits (0 Sry Of Is cither even in Case of even parity, or od! “6 oF odd parity Shown in Fig, 2.14 (a) and (b) : (SP Tech-Neo Publications... SACHIN sHAH Vet!Computer Networks, Bata bis Toe ‘Security (SPPU-Sem §-Comp) > GBECEEo (ves9ig, 211.1(0) imple parity cheek {@) The sender while creating a frame counts the number of ts in (@) Ifthe number of Is bits is od, then parity bit 1 is appende and if the number of 1s bits is even, then party bit 0 is appended atthe end of the data unit 4) At the receiving end, the parity bit is calculated from the Dit. This technique itis known as even-parity checking red data bits and compared with the received parity enerates the total number of 1s even, so 2 2.11.1(A) Disadvantages of Simple Parity Check Iican only detect single-bit errors which are very rare. 2. _Iftwo bits ae interchanged, then it cannot detect the errors. Accept data } Reject data f011011 [Compute Compute Parity bt parity bit 011011] 4 or1014] 4 Transmission | media (1860 ig. 2.11.1(b) : Flowchart of working of simple pari check 7%. 2.11.1(B) Two-Dimensional Parity Check « Performance can be improved by using T' Parity Check which organizes the data inthe form ofa table Shown in Fig. 2.11.2. (SPPU-New Syllabus w.ef academic year 2021-22) (5-34) Data Link Layer... Page no. (2-17) 2) Parity check bit ted for each row, which i ate comp le-party check @) In Two-Dimensional Party cheek, a block of bits is divided whole block. (4). At the receiving end, the party bits are compared with the parity bts compited from the received data, (5) _2 sets of parity bits generated known i) Longitudinal redundancy bits (LRC) al redundancy bits (VRC) B (6), LRC indicates the parity of rows and VRC indicates (ii) Vert party of columns. Original data [44001110 40111010 01110010 o1010010 11001110 404114 01014 Row parties 0111001010 0101001 olf Ono tomo Column parities (asnFig. 2.112 : Two yensional parity check ‘WS 2.11.2 Longitudinal Redundancy Check (uRc) (1) Longitudinal Redundane) Check (LRC) is also known as 2-D parity check (2) In this method, data which the user want so send is organizs into tables of rows and columns. Shown in Fig. 2.11.3. (3) Meaning’: A block of bit is divided into table or matrix of rows and columns, In order to detect an dant bi is added to the whole block and this block is transmitted (4) The receiver uses this redundant row to detect error (5) After checking the data for errors, receiver accepts the data and discards the redundant row of bits (6) Example + If a block of 32 bits is to be transmitted, it is divided into matrix of four rows and eight columns which as shown in the following figure [el rech.eo Publications..A SACHIN SHAH VentureData Link Layer... = sees compute enputerN 100 11100011 = frremrore rorororo 110011 == Disad' (1) The m 11001010 error i 401010140 exactly 11001100 — 11100014 o1o01¢ Redundantrow/LRC 01004 414 4 ot . — [aa ae sore Tea oe To Fig. 2. receiver Parity bits/LRC_ Fig 2.113 : Longitudinal Redundancy Check oe Also Oar * iy Bit (dor even) i ealeulated foreach column. I means 32 bts data plus 8 redundant bis ‘tansmitted to receiver. mm 2.11 (© Wiss dtreaches the destination receiver wes LR to detect error in data % 2:11.2(A) Advantage and Disadvantage of LRC © Advantage (0) ARC is med w detect bust errors. seat: Seppose 32 bit dat pls LRC tha wa bei, Ee 1s Beng transite is hit by a burst err of length 5 and some bits are comp Data + LAC To e Hit by Receiver [11001010 10101014 410111100 14100014 01001111 | burst error of length 5 OO1n, 444 0 TT ‘AC computed at destination TRC received at ‘destination yee a [PPL New Stab Wel aca1 Disadvantage | eat alt + on + ee EEE) Fig 2115: Two bits at same bit postion damaged in : ys. 2.11.3 Vertical Redundancy Check (VRC) or Parity Check (GQ. Explain the VRC with example. icdundancy Check is also known as Parity Check. redundant bit also called parity bit i cach data unit. This method includes even parity Shown in Fig. 2.11.6. added Meaning : Even parity means the total number of 1s in data fe even and odd parity means the total number of 1s in dats i to be 0d. O the source wants to transmit data unit 1100111 city 1o the destination. The source will have to | ough Even Party Generator. 5) Parity generator will count number of Is in will 28d parity bit. Inthe above example, number of 1s in data unit and 5, party generator appends a parity bit 1 2.41.3(A) Advantage and Disadvantage of vec Advantages VRC can detect all single bit error. Itcan also detect burst error Disadvantage The major disadvant detection is that it is not able to detect burst exror number of bits changed i 4,6 Example : Ifthe original : z data unit that will b , way 2bits are 01011111 When this data will react Destination will a though data aTafefolsT[som nal number of 1s even ic. 6 which i | 1] 7H Fig 2.116 : Even parity VRC (SPPU-New Syllabus we academic year 2021-22) (P5-24) ‘Tei Fig. 2.11.7 : Erroneous data accepted by receiver with number of changed bits 2.11.4 Drawbacks of 2D Parity Check If wo bits in one data us the same position in another data ws AD Parity checker will not be able This technique cannot be used to detect the 4-bit errors [Ba rech- neo PubComp), som’ sep = peoded usin VET merit i resem 10110010, anjo1011 11100001 peck ERC parity bis find wrong parity nas SSS hs eine wy. pvt and 11001 nano oem { ot 101010 apoio 00101 vas vertically and o eae ry is vertically a vat and VRC a = Byte Tele es Baya eaeien re 0 | opitin | ia iss 0 1 o 0 0 | wrong see ee ee | | | party | which sa wrong party (odd parity), ofthe $* bye isin ero. v1 | 2.12 CYCLIC CODES T vRC and @ % 212.1 Cycle Redundancy Check (CRC) R_Eplan RCwih omni RCs edndaney enor tech fer. Folowing are the a Stow in Fig. 2121 ique used to determine the PS used in CRC for error detection its ina predetermined numbe, +1 bits » Siep 11: Seconly, ty ma 7, Be Devly extended data is divi, % Nog a proces, is known seman BeterHed fom tis dvs led by a 8 binary division, The sion is known as CRo places the appended WY Benerted uni ig ) CR remainde, Remainder} Zor, aecopt |nénzoro, rect & Sender Receiver (1859)Fig. 2.12.1 : Working of CRC Step VI : If the resultant of this division is not zen means that the data consists of an error. There! discarded. Step VII: Let's understand this concept th an exany ‘Step VIII : Suppose the original data is 11100 and diy 1001 2B 2.12.1(A) CRC Generator A CRC generator uses a modulo-2 division. Firstly, te zeroes are appended at the end of-the data as the the divisor is 4 and-we know that the len to be appended i divisor lengt igth of the stig always"One"Iéss than’ the length of Now, the string becomes 11100000, and the resultant stig divided by the divisor 1001 remainder is 111. (Refer Fig 1444 11700000 too1 10044} 110 1004, 1110 1001 2112.2 1001 , (SOF Ie. 2.122: CRE replaces the ompute Zs 2A @ The gene whe then Gi) Ast Int The @ Gi) Hicomputer Networks & Secunty (SPPU-Som, Comp) Data Link Layer.;Page no. (2-21) Ya 2.12.1(B) CRC Checker They 5 it defined mathematical structure whic When the string 11100111 is received at the receiving end Wid lata BSR a then CRC ched ns the modulo:2 divisioy haa Ras hs hat i A string is divided by the same divisor, Le., 1001 Tee ere ay A ed 5. Its possible detet the enor bursts using the cyclic ed In this case, CRC checker generates the remainler of zero, Therefore the data is accepted "Disadvantages ttt 1. Ben though the error detection is simpler, the error Unit correction slightly mon 4 This is due tothe complexity ofthe combination logic circuit ((Be0)Fig. 2.12.3 : RC Checker 2 2.12.1(C) CRC is Appended at Trailer Not at Header $ _atrallerrather than in a header ? ea the CRC at the end of a frame reduces packet latency and reduces hardware buffering requirements. On the transmit side, hardware can read and transmit bytes of the frame immediately. The transmitter ealeulates the CRC on the fly as data p through, then simply appends the CRC the tal of the fran Consider the alternative where the CRC comes somewhere in the Ethernet header. Gil) Hardware must read and store the entire fran calculate the CRC. This amounts to a large look-ahead operation and adds significantly to transmit latency and hhardware cost. The situation also becomes more complex for the receiver as well. 3. 2.12.2 Advantages and Disadvantages of Cyclic Codes © Advantages in order to The advantages of cyclic codes over most of the other codes areas under 1. They ate easy to encode, (SPPU-New Syllabus w.e f academic year 2021-22) (P5-34) {UQ._ Why does the data link protocol always put the CRC in! 2.12.3 Solved Problem ex. 2.424 The data word 1101011011 is to be’ sent using. generator Polynomial x! +41 Sender side. OR A bit stream 1101011011 is transmitted using the standard CRC method. The generator polynomial is xf-+x + 1. What isthe actual bit string transmited ? {Use CRC to compute the code word atthe © sotn. : The generator polynomial Gx) = 10011, Clearly, the generator polyno The resuting bitstream is 110101101 10000, Now 1100001040 11010110110 foots ded a he binary 10011 AAO — Remainder From here, CRC Talrech:tco pubications.A SACHIN SHAH Ventureond 10 code Wo reso 1101011011 ete site vord_ ans wast + Ts te 2 yoo1t0r1108 af am 1001170 pias att a esd ORC mtd. The pene Pol {watt acl BSE APs invented during K "oe, fom the et iin 2 Sem on How wl ceiver detec hi ison: pert=1 ‘he pera polynomial eChey, he gern polynomial consists of 4 its. © So wing of 3 eres is appended tothe bit steam to be snes + Therein bit seam i 10011101000, + Now te ary vison is performed a- 40001100 soo1Jto011101000 1001f 0004 0000, oot 0000 0 0 (i) onde 1 is transmitted using the ynomial is x°+ 1 3 + 1is encoded as 1001 4100 —> CRC remainder From bere, CRC = 199, Now, The code wi re, Mord 10 be transmitted is obtained by on 2280 of 0011101000 wih te Cae? PEIN the ts the Code word i 10011101109," ¥ B18 transmited 0 the recciver = Part-2 Accontng othe question, Third bit from Receiver receives * Recsiver peony. tp SE0¢Faor Polynomial ag the bit steam = ———T_ EE aaa UEx. 2.12.3 UOTE CEE ) ‘Consider a message 1101001 1101100, divisor 1011. Compute binary CRC. © sotn. : Data Link Layer. 70101000 Page n (2, 1001 —> CRC remainder From here, The remainder obtained on division is a non-zero valu. indicates to the receiver that an error occurred in the during the transmission. Therefore, receiver rejects te ig and asks the sender for retransmission. Message 1 1 0 1 Divisor 10 1 1 3 zer0s to be appended (divisor -1) 1141000411110 110100111401100000 1044 1100 1044 1 4 0Ol1L101100 1014 oF 00—~ Remainder E 1 1