epson “Obi, me ‘% © es VOUSOL » Yotal Solution visit ; www.totsoLimfo email: to_arnabch@satyam.net.in Phone/Fax: (033) 2577-3821,(033) 2578-3034, Mobile: 98300-37439.‘TOTSOL - Total Solution totsol info. DCN +1 MESSAGE TRANSPORT TECHNOLOGIES, CIRCUIT - SWITCHED , MESSAGE ~ SWITCHED , AND PACKI NETWORK: SWITCHED Networks May be conveniently classified by the techniques employed in trans- porting messages between the nodes. We first introduce these classes of net- works, and then describe their comparative advantages and disadvantages. A Circuit ~ Switched network transmits a message by providing a complete path of transmission Links from the message origination node to the destination node. This path is set up by a special signaling message sent by the origin node to the destination node. A response to this signaling message from the destination node informs the originating node to proceed with data transmission. The data is trans- mitted progressively over all the channels in the path with no intermediate store ~ and - forward delays. The entire fixed — delay path is allocated to this transmis- sion, until the sender releases this path. A message - Switched network transmits a message among the nodes by moving the message through various transmission links and message buffers. A mes- sage is stored and then transmitted to the next node along the message path. A message transmission from a node does not start until a buffer at the next node on the route has been allocated for it. The path ( or route ) for message transmis- sion may be fixed, or it may be determined dynamically as the message progresses toward its destination node. This class of networks is also called store - and - forward networks, since message are stored in each node and then forwarded to the next node on its route. Due to the message storage in the intermediate nodes on a message route, some nodes may experience massage queuing and conges- tion. An important attribute of a message ~ switched network is the wide varia- tion in message delay through the network.TOTSOL- Total Solution totsol inf DCN - = A packet - switched network differs from a message switched network in that long message are first decomposed into fixed - size segments called packets. These packets independently traverse the network until they reach the desired node, where they are reassembled into the corresponding message. Thus, many packets of the same message may be in transmission simultaneously, thereby providing an important advantage of packet switching : the pipelining effect. Packets belonging to a particular message are held in the memory buffers of a destination node until all the packets required to reassemble the entire message the entire message have arrived. Thus, besides the store - and - forward buffers, nodes in packet - switching networks ( that allow packets to get out of order | also require reassemble bullers. In some cases, especially when all the packets of a message traverse the same physical path, out - of - sequence packets are discarded at the receiving node. @ © Serial polling Here the primary station ( PS ] maintains a list of all secondary stations ( SS } The PS sends a poll to each SS, one at atime. The PS includes the address of the poll message. If the polled SS has no data to send, it sends a negative response. If the station has data to send, then it sends a positive response, following by data. Next, the PS sends a poll to another SS in the list. Here the principal problem is to bear the over head of processing of non — productive polling. © Hub polling : The sent poll from PS is received by first SS si. If s: has no data to transmit it sends a poll to the next SS, on behalf of PS. This process is repeated by s2 and so on, until a SS is found which has data to send. Such SS will send data to PS and the process terminates. The PS resumes with next SS transmission. Otherwise the process continues with the last sn where the poll terminates. The PS resumes. This method is useful when there is a long distance between PS and SS; . this method improves response time. Transmission Simultaneous Number of wires direction transmission * Simplex Two One No * Half duplex Two Both No © Four wire half duplex four Both No © Full duplex or duplex four Both Yes * Separate wires may be used for control signal transmission. Two wire facility may be used for full - duplex with FDM for two directions.DOTSOL - Total Solution totsol info DCN -3 ‘Comparison of the characteristics of circuit — switched message — switched, and packet — switched networks: Description Value Circuit - Message - Packet — Code* Switched Switched Switched 1] Allow different Speed links = No Yes Yes 2_| Provide pipelining Effect = No No Yes 3. [Can change Format/Code during store- * No Yes Yes and- forward transmission process 4 [Can adaptively select faster path for + Ne Yes Yes transmission as a function of network congestion 5, | Can wansmit Shorter messages first due to ¥ No No Yes fewer packets ©_| Can provide Ervor-contral = Some Yes Yes 7._| Require buffering 5 No Yes Yes & | Possible queving and congestion in the - NO Yes Yes Communication sub network 9 [Require routing and Flow-control - Ne Yes Yes procedures TO. | Require sequencing and aumberin = No Some Yes TL | Require Packetizing - Segmentation and ~ No ‘No Yes Reassembly overhead 12 | Applicability to time dependent uses ¥ Yes (Under Study) such as voice * Plus and minus sign represents an advantage and a disadvantage respectively. | @ Except when setting up message transmission path.DOTSOL - Total Solution awww totsol info DCN }__—— svc tr primary station s1 Pp Selection p__Selection _, 5 Ready a ‘Transmit data ‘Ack — s4 Poll Transmit Data ‘Ack P ‘Ageneral multipoint discipline (in serial polling) primary station Secondary stations (Hub) Poll No data to send Pal” _, 5 No data to send —Poll_, 3 5 Data Data to send’ Ee Poll *Porwarded poll ee Hub Polling.DOTSOL - Total Solution totsol inf ec 480 Hz guardband 300 bps channel y 00 - 1013 Hz 1493 -2206 He 2686-3400 Hz 480 Hz guardbant Guardbande used in FD (a) Using FDM Host Moder n\n (b) Using TDM Modem Terminale (b) Network using TDMs Satelite aN Messare transmission with oie time slot for each earth- station / 7 7 esuth:DOTSOL - Total Solution totsol info DCN - 6 Host © Features of Modems s Speed, Automatic dialing / Reading, Auto- T Modem matic answering, self-testing, voice-over- D data and others are : i (1) Auto-Disconnect. t (2) Manual Connect / Disconnect. (3) Speaker. 1200 bps (4) Full or Half - Duplex. (5) Reverse Channel. (6) Synchronous or Asynchronous ‘Transmission, 9600 bps (7) Multi - port. s % D M Modem va Terminals 1200 bps & Network using STDMs: Modem Modem }— M M u U x x Fo FM Host u Modem x Fig: Cascading Multiplexers: M u fp xe ModemDOTSOL - Total Solution DTE Modem totsol info. Telephone Network Modem DTE T) i DIE proneiteh _. To other i {Modulator} > Modi Digital Line i cous interface interface, i Receiver (Demodulator} Telephone instrument Fig: Building blocks of a Modem Request to transmit data Incoming Data Incoming carrier indication, Incoming = carrier indication Received Data + (a] Half-Duplex Modem Request to transmit data Data “~ Outgoing Incoming carrier oe (b] Full-Duplex Modem: oe Modem f j---> Start bit (a] Asynchronous Modem Received Clock-¢. Modem [> _ Received 4 ‘Transmit clock is supplied either data by Modem or by the DTE (b] Synchronous Modem‘TOTSOL - Total Solution totsol info. DCN ~ FREQUENCY DIVISION MULTIPLEXING ( FDM) Users of existing voice grade lines (phone lines) can multiplex low speed circuits into the standard voice grade channels by using FDM. In FDM, a modem and a frequency division multiplier are used to break down the frequency of available bandwidth of a voice grade circuit, dividing into multiple smaller bandwidths. The bandwidth is the measure of the amount of data that can be transmitted per unit of time. It is determined by the difference between the highest and lowest allowed frequencies in the transmission medium. ‘Assume that a telephone circuit has a bandwidth of 3100 Hz, and a line capable of carrying 1200 bits per second (bps). Suppose that instead of running a terminal at 1200 bps, it is desired to run 3 terminals at 300 bps. If three terminals are going to use the same communication line, then some type of separator is required to avoid cross talk (interference of signals from one to another). This separator is called guard band, For transmission at 300 bps the standard separation is 480 Hz, Therefore, in this situation 2 guards of 480 Hz each is required. Since the guard bands now occupy 960 Hz, and the original bandwidth is of 3100 Hz, then the frequency left for the 300 bps transmission is 2140 Hz. If three terminals are required, then 2140 Hz divided by 3 gives a frequency of 713 Hz to be used per channel. With FDM it is not necessary for all lines to terminate at a single location, Using multidrop techniques, the terminals can be stationed in different locations within a building or a city. ‘TIME DIVISION MULTIPLEXING (TDM ) Time division multiplexers are digital devices and therefore sclect incoming bits digitally and place each bit into a high speed bit stream in equal time intervals. The sending multiplexer will place a bit or byte from cach of the incoming lines into a frame. The frames are placed on high speed transmission lines, and a receiving multiplexer, knowing where each bit or bytes at appropriate speeds. ‘Time division multiplexing is more efficient than frequency division multiplexing, but it requires a separate modem, To the sending and receiving stations it always appears as if a single line is connecting them. All lines for time division multiplexers originate in one location and end in one location, TDMSs are easier to operate, less complex, and less expensive than FDMs. STATISTICAL TIME DIVISION MULTIPLEXERS (STDM) In any terminal-host configuration the terminals attached to the host CPU are not always transmitting data, The time during which they are idle is called down time. Statistical time division multiplexers are intelligent devices capable of identifying which terminals are idle and which terminals require transmission, and they allocate line time only when itis required. This allows connection of many more devices to the host than is possible with FDMs and TDMs. ‘The STDM consists of a microprocessor based unit that contains all hardware and software require to control both the reception of low speed data coming in and high speed data going out. Newer STDM units provide additional capabilities such as data compression, line priorities mixed speed lines, host port sharing, network port control, automatic speed detection, internal diagnostics, memory expansion and integrated modems, The number of devices that can be multiplexed using STDM depends on the address field used in an STDM frame. If the frame is 4 bits long, then there are 16 terminals (2 in the power of 4) that can be con- nected. If bits are used then 32 devices can be connected.DOTSOL - Total Solution totsol info. alt] Digital Encoder xit) Decoder alt) or >| Digital Analog at aft (a) Encoding onto a digital signal mit} Digital s(t} oe Modulator Analog ‘Analog | Decoder of I, (b] Modulation onto an analog signal Encoding and Modulation Techniques: Key data transmission terms ‘Term Unit Definition Data element bits ‘A single binary one or zero Data rate bits per second(bps) ‘The rate at which the data elements are transmitted. Signal element Digital. a voltage pulse ‘That part of a signal that occupies of constant amplitude the shortest interval of a signaling Code ‘Analog: a pulse of constant Frequency, phase and amplitude Signaling rate or Modulation rate signal elements per second(baud) The rate at which signal elements are transmitted 5 bits = 5 usec D=R/b=R/ loge where, D = Modulation rate, baud NRZI R = Data Rate, bps L= No. of different signal — { { elements !1 bit = 1 signal element = 1 psec b= No. of bits/signal element Manchester A stream of ones at 1 Mbps: —> > 1 bit = 1 psec 1 Signal Element = 0.5 useco 1 60 0 1 1 0 0 0 1 NRZ-L, RZ Bipolar-AMI Pseudo-ternary Manchester Differenti: Manchester TOTSOL- Total Solution totsol inf DCN 10 Non-return-to-zero-Level{NRZ-L) 0 = High Level 1 = Low Level Non-return-to-zero-Inverted(NRZI): 0 = no transition at the beginning of interval(one bit time) 1 = Transition at the beginning of interval Bipolar-AMI: 0 = No Line signal 1 = Positive or negative level, alternating for successive ones. Pseudo-ternary: 0 = Positive or negative levels, alternating the successive zeroes 1 = No Line signal Manchester: 0 = Transition from High to Low in middle of interval 1 = Transition from Low to High in middle of interval Differential Manchester: Always a transition in the middle of interval 0 = Transition at the beginning of interval 1 = No transition at the beginning of interval Digital signal encoding formats A Manchester Stream‘TOTSOL - Total Solution totsol infor DCN ‘Amplitude shift keying NN 3) Acos(2nft) Binary 1 0 Binary 0 Frequeney-shift keying s()= J Acos@nfit) Binary 1 Acos(2nft) Binary 0 HUNK? FIGURE 4.7 Modulation of analog signals for digital data. om 00% onto. s(t) = Acos(2aet+ 45°) 11 Acos(2nkt + 135°) 10 Acos(2nft + 225°) 00 Acos(2nft+ 315%) O1 Phase-reference ae signal (previous 4 bits) 1101 vou segeee wee ue vue” ps — Se Mt A LMS SHUIL‘TOTSOL - Total Solution totsol info. * Analog Data, Digital Signals (NRZ-L) Digitizing analog data. Modulator)» DCN -12 ‘Analog signal TASK) 6 ‘ ‘4 3 3 [ iil w | on oor 110 oor on 110 7100 PCM pulses, PCM output ; 011001110001011110100 2ulse - code modulation PAM Cuantizer Encoder ‘Continuous-time * | Discrete-time Discrete-time Digital bit- continuous. continuous acre. ream ouput ‘mpiitude (analog) signal fampltede signal signal input signal (PAM pulses) (PCM pulses) \nalog - to - digital conversionTOTSOI.- Total Solution totsol info. DCN Signal amplitude ‘Analog. input Slope overload Quanizing uirease oe function 13 Time Sampling time Delta 1 modulation output 0. ‘Analog >| comparator a S < a OO] inpat L [Reconstructed Waveform. wit day -— (0) Reception et modulation One time Reconstructed unit delay waveform (4) Transmission SQ)=[1 +m» O)] cos 2akt (4) Sinaia modulating wave () Resulting AM signal Amplitude Modulation‘TOTSOL - Total Solution totsol info. DCN 14 Angle Modulation Basic Definitions : Let 6(t) denote the angie of a modulated sinusoidal carrier, assumed to be a function of the message signal. We express the resulting angle modulated wavaas s(t) = A, cos[6,(#)] where A, is the carrier amplitude. A complete oscillation occurs whenever 0,(¢) changes by 2zr radians. If 6,(1) increases monotonically with time, the averdge frequency in Hertz, over an interval from {to t + At, is given by _ at + a — a0) Sal) = = ‘We may thus define the instantaneous frequency of the angle-modulated signal s(t) as follows: S(t) = him fit) ano = tim [Met 49 = Bo 40 QnAt _ 1 d8xt) Qn at Thus, according to Eq. (4):, we may interpret the angle-modulated signal s(t) as a rotating phasor of length A, and angle 6(t). The angular velocity of such a phasor is dé,(1)/dt measured in radians per second, in accordance with Eq. (3):. In the simple case of an unmodulated carrier, the angle 0,(t) is 8(t) = 2nft t+ o, and the corresponding phasor rotates with a constant angular velocity equal to 2nf',. The constant ), the value of Q,(t) at t=0. ‘There are an infinite number of ways in which the angle Q(t) may be varied in some manner with the message (baseband) signal. However, we shall consider only two commonly used methods, phase modulation and frequency modulation, as defined below. 1, Phase modulation (PM) is that form of angle modulation in which the angle Q,(t) is varied linearly with the message signal m(t), as shown by : 0 Inf, +k, ——‘TOTSOL - Total Solution totsol info. DCN ~ 15 The term 2a7f,t represents the angle of the unmodulated carrier; and the con- stant k, represents the phase sensitivity of the modulator, expressed in ragians per volt on the assumption that m(1) is a voltage waveform. For convenience, we have assumed in Eq. (4) that the angle of the unmodulated carrier is zero at t = 0. The phase-modulated signal s(?) is thus described in the time domain by S(t) = A, cosl2afit + kym(t)] 2. Frequency modulation (FM) is that form of angle modulation in which the instan- taneous frequency f(t) is varied linearly with the message signal m(t), as shown by S(t) = f+ y(t) oe The term f, represents the frequency of the unmodulated carrier, and the constant k, represents the frequency sensitivity of the modulator,’ expressed in Hertz per volt on the assumption that m({) is a voltage waveform. Integrating Eq. (6) with respect to time and multiplying the result by 27, we get 5) a0) = 7) aft + ant, [ m(t)dt —— 5 where, for convenience, we have assumed that the angle of the unmodulated carrier wave is zero at t = 0. The frequency-modulated signal is therefore described in the time domain by s(t) = A, cos [2mf,t + on, | m(t) dt] (8) A consequence of allowing the angle 6,(1) to become dependent on the message signal m(t) as in Eq. (4) or on its integral asin Eq. (7) is that the zero crossings of a PM signal or FM signal no longer have a perfect regularity in their spacing; zero crossings refer to the instants of time at which a waveform changes from a negative to a positive value or vice versa. This is one important feature that distinguishes both PM and FM signals from an AM signal. Another important difference is that the envelope of a PM or FM signal is constant (equal to the carrier amplitude), whereas the envelope of an AM signal is dependent on the message signal. These differences between amplitude-modulated and angle-modulated sig- nals are illustrated in Fig. (1) for the case of sinusoidal modulation. Figures (1.a) and(1.b) refer to the sinusoidal carrier and modulating waves, respectively. Figures (1.¢; 1.d and 1.e) show the corresponding AM, PM, and FM signals, respectively. These waveforms indicate that a distinction can be made between PM and FM waves only when compared with the original modulating signal, which shows that there exists a close relationship between PM and FM signals. Furthermore, comparing Eq. (5) with (8) reveals that an FM signal may be regarded as a PM signal in which the modulating wave is f5, m(dt in place of m(t). This means that an FM signal can be generated by first integrating m(t) and then using the result as the input to a phase modulator, as in Fig. (2.a) Conversely, a PM signal can be generated by first differentiating m(t) and then using the result as the input to a frequency modulator, as in Fig (2.b) We may thus deduce all the properties of PM signals from those of FM signals and vice versa. Henceforth, we concentrate our attention on FM signals.‘TOTSOL - Total Solution vtotsol info DCN 1 @ e fe — time Figure (1) Illustrating AM, PM, and FM signals produced by a single tone. (a) Carrier wave. (b) Sinusoidal modulating signal. (c) Amplitude-modulated signal. (d) Phase-modulated signal. (e) Frequency-modulated signal.‘TOTSOL - Total Solution totsol info. DCN 17 Phase Integrator FH msdatstor FFM wave ° ft A608 (eft) te Modulteg sloitereniatorf—of F°S0%PY LS put wave f Accos (eft) ow Figure (2) Illustrating the relationship between frequency modulation and phase modulation. (@) Scheme for generating an FM wave by using a phase modulator. (b) Scheme for generating a PM wave by using a frequency modulator. FREQUENCY MODULATION ‘The FM signal s(t) defined by Eq. (8) is a nonlinear function of the modu- lating signal m(2), which makes frequency modulation a nonlinear modulatin proc: ss, Consequently, unlike amplitude modulation, the spectrum of an FM signal is not related in a simple manner to that of the modulating signal; father, its analysis is much more difficult than that of an AM signal. How then can we tackle the spectral analysis of an FM signal? We propose to provide an empirical answer to this important question by proceeding in the following manner: + We consider the simplest case possible, namely, that of a single-tone modula- tion that produces a narrow-band FM signal. + We next consider the more general case also involving a single-tone modula- tion, but this time the FM signal is wide-band, We could, of course, go on and consider the more elaborate case of a multitone FM signal.® However, we propose not to do so, because our immediate bbjective is to establish an empirical relationship between the transmission bandwidth of an FM signal and the message bandwidth. As we shall subsequently see, the two stage spectral analysis described above provides us with enough insight to pro- pose a solution to the problem. Consider then a sinusoidal modulating signal defined by m(t) = Ay cos(2fqt) ate.‘TOTSOL - Total Solution totsol info. DCN - 18 The instantaneous frequency of the resulting FM signal equals St) =f, + hpAg, cos(2mf,,t) -—-—-———-—(10) = + Af cos(2mf,t) where AS = hyn (11) The quantity Af is called the frequency deviation, representing the maximum de- parture of the instantaneous frequency of the FM signal from the carrier fre- quency Jf. A fundamental characteristic of an FM signal is that the frequency deviation Af is proportional to the amplitude of the modulating signal and is independent of the modulating frequency. Using Eq. (40), the angle 6,(:) of the FM signal is obtained as a(t) = anf soa ——(12) = Qaft + & ancanfan The ratio of the frequency deviation Af to the modulation frequency f,, is com- monly called the modulation index of the FM signal. We denote it by B, and so write af as 13) and 0(t) = 2aft + Bsin(2nf,t) —— (14) From Eq. (14) we see that, in a physical sense, the parameter B represents the phase deviation of the FM signal, that is, the maximum departure of the angle 6,(t) from the angle 27f,t of the unmodulated carrier; hence, 8 is measured in radians. ‘The FM signal itself is given by s(t) = A, cosl2afit + B sin(2mf,0)] ——————(15) Depending on the value of the modulation index B, we may distinguish two cases of frequency modulation: + Narrow-band FM, for which is small compared to one radian. + Wideband FM, for which B is large compared to one radian. These two cases are considered next, in that order. row-Band Frequency Modulation Consider Eq. (15) , which defines an FM signal resulting from the use of a sinusoidal modulating signal. Expanding this relation, we get s(t) = A, cos(2f,t)cos[B sin(2mf,0)] — A, sin(2mf.t)sin[B sin(2m7f,,)) ———‘TOTSOL - Total Solution totsol info. DCN 19 Hammii le (SECDED Code) ‘SECDED: Signal error correction double error detection. Parity bit |» Even parity (total no of I, in the message and the parity bit is even) ‘+ Odd parity ((otal no of I, in the message and the parity bit is odd) Let us use even parity logic in the following example: Sender station Receiver station Sender station | Receiver station ‘Transmission ‘Transmission 10111 ui 1 10111 101 0 ized 1 De Ds Di Ds Dr Dy Parity bits: Py Ps Pr Py Error neous place no. in decimal | Parity bits Inference / Comment PoP PP Po Rarity bit (PI error) | 2 — _ 1 __ [Parity bit 2 (P2 error) 3 | 1_| Data bit 1 (D1 error) 4 1 Parity bit 3 (P3 error) 5 1 1__ Data bit 2 (D2 error) 6 1 1 Data bit 3 (D3 error) 7 1 1 1 Data bit 4 (D4 error) 8 1 Parity bit 4 (P4 error) 9 fa _____1_| Data bit 5 (D5 error) 10 fa _ 1 | Data bit 6 (D6 error) i 1 1 1 __| Data bit 7 (D7 error) 12 1 1 Data bit 8 (D8 error) Py = {9,10,11,12} => {5,6,7,8} Py = (11,01) =1 Ps 5,6.7,12} => {2.3.4.8} Ps (0,1,0,1} =0 Py (3,6,7,10,11} => {1,3,4,6,7} Py 1,1,0,1,0} = 1 P, (3,5,7,9,11} => {1,2,4,5,7} Pt {1,0,0,1,0} = 0 Data Bits Are : 7 6 5 4 3 2 1 1 0 1 1 0 1 0 1 ue 6 = 4 3 2 7 0 o 1 0 1 0 7 Pa = {1,0,0,1} Pa Ps P2 Pr 1010 Ps’ = (0,1,0,1} Pie {1,1,0,0,0} @ru Ps'P;'P,! => 0000 Pi/ = {1,0,0,1,0} =0 Syndrome => 1010 => 100 So this bit place no, has become corrupted. Class Work : Original message: 11011011 1001011 Received message : Use SEC to detect the erroneous bit place number.‘TOTSOL - Total Solution totsol info. DCN +21 In ease of DED operation, double bit errors are detected but could not be corrected, Here in this DED scheme an extra generate parity bit will be attached with the old scheme. And the inference will be drawn as shown below. General parity bit detects Syndrome is. Inference / Comment 1. Noerror Non—zero DED 2. Error Non — Zero SEC 3. Noerror Zero No bit corruption has taken place Original Massage : 8765 4321 1101 1011 8765 4321| GP | P’y={0,1,1,1}=1 1110 1011] O 0*3={1,0,1,1 Py Ps P2 Py 2 Py PsP,’ Py! § Syndrome syndrome is pointing to location 3 but errors have occurred at locations 5" and 6" . The general parity bit (GP) detects no error, because errors have occurred in even number of places. So it proves that double bits error has occurred. So DED commences,‘TOTSOL - Total Solution totsol info. DCN ~ Aa , 4 | comics eee eed 1 1 0 1 0 0 1 0 1 o 0 1 0 0 1 1 0 1 1 1 0 0 0 1 0 OO Oe 1 0 1 0 1 0 1 0 1 1 0 0 1 1 0 0 oo 1 0 ©O 1 0 0 1 0 0 1 1 1 i) 0 © Checksum Byte Transmitting End Po 00000000 10100101 Q 00000000 B, 10100101 —~*y so100101 Pi 10100101 10100101 00100110 Q B, 00100110 —“'» Jo01011 Po 11001011 01110000 11100010 Q Bs 11100010 —* jo101101 Ps 10101101 OOOIIIOL 01010101 B, 01010101 —2: goo00010 P; — 00000010 Ooo 00000000 ° X 00000000 —~y 90000010 Ps 00000010 00100001 00000000 y 00000000 —2:+ oo00010 Pe 00000010 =, 00100011 X= Pe- Qe 00000010 00100011 x 11011111 Y= Qo - 2P6 00100011 00000100 00011111 ® Receiving End Y & Summation up to Ps and Q, will be same as above. Rest of the steps are shown below P, 00000010 ~~~ ~~ ooo =X Loni oon P, 11100001 11100001 (00000000 OOL1IIT Y QOOL P. 00000000» 00000000 * 00000000 Thus P, and Q, are zero indicating no errors,DOTSOL - Total Solution totsol info’ DCN > L Input es Bits L Ny ou Half-rate convolutional encoder. Input —» Upward Transi 1” Input —» Downward Transition Trellis diagram of conventional encoder‘TOTSOL - Total Solution totsol info. DCN - 24 © Bit Error Rate (BER): In analog transmission, signal quality is specified in terms of signal-to-noise ratio (SIN), which is usually expressed in decibels. In digital transmission, the quality of received digital signal is expressed in terms of Bit Error Rate (BER) which is the number of errors in a fixed number of transmitted bits. A typical error rate on a high quality leased telephone line is as low as | error in 106 bits or simply 1 x 10°. = PROBI If the average BER is 1 in I 1. single bit error, 2. single bit correct, 3. at least one error in an eight-bit byte? © Solution: Probability of having single bit error 1/10° = 0.00001 Probability of having single bit correct | - 0.00001 = 0.99999 Probability of having one or more errors in an 8-bit byte 1 - (0,999) § = 0.00008 Just like BER, Character Error Rate (CER) and Frame Error Rate (FER) can be defined. CER is the average number of characters received with at least one error in a large sample of transmitted characters. The probability of having at least one error in a byte calculated in the above example gives a CER of 8 in 10° characters, FER, likewise, refers to the average number of frames received with at Teast one error in a lar sample of transmitted frames. It can also be calculated on the same lines as CER. For low values of BER, CER and FER can be calculated from BER as below: CER=bx BER FER=f x BER where bis the number of bits per character and. fis the number of bits per frame, Whatever be the methods of error control, errors cannot be completely eliminated, There is always some residual error, which goes undetected, Residual Error Rate (RER) refers 10 the error rate in the data bits after error control has been performed & Cyclic Redundancy Check: Cyclic Redundancy Check (CRC) codes are very powerful and are now almost universally employed. These codes provide a better measure of protection at the lower level of redundancy and can be fairly easily implemented using shift registers or software. ‘A CRC code word of length N with m-bit data word is referred to as (N, m) cyclic code and contains (N- m) check bits. These check bits are generated by modulo-2 division. The dividend is the data word followed by the n= N - m zeros and the divisor is a special binary word of length n + 1. The CRC code word is formed by ‘modulo-2 addition of the remainder so obtained and the dividend,DOTSOL - Total Solution totsol infor DCN ~ 25 © PROBLE! ‘Generate CRC code for the data word 110101010 using the divisor 10101 © Solutio ¥ Data Word: 110101010 ¥ Divisor: 10101 jooly — L4.000114 Quotient 1101010100000 Dividend 10101 ni Lo10L T0100 Jo1oL 11000 Lo1or 11010 10101 TI110 1101 Tou Rerminder 1101010100000 1011 TroToToT01011 Cade Word In the above example, note that the CRC code word consists of the data word followed by the remainder. The code word so generated is completely divisi le by the divisor because it is the difference of the and the remainder (Modulo-2 addition and subtraction are equivalent). Thus, when the code word is again divided by the same divisor at the receiving end, a non-zero remainder after so dividing’ will indicate errors in transmission of the code word, PROBLEM: The code word of Example 6 be received as 1100100101011. Check if there are errors in the code word. = Solution: jing the code word by 10101, we get 111110001 Lolo 1100100101011 10101 T1000 Lo1ol 11010 Lo1o1 Mit 1o101 10100 LoloL 11011 1o1o1 1110 Remainder ‘Non-zero remainder indicates that there are errors in the received. code word.TOTSOL- Total Solution totsol inf DCN +26 © Undetected Errors in CRC: Not all the types of errors can be detected by CRC code. The probability of error detection and the types of errors, which can be detected depends on the choice of the divisor. If the number of check bits in CRC code is 1, the probabilities of error detection for various types of errors are as given below: Single errors 100% % Two bit errors 100% Odd number of bits in error 100% + Error bursts of length <71+1 100% + Error bursts of length =” + 1 1-1/2" “Error bursts of length > 7+ 1 1-1/2)" For the purpose of analysis, the binary codes are represented using algebraic polynomials. In a polynomial of variable x, coefficients of the powers of x are the bits of the code, the most significant bit being the Coefficient of the highest power of x. The data word 110101010 can be represented by a polynomial M(x) as: M@)=x8 +x" 4x5 +0 +x © Some of the common generating polynomials and their applications are: vc VAL xterra T Ttis used in HDLC/SDLC/ADCCP protocols. v¥ CRC-12: xPaxlty tt eters] It is employed in BISYNC protocol with 6-bit characters. ¥ CRC-16: xa path] Itis used in BISYNC protocol with 8-bit characters, ¥ CRC-32; MB Mey Sey basta ta bet] It is used with 8-bit characters when very high probability of error detection is required.DOTSOL - Total Solution totsol inf DCN ~ Firewalls For corporate security managers, it is a nightmare. Most companies have large amounts of confidential information on-line-trade secrets; product development plans, marketing strategies, financial analyses, etc, Disclosure of this information to a competitor could have an adverse effect on the company. It is danger of information leaking out and leaking in, In particular, viruses, worms, and other digital pests can breach security, destroy valuable data, and waste large amounts of administrators’ time trying to clean Packet Packet the mess they leave. fitemng Applicaton tern ee they router gataway router Consequently, mechanis- ims are needed to keep "good" bits in and "bad" bits out, One method is to uuse encryption. This approach protects data in transit between secure sites. However, encryption does nothing to keep digital pests and hackers Corporate Secutty Inside Outside rope ‘out. To accomplish this Tae teat BAL / 8 A frevall consisting of vo packet files and an piston stenay ae Its old implementation was to dig a deep moat around a castle, This design forced everyone entering or leaving the castle to pass over a single drawbridge, where they could be inspected by the 1/0 police. With networks, the same trick is possible: a company can have many LANs connected in arbitrary ways, but all traffic to or from the company is forced through an electronic drawbridge (firewall) as shown in Fig. ‘The firewall in this configuration has two components: two routers that do packet filtering and an application ‘gateway. Simpler configurations also exist, but the advantage of this design is that every packet must transit two filters and an application gateway to go in or out. No other route exists, Each packet filter is a standard router equipped with some extra functionality, The extra functionality allows every incoming or outgoing packet to be inspected, Packets meeting some criterion are forwarded normally, Those that fail the test are dropped. Packet filters are typically driven by tables configured by the system administrator. These tables list sources and destinations that are acceptable, sources and destinations that are blocked, and default rules about ‘what to do with packets coming from or going to other machines. Blocking outgoing packets is trickier because although most sites stick to the standard port naming conventions, they are not forced to do so. Furthermore, for some important services, such as FTP (File Transfer Protocol), port numbers are assigned dynamically. In addition, although blocking TCP connections is difficult, blocking UDP packets is even harder because so little is known a priori about what they will do. Many packet filters simply ban UDP traffic altogether. The second half of the firewall mechanism is the application gateway. Rather than just looking at raw packets, the gateway operates at the application level, A mail gateway, for example, can be set up to examine each message going in or coming out. For each one it makes a decision to transmit or discard it based on header fields, message size, or even the content (e.g, at a military installation, the presence of words like “nuclear” or “bomb might cause some special action to be taken).TOTSOL - Total Solution sotsol info DCN : 28 —_—— l _ & A B e D gy |e a |e 2 + — -_-_— g 5 i 3 s e a |Ss & Se) g 2 A B e D 3 8 $ 3 + —_—_ > a]? 8 Fi 2 3 a « © . Z Ss | a {8 a & A B @ D 5 —_—_— > > ele — 2 sia @ 8 a Me ® 1 .fe] £8 1 8 A B c D 2 jy ys oi, 2 @ CSMA / CD operation. Sis ]8 ig 7 4 2 3 " 5 = < 5 a TEE 802 5 10-Mbps Physical Layer Medium alternatives TO Based [10 Base? To Fase T [10 Based [10 Base FF Transmission | Coaxiabcable | Coania-cable | —Unahieided | Coaxia-eatle | — 50am Medium 0 cha) GOohn) | Twistedpar | Sohn) | opticarfiber pair Signaling Bussband | Baseband | Basoband | Broadband | Manchestez Technique Manchestes | Manchest) | (Manchestey | @PSK) oof? Topology Bas Bas Star BusrTree Star Heiman 300 Tes Tao Tao 300 Segment Length) Nodes per 00 a - : 5 Segment Cable ri 3 Ta08 Ta10 | asia am Diameter) IEEE 802.3 frame format.‘TOTSOL - Total Solution totsol info. DCN 29 Description of CSMA/CD: CSMA, although more efficient than ALOHA or slotted ALOHA, still has one glaring inefficiency. When two frames collide, the medium remains unusable for the duration of transmission of both damaged frames. For long, frames, compared to propagation time, the amount of wasted capacity can be considerable. This waste can be reduced if a station continues to listen to the medium while transmitting. This leads to the following rules for CSMACD: 1. If the medium is idle, transmit; otherwise, go to step 2. 2. Ifthe medium is busy, continue to listen until the channel is idle, then transmit immediately. 3. Ifa collision is detected during transmission, transmit a brief jamming signal to assure that all stations ‘know that there has been a collision and then cease transmission, 4, Afier transmitting the jamming signal, wait a random amount of time, then attempt to transmit again, (Repeat from step 1.) &FNon-Persistent CSMA: In this scheme, when a station having a frame to send finds that the channel is busy, it backs off for a fixed interval of time. It then checks the channel again and if the channel is free, it transmits. The back-off delay is determined by the transmission time of a frame, propagation time and other system parameters, There is likelihood of some wasted idle time when the channel is not in use by any station. & | Persistent CSMA: In this scheme, stations wishing to transmit monitor the channel continuously until the channel is idle and then transmit immediately. The problem with this strategy is that if two stations are waiting to transmit, then they will always collide and require retransmission. 'p-Persistent COMA: To reduce the probability of collision in I-persistent CSMA, not all the waiting stations are allowed to transmit immediately after the channel is idle. A. waiting station transmits with probability "p” if the channel is idle. For example, if p = 1/6 and if 6 stations are waiting, on average only ‘one station will transmit and the rest will wait, It is equivalent to throwing a dice, and if a station gets six, it transmits, If two stations get six, then both will transmit and collision will take place. Likelihood of such occurrences can be reduced by ‘Throughput as a function of the offered traffic and persistence. reducing the transmission probability. Optimized p-persistent CSMA can give throughput of 0.8-0.9 while I-persistent CSMA achieves throughput of 0.53. The graph shows the throughput (S) versus the offered traffic (G) for various contention schemes.TOTSOL- Total Solution totsol inf DCN +30 © Example of FDDI Token Ring operation: TY “Samana Rl yNemAg op SAe Bits 64 8 8 16o0r48 160r48 20 32 4 1 Preamble sD | FC DA SA info | Fes | ED |Fs (a) General Frame Format Preamble) sD | Fo | FS (b] Token Frame Format Legend: tart-Frame Delimeter SA = Source Address ED = Ending Delimeter FC = Frame Control FCS = Frame-Check sequence FS = Frame Status DA = Destination AddressDOTSOL - Total Solution totsol info DCN ~ Al 802.10 Security and Privacy 802.1 Overview, Architecture, Management 802.2 Logical Link Control (LLC} 802.1 Bridging MAC CSMA cD 802.3 MAC Token Bus 802.4 MAC Token Ring 802.5 MAC MAN 802 MAC IVD 802. MAC Wire -less 802. ql MAC Future 802.7 Broadband Tag 802.8 Fiber optic Tag a oo Fig 6-1: The IEEE 802 standards: CSMA / CD => Carrier Sense, Multiple Access MAN = Metropolitan Area Network IVD & Integrated Voice Data Collision Detect Data-Link Layer Physical LayerTOTSOL- Total Solution totsol inf DCN =32. IEEE LAN Standards: The Institute of Electrical and Electronics Engineers (IEEE) published several widely accepted LAN recommended etandarde, These standards are very important because they encourage use of common approaches for LAN protocols and interfaces. The IEEE LAN Committees are organized as follows: '1,IBBE 802.1 High Levelinterface (MAC bridges} 2,1BBE 802.2 Logical Link Control (LLC) 3. IEEE 802.3 Carrier sense multiple access/collision detect (CSMA/CD} 4.1BEE 802.4 Token bus S.IBBE 8025 Token ring 6.1EEE 8026 — Metropolitan Area Network 7.1BBE 802.7 Broadband LANs B.IEEE 8028 Fiber optic LANe Q.IEEE 802.9 Integrated data and voice networks 10.1EEE 802.10 Security 11.IEEE 802.11 Wireless network Connection options with LANe: At the onset of the IBEE 802 work, it was recognized that the connection-oriented aspect of the OSI approach would limit the scope and power of a local area network. First, many local applications do not need the data integrity provided by a connection-oriented network. Second, high-speed application processes cannot tolerate the overhead in establishing and disestablishing the connections. ‘The connection-oriented model of the ISO's Open system interconnection ig depicted in figure a. ‘The connection mode transfer here requires an agreement between three partiee, ie., the two users A and B, and the service provider (SP] {i., LAN). The two ucers here have to agree upon the SP they will uge for connection before they actually connect each other. Once, thie iz done, the users connect each other through SP, The two users, A and B, then communicate in thie faghion ~ 1, User A sends a request to user B for connection 2, This request is transported across the SP and received by user B ae an indication, 3. Ifuser B wants to connect, then B provides a response. 4. This is transported acrose the SP and received by user A as confirm. 5, Then communication begins. In the other models, only two party agreement ie required, i.e., either between A and B or between uger A and the SP. When the SP knows of the connectione, ie., agreement between user and SP, either remote SP confirms (figure B] or local SP confirms figure c] for user A's connection request to user B. Figure d depicte the model where the SP ahe no prior of the individual agreement between A and B. In these three modele of the IEEE approach, all the information required to deliver the data unit ie presented to the SP, the n layer, along with the data. Consequently, destination address, protocol information, error-check fields are all sent as a single entity through the network, Once this information is exchanged between A and B and the SP, no further communication occurs between the SP and the user layers regarding the disposition of the data unit. User A and B can agree beforehand to establish certain acceptable error ratee and transfer rates. ‘The SP, or n layer, is not involved in any prior agreements between the n#1 A and B entities. Ina gense, the provider ie a passive conveyor of the data between A and B. Because of thie approach, the connectionlees orientation provides for a relatively imple environment between A and B, through the SP. ‘The SP considers the data unit tranemitted to it completely unrelated to any other data unit coming from or going to A and B. Consequently, the data unite provided to the SP will not be delivered in any particular order, Sequencing ie not part of the connectionless data environment. This provides for flexibility, because the SP need not become involved in the details of the connection of DTE A and DTEB.TOTSOL- Total Solution totsol inf DCN +33 A B = Confirm “= Request | 4 ne B Request Response Indication n+1 eee - - Indica- : n Es tion Service- | Service- Provider { Provider for A ; for B - : Confirrg—| Res- ponse (a) Connection-oriented Model - A Service- B Fi a Provider x Request Request Confirm Indication n+1 ao Indica- ¥ = + Service H Service- if Provider | Provider Confirrf forA | bor I (b] Remote-service provider confirms a || Service- ]/B * 5 Request Provider Request Confirm J F indication n+1 s] Indica- | tion Service Service- ? Provider Provider | for A for B (c] Local-service provider confirms A Service- B A B Request Provider Request F Indication n+1 = Indica- a Aa tion n Service- Service- Provider Provider for A for B (d) Service-provider provides no confirm ISO and 802 Connection and Connectionless ModesTOTSOL- Total Solution totsol inf DCN 34, Station 1 Station 2 Station 3 eS Ue CU LAN-A Bridge Bridge Bridge 101 107 102 LAN-B. LAN-C Bridge Bridge Bridge Bridge 103 104 105 106 LAN-D LAN-E| LAN-F LAN-G Station 4 Station 5 Station 6 Station 7 Fig: Configuration of Bridges and LANs, with alternative routes:TOTSOL- Total Solution totsol inf DCN 35 Central Routing Directory: Source Lan: A B ce D E F G aL = i0i_| 102 | 103 | 107 | 105 | 106 BL_i0i = 102 | 103_| 104 | 105 | 106 Destination C|_102 | 101 = 103_| 107 | 105 | 106 LAN D 101 103 102 = 104 105, 106, =" pl-i107_|_104_[_102_[_105 = 105_| 106 F[102_[i01 | 10s [103 [107 = 106 c[-102_ [101 [106 [103 _[_107_|_i05 = Bridge 101 Table: Bridge 102 table: Bridge 103 table: FromLanA FromLanB FromLanA FromLanC FromLanB From LanD Dest Next |Dest Next Dest Next | Dest Next Dest Next |Dest Next B B/A A B -|A aA A -|A B ec -|/c aA oo /]B & c -|B B D Bi|D = dp -|D a BD pilsc. .B BE - |B - BE -/E A BE -|E B Fo- | FOoA Fle kos F -|F B @¢ -|cG aA @ cl{[a - e@ -|c¢ B Bridge 104 Table: Bridge 105 table: Bridge 106 table: From Lan B FromLan& From LanC From Lan F From Lan C From Lan G Dest Next] Dest Next Dest Next|Dest Next Dest Next |Dest Next A -|A = & = a os A -|a c c -|B B B -/|B ¢ B -/|B ¢ D-|c = D'-|c .¢ Dp '—|e.¢ E E|D B E -|bD c B -|D c¢ Fo- | Fo = F F|E co F -|E oc q@-|¢o - qo -|c oc Go c|F_c Bridge 107 Table: From Lan A From Lan Dest Next |Dest Next B =| 4 A c =| BS D -| Cc oA E E| D - F -| F A G -|¢ A Fixed routingtotsol infor DCN +36 2, ‘TOTSOL - Total Solution Host A: Host B: 4 &£ & 4 Networle Piet Bridge Pit — Pht _4| Pe Pit ¥ ¥v t ao 3023) Piet so23 Pit | [soa4 Pit 5024 Pre ¥ Physical = + Ehysical 802.3. Pict 802.3] Pkt 802.4) Pkt 802.4 Pkt ¥ R of 02.4] Pt —> 2023] Pre Token Bus LAN CSMA/CD LAN Fig:_Operation of a LAN Bridge from 802.3 to 802.4 RDOTSOL - Total Solution totsol info DCN _- 37 Backbone Bridge LAN ie B B B B tT] Cluster ona File single Server LAN Work Station LAN Multiple LANs connected by a backbone to handle a total load higher than the capacity of a single LAN: z 7 6 6 5 5 Open 4 Bridge 7 Open System 3 3 System z 2 z 1 Tew ¥ Open system Representation of a bridge Bridge Bridge LAN B Point-to-point Link (a] Bridge over a point-to-point links X.25 Packet- Se Bridge Switched Bridge at Public Network (b] Bridge over X.25 Wide Area Network: & Remote BridgesTOTSOL- Total Solution tots infin DCN -38 Host an Router aan Host (a) LAN to LAN: Long haul communication Linke — Router LAN = for} Can) Cromer FO Common Linke wan) Host () WAN to WAN. Gateway ta T 7 7 T 7 6 6 os! 6 6 6 6] ost S s ‘System S S 5 S| System 7 Router a $ 4 z a 3 s 3 3 3 3 3 2 Se 2 2 2 2 2 T TT T I 1 1 OSI Model of a Router: OSI Model of a Gateway: city 64 Kbps CityB aa Router Router Cater) Leased 64 Kbps \ 19.2 Kbpe Router Router, Router ao 64 Kbpe city c city D y Fig A large IntemetworleDOTSOL - Total Solution totsol info DCN_-39. A B ¥ Connect Phase + Link Establishment Phase RRM |} [ Data-n }-» Data-Linie Laver B Puaations ¥ Data Transfer Phase t <—{_DataP Q ‘Termination Phase 4 Clear Phase M[ Daten [> N,P: Data Frame Sequence Numbers M, Q: RR Sequence Numberz Link Manag¥ment Phases Fig: Piggybacking of Acknowledgements: Level: Layer: Primary Functions: Services Provided to next higher layer: 7 Application | Support the end-user, This is the highest layer and provide LOGIN, password, user-oriented services, File Transfer. 6 Presentation | Code and format conversion | Freedom from compatibility problems. 5 Session Session management Dialogue Management and Synchronization, 4 ‘Transport Optimum utilization of End-to-end transport connection of the network resources. ‘The required quality in a cost- effective manner, 3 Network Interaction with the Network connections linking the Sub network, Routing End-systems. and Relaying 2 Data Link Error control, Flow control | Reliable transfer of bits across the Physical connection. 1 Physical Conversion of bits into Transmission of bits Electrical signals of suitable characteristics. 0 Media Transmission of Electrical | Transmission of Electrical signals signals, Summary of functions and services provided by various layers of the OSI Reference Model:DOTSOL - Total Solution totsol info. DCN 40 tq | DLE || st A DLE B Die || etx (b) | DLE || stx 4 DLE || DLE B DLE | | ETX ™ Stuffed DLE (J) DLE || sTx A DLE B DLE || ETx {a} Data sent by the network layer. (b] Data after being character stuffed by the data Link layer. {c} Data passed to the network layer on the receiving side. la) 9 (b] 0 PICT PTET PAT 4A24 144 0 DOD oh Oe Oa Adee CHT, 31k 3 Stuffed bits (] O1101111111111111111 OO10 oo10 1010010 1 0 (a} The original data. (b) The data as they appear on the line. (c] The data as they are stored in the receiver's memory after destuffin, To: A T H K allo 24 20 21 8 A os 5 e D «(aie 36 31, 28 20 A ce [2s 18 19. 36_|| 28 I we D (40 27 3 24 || 20 | _H eB [4 7 30 22 17 I E H F* [23 20, 19 40 30 T ¢ rts 3i 6 31 18 H Hi [rire 20 0 19 12 H I 21 0 pS 22 10 I a it 7 10. 0 - 1 a K L « [24 22 23 o 6 K L [23 33 3 3 15 K (2) , JA JI JH JK ‘= Bellmanford’s Algorithm: Delay Delay = Delay Delay is is is is 8 10 12 6 {a] A Subnet Vectors received from J's four neighbors (b) (b] Input from A, I, H, K, and the new Routing Table for J New routing table for JDOTSOL - Total Solution totsol info. DCN - 41 Apee ena #* Dijkstra’s Algorithm */ status [wo] <1; disk [vo] — 0; next [vo] —*; for (all [v] #¥0) ( status [v] —? ; dist [v] —L(v, wo); next [v] — vo; ch while (There are remaining ? Vertices) ( vs? Vertex for which dist is smallest ; status [v«] 1; for (cach remaining ? vertex v) Cif Gist [v] > dist [ve] + L (ve, v)) ( dist [v] — dist [ve] +L (ve) ; next [v] —v+; initialize ve vertex with smallest dist + Status [ ve] <1 ¥ update list next tablesDOTSOL - Total Solution totsol infor DCN 42 Dijkstra’s Algorithm Vertex x B i D E F S ‘Status T 7 7 7 7 7 7 dist 0 3 6 a e & @ next = A A A A A A Status T T 7 7 z 7 z dist 0 3 5 7 o & % next = A B B rN A A ‘Status T T T t 7 7 z dist 0 3 53 6 9 a a next * A B c fi c A ‘Status I i i i 7 F 7 dist 0 3 3 6 7 10 next * A B c D c D ‘Status T T T T 7 T 7 dist 0 3 5 6 3 a 8 next = A B c D c F ‘Status I 1 1 1 1 i 7 dist 0 3 5 6 8 i 8 next = A B o. D ral F Status T T T T i T T dist 0 3 5 6 g a 8 next = A B c D Cc ETOTSOL- Total Solution totsol inf DCN -43 Nyquist’s and Shannon’s theorem The Maximum Data Rate of a Channel: As early as 1924, H. Nyquist realized the existence of this fimdamental limit and derived an equation expressing the maximum data rate for a finite bandwidth noiseless channel, In 1948, Claude Shannon carried Nyquis’s work further and extended it to the case of channel subject to random (that is; thermodynamic) noise (Shannon, 1948). Nyquist proved that ifan arbitrary signal has been run through a low-pass filter of bandwidth H, the filtered signal can be completely reconstructed by making only 2H (exact) samples per second, Sampling the line faster than 2H times per second is pointless because the higher frequency components that such sampling could recover have already been filtered out, Ifthe signal consists of V discrete levels, Nyquist’s theorem states: 2H logs V bits/sec For example, a noiseless 3-KH1z channel cannot transmit binary (i.e, two-level) signals at a rate exceeding (6000 abps. So far we have considered only noiseless channels. If random noise is present, the situation deteriorates rapidly. The amount of thermal noise present is measured by the ratio of the signal power to the noise power, called the signal-to-noise ratio. If we denote the signal power by S and the noise power by NV, the ignal-to-noise ratio is S/N. Usually, the ratio itself is not quoted; instead, the quantity 10 logo S/N is given. ‘These units are called decibels (dB). An S/N ratio of 10 is 10 dB, a ratio of 100 is 20 dB, a ratio of 1000 is 30 B and so on. The manufacturers of stereo amplifiers often characterize the bandwidth (frequency range) over which their product is linear by giving the 3-dB frequency on each end, These are the points at which the amplification factor has been approximately halved, Shannon's major result is that the maximum data rate of a noisy channel whose bandwidth is MHz, and ‘whose signal-to-noise ratio is S/N, is given by ‘maximum number of bits! sec = H logs (I + S/N) For example, a channel of 3000-Hz bandwidth, and a signal to thermal noise ratio of 30 dB (typical parameters of the analog part of the telephone system) can never transmit much more than 30,000 bps, no ‘matter how many or few signal levels are used and no matter how often or how infiequent samples are taken. Shannon's result was derived using information-theory arguments and applies to any channel subject to Gaussian (thermal) noise. Counter examples should be treated in the same category as perpetual motion machines. It should be noted, however, that this is only an upper bound and real systems rarely achieve it. maximum data rate Bps_[T (msec)]First harmonic (Hz) |# Harmonics sent 300 [_ 2667 375) 80 ooo [1333 5 0 1200 [667 150 20 2400 [3.33 300 10 4900 | 1.67 600 5 600 [083 1200 2 19200 [0.42 2400 1 36400 [021 4800 0DOTSOL - Total Solution totsol inf DCN - 44 Layer: Name of unit exchanged: Application Protocol 7 | application |Q_ Application | appu Interface ; Presentation Protocol 6 | Presentation Presentation | PPDU Interface ae Session Protocol 5 | Session SPDU Tranzport Protocol t 4 | Transport ‘Transport Communication Subnet Boundary TPDU 8 Packet 2 | Data-Link Data-Link Data-Link Data-Link | frame 1 | Physical Lp] Physical [jw | Physical Physical | pit QS ssewecceerece tt cree Host-A Host-B Networle Layer Host-Router Protocol Data-Link Layer Host-Router Protocol Physical Layer Host-Router Protocol Fig: 1.16: The OSI Reference Model: 1. A layer should be created where a different level of abstraction is needed. 2. Each layer should perform a well-defined function. 3. The function of each layer should be chosen with an eye toward defining intemationally standardized protocols 4. The layer boundaries should be chosen to minimize the information flow across the interfaces. 5. The number of layers should be large enough that distinct functions need not be thrown together in the same layer out of necessity, and small enough that the architecture does not become unwieldy.DOTSOL - Total Solution totsol info. F F F F ACK F F F ACK F F WACK ACK F Asends a Frame B sends ACK, Asends next frame. B withholds ACK. Awaits for ACK. B releases ACK. Asends next frame. B sends WACK, Awaits for ACK. B releases ACK. Asends next frame. © Stop-and-Wait Flow Control Mechanism: (1 Att=o A starts sending Frame. (2) Att = t A completes trans- mission of the frame. (3) Att =te+ te B receives the frame. (4) Att=te+ ty B sends ACK. (5) Att = te+ 2tp A receives the ACK. Frame © Time Analysis of Stop-and-Wait Flow Control Mechanism:DOTSOL - Total Solution totsol info. DCN -46 Sender Receiver Data-block with Check bits 1 1 No Errors 2 Errors detected 3 3 2 2 | Request for retransmission of data-block 2 2 2 3 3 © Reverse error correction by Go-Back-N Mechanism: Sender Receiver Data-block with Check bits t 1 No Errors 3 Errors detected 3 3 2 2 Request for retransmission of data-block 2 = Reverse error correction by Selective Transmission Mechanism:DOTSOL - Total Solution totsol inf DCN _- 4: Window es | r4 [ra | ro [Fi 3 a 3 a Fa RRS Fil Fo RNRO_}-- FIT Fl6 A waits for RR Window BRO FI7 Fle Sliding Window Flow Control Mechanism Le > o A transmits all the frames in the window and awaits acknowledgement, ‘When window size RR-3 equals count sequence, After time out A challenges B. Ambiguous acknowledgementDOTSOL - Total Solution totsol inf DCN - 48 & PROBLEM ON FLOW CONTROL: Calculate the maximum link utilization efficiency for stop-and-wait flow control mechanism if the frame size is 2400 bits, bit rate is 4800 bps and distance between the devices is 2000 km. Speed of propagation over the transmission media can be taken as 2,00,000 k's © Solution: We consider t= Frame copy time = N/R and t, = Frame propagation time and A= fp/ ty Frame transmission time 2400/4800 = 0.5 s. Propagation time 20001200000 = 0.01 s. =0.01/0.5 = 0.02 U= (1+ 0,04) = 96% The following table gives link utilization efficiencies for some typical cases. Speed of propagation in cable ‘media is assumed to be 200,000 km/s, Note that link utilization is very poor in case of a satellite link. The link utilization can be improved by keeping a large fame size and thus reducing . But it may not be always possible. It is more likely that transmission errors will occur in a data frame of large size and the effective throughput will be reduced, Therefore, the stop-and-wait mechanism is suitable only when the propagation time is less than or comparable to the frame transmission time. The sliding window flow control mechanism ‘overcomes this limitation and is universally adopted. ‘Table: Link Utilization in Stop-and-Wait Mechanism e Tocal cable lnk = 100% (10 km) Coaxial cable link oT 0.005 0.05 ~91% (1000 ken) Satelite link oT 270 270 ~ 15.0% * N= 960 bits, R = 9600 bits/s SLIDING WINDOW FLOW CONTROL In sliding’ window flow control mechanism, multiple data frames can be transmitted without waiting for acknowledgements of individual data frames. To understand its operation, let us examine its basic features: ‘© Each data frame carries a sequence number for its identification. ‘© The sending end maintains a window containing a fixed number of data frames ready for transmission. ‘These frames can be sent without waiting for any acknowledgement. But a copy of each transmitted frame is retained in the window tll it is acknowledged ‘©The number of frames in a window is called its size. It typical value is seven. © The receiver acknowledges receipt of one or more data frames by sending back a numbered acknowledgement (Receive Ready, RR-N) frame, N is the sequence number ofthe next frame it expects to receive. * All previous data frames are assumed acknowledged on receipt of an acknowledgement, For example, by sending RR-S, the receiver is acknowledging receipt of frames bearing numbers 4, 3, 2, ete. Note that all data frames are still acknowledged - but not individually. ‘* When an acknowledgement is received by the sending end, it slides the window deleting the copies of acknowledged data frames and inserting the same number of new frames from the queue of data frames awaiting transmission,DOTSOL - Total Solution totsol inf DCN - 49 ‘© To stop the transmission temporarily, the receiving end can send another type of acknowledgement, Receive Not Ready (RNR-N). RNR-N is acknowledgement up to frame N-1 and a request to stop further transmission temporarily. Transmission can be resumed when RR-N is released by the receiving end. ‘© A timer is provided so that the sending end can request for acknowledgement after time out. ‘© RRand RNR insliding window flow control are equivalent to ACK and WACK in the stop-and-wait flow control. Sequence Numbering of the Frames: In the sliding window mechanism, all data frames are given a sequence number, which is a binary number of a fixed number of bits. Any numbering scheme of fixed size has a finite count sequence afier which it must start all over again from the beginning. If there are "tn" bits, the length of the count sequence would be 2". The two-bit numbering scheme counts from 0 to 3 and then starts again from 0, Similarly, the three-bit numbering scheme counts from 0 to 7 and then starts again from 0. Since no sequence number may be repeated within the window, the length of the count sequence must be at least equal to the size of the window. But window size equal to the count sequence results in some ambiguous situations as illustrated below. Let us suppose window size is eight and the three-bit numbering scheme is used. The status of window on receipt of RR-3 is shown in the following figure: A tesa all the frames in the window ‘od awaits ve ackoowiodgement Aller time out A challenges B ‘Ambiguous acknowledgement when ‘window size equals count sequence. On receipt of RR-3, A sends all the eight frames in the window and waits for acknowledgement, After time out it challenges B, which replies with RR-3. This RR-3 can be a repetition of the last RR-3 if all the eight frames just sent did not reach B at all or it can be the acknowledgement of these 8 fiames. This ambiguity can be resolved by restricting the maximum window size to 2" - | for an n-bit sequence number, & Link Utilization, Unlike the stop-and-wait mechanism, in the sliding window flow control, each data frame is not individually acknowledged, and therefore, the sending end can send a number of frames one after the other without waiting for acknowledgement, which results in better link utilization, To calculate link utilization, let us consider the following two possible situations, 1. Areceives an acknowledgement before it exhausts the window. 2. Aexhausts the window before it receives an acknowledgement, In the first situation, A can keep sending data frames’ without interruption and the link is never idle. The link utilization is unity. If the window size is W, this situation will occur when the time required to transmit 17 frames is more than the earliest possible arrival of an acknowledgement, i.¢., 4p +2. It is assumed that the size of acknowledgement is very small.DOTSOL - Total Solution totsol inf DCN -50 Wy zy +2'p Or W>1424, A= p/p In the second situation, A sends W frames in time 17 iyand then suspends further transmission of the frames until an acknowledgement is received. If we assume that B sends the acknowledgement at the earliest ‘opportunity, i.e., immediately following the receipt of the first frame, A shall receive the acknowledgement afier time fy + 2. Therefore, this situation will occur when Wir sty +2 ty Or Welt2A In this case the link has been engaged for time fy + 2 %, while A has utilized it for time 1’ t Therefore, link utilization is given by U=Wh/ly +24)=Wi(1+2A) 10 ¥00 1000 zation efficiency. ‘PROBLEM: Calculate link utilization efficiency if 1. Bit rate = 19.2 Kbps, frame size = 960 bits, window size = 3, propagation time = 0.06 s. 2. Bitrate = 19.2 kbps, fiame size = 960 bits, window size =7, propagation time = 0.06 s, ‘What is the minimum window size for 100 per cent link utilization? © Solution: 1. = 960/19200 = 0.05 s, A = 0.06/0.05 = 1.2 2A+1=34, W<24+1,u=3/3.4=88% 2,W=7>24+1, U=100% Minimum size of the window for U= 100% is given by W=24+1=3.4 ‘So window size should be at least equal to 4 for 100% efficiency,DOTSOL - Total Solution totsol inf DCN 51 The Leaky Bucket Algorithm Imagine a bucket" with a small hole in the bottom, as illustrated in Fig. (a). No matter at what rate water enters the bucket, the outflow is at a constant rate, p, when there is any water in the bucket, and zero when the bucket is empty. When the bucket is full, any extra water entering it spills over the sides and is lost. The same idea can be applied to packets, as shown in Fig. (b). Conceptually, each host is connected to the network by an interface containing a leaky bucket, that is a finite internal queue. If a packet "arrives at the queue when it is full, the packet is discarded. In other words, if one or more processes within the host try to send a packet when the maximum numbers are already queued, the new packet is discarded. This arrangement can be built into the hardware interface or simulated by the host operating system. It was first proposed by ‘Tumer (1986) and is called the leaky bucket algorithm. In fact, itis nothing other than a single-server queuing system with constant service time. The host is allowed to put one packet per clock tick onto the network. Again, this ean be enforced by the interface card or by the operating, system. This mechanism tums an uneven flow of packets from the user processes inside the host i erace i, xiang —| \ aleny bucket ¢ Wit iesonctte 4) hale at acest le ® |a (3) Aeaky bucket wih wate, 0) A leaky bucket with packets Neto into an even flow of packets onto the network, smoothing out bursts and greatly reducing the chances of congestion. When the packets are all the same size (eg. ATM cells), this algorithm can be used as described, However, when variable-sized packets are being used, it is often better to allow a fixed number of bytes per tick, rather than just one packet. Thus if the rule is 1024 bytes per tick, a single 1024-byte packet can be admitted on a tick, two 512-byte packets, four 256-byte packets, and so on. Ifthe resus byte counts too lw, the net packet mist wait uni the next tick, The byte-counting leaky bucket is implemented almost the same - way. At each tick, a counter is initialized ° —--) ‘0 to 7. If the first packet on the queue has fewer bytes than the current value of the counter, it is transmitted, and the counter is decremented by that number of bytes ¢ Tie (ro) 0 Additional packets may also be sent, as long as the counter is high enough. When the counter drops below the length of the next packet on the queue, transmission stops until the next tick, at which time the residual byte count is overwritten and lost. AS an example of a leaky bucket, imagine that a computer can produce data at 25 MB/sec (25x8 = 200 Mbps) and that the network also runs at this speed. However, the routers can handle this data rate only for short intervals. For long intervals, they work best at rates not exceeding 2 MBYsec. Now suppose data comes in I-million-byte bursts, one 40-msec burst every second, To reduce the average rate to 2 MBisee, we could use a leaky bucket with p = 2 MB\sec and a capacity, C, of 1 MB. This means that bursts of up to 1 MB can be handled without data oss, and that such bursts are spread out over 500 msec, no matter how fast they come in. In Fig. we see the input to the leaky bucket running at 25 MB’sec for 40 msec. In Fig. we see the output draining out at a uniform rate of 2 MB/sec for 500 msec. Fe ec tre » 2 Wel e500 eee =DOTSOL - Total Solution totsol inf DCN _- ‘The Token Bucket Algorithm The leaky bucket algorithm enforces a rigid output pattem at the average rate, no matter how bursty the traffic is. For many applications, it is better to allow the output to speed up somewhat when large bursts arrive, so a more flexible algorithm is needed, preferably one that never loses data. One such algorithm is the token bucket algorithm, In this algorithm, the leaky bucket holds tokens, generated by a clock at the rate of one token every AT'sec. In Fig, (a) we see a bucket holding three tokens, with five packets waiting to be transmitted. For a packet to be transmitted, it must capture and destroy one token. In Fig. (b) we see three of the five packets have gotten through, but the other two are stuck waiting for two tokens to be generated. The leaky bucket algorithm ‘does not allow idle hosts to save up permission to send large bursts later, The token bucket algorithm does allow saving, up to the maximum size of the bucket, n. This property means that bursts of up to 1” packets can be sent at once, allowing some burstiness in the output stream and giving faster response to sudden bursts of input. Another difference between the two algorithms is that the token bucket algorithm throws away tokens when the bucket fills up but never discards packets. In contrast, the leaky bucket algorithm discards packets when the bucket fills up. Here too, a minor variant is possible, in which each token represents the right to send r not one packet, but & bytes. A packet can only be 1 a ie ‘transmitted if enough tokens are available to cover its length ‘computer ‘computer in bytes. Fractional tokens are kept for future use. | The leaky bucket and token bucket algorithms can also be used to smooth traffic between routers, as well as ren being used to regulate host output. However, one clear Sn aha Q difference is that a token bucket regulating a host can make NI the host stop sending when the rules say it must. Telling a ia router to stop sending while its input keeps pouring in may a result in lost data, a 5 @) The implementation of the basic token bucket algorithm is just a variable that counts tokens. The counter is incremented by one every AT and decremented by one Networks Networks whenever a packet is sent. When the counter hits zero, no The token bucket algorithm. (a) Before. (b) Afr. packets may be sent. In the byte-count variant, the counter is, increment by k bytes every A/’and decremented by the length of each packet sent, Let we have a token bucket with capacity of 250KB. Tokens arrive at rate allowing output at 2 MB/sec. Assuming the token bucket is full when the IMB burst arrives, the bucket can drain at the full 25 MB‘sec for about 11 msec. Then it has to cut back to 2 MB/sec until the entire input burst has been sent. Calculating the length of the maximum rate burst is slightly tricky. It is not just I MB divided by 25 MB/sec because while the burst is being output, more tokens arrive, If we call the burst length S sec, the token bucket capacity C bytes, the token arrival rate p bytes/sec, and the maximum output rate M bytes/sec, we see that an output burst contains a maximum of C + pS bytes. We also know that the number of bytes in a maximum-speed burst of length S seconds is MS. Hence we have: C + pS ~ MS So S = C(M- p). And for C= 250 KB, M=25 MBisec, and p = 2 MB/sec, we get a burst time of 11 msec. A potential problem with the token bucket algorithm is that it allows large bursts again, even though the ‘maximum burst interval can be regulated by careful selection of pand M. Frequently itis desirable to reduce the peak rate, but without going back to the low value of the original leaky bucket. One way to get smoother traffic is to put a leaky bucket after the token bucket. The rate of the leaky bucket should be higher than the token bucket's p but lower than the maximum rate of the network.DOTSOL - Total Solution totsol info DCN 53 Oxoowox | LLLLLLLL | LLLLLLLL | LLLLLLLL | Gyn 9 addresses: Values 0-126 O11 Reversed Loopback address value: 127 1o%0coo% |AAAAAAAA | LLLLLLLL | LLLLLLLL | Class B addresses: Values 128-191 AAAAAAAA 11oxxxxx |AaAAAAAA |AAAAAAAA|LLLLLLLL | Class C addresses: Values 192-223 AAAAAAAA — 1110xxxx Reversed Multicast addres: Values 224 - 239 Saeae Reversed Experimental addresses: ae Values: 240 - 255 + A= Assigned by NIC © L= Locally administered IP 199,84.57.20 EN 25 IP 156.40.10.50 = EN 115 IP.156.40.10.8 EN 1000 IP 199.84.57.10 EN 14 IP 156.40.10.60 ENS 9) \@ a" ce IP 199.34.57.40 ELL IP 156.40.10.14 EN 231 © Segment “Rome” © Segment "Carthage"‘TOTSOL - Total Solution totsol info. DCN +54. ‘TCP/IP and its addressing format © An intranet is made of at least two sub-nets. SF Ethemet Addre Each Ethemet board’s Ethemet address is a unique 48-bit identification code. If it sounds unlikely that every Ethemet board in the world has a unique address, then consider that 48-bits offers 280,000,000,000,000 possibilities. Ethemet itself only uses about one quarter of those possibilities. (two bits are set aside for administrative function), but that’s still a lot of possible adresses. In any case the important thing to get here is that a board’ Ethemet address predetermined and hard-coded into the board. Ethemet addresses, which are also called Media Access Control (MAC) addresses are expressed in twelve hex digits. (MAC address is synonymous with Token Ring. address or Ethemet address). A computer with more than one network card is called multi-homed computer. EPP Addresses and Quad Format In contrast to the 48-bit in a MAC address, an IP address is a 32-bit value, IP addresses are number set at a ‘workstation (or server) by a network administrator ~ they are not a hard coded hardware kind of address, unlike the four billion distinct Internet addresses. IP addresses are usually represented in dotted quad notation e.g. w.x.y.z where w, x, y, Z are all decimal values between 0 and 255. An IP address of 11001010 00001111 10101010 00000001 is equivalent to 202.15.170.1 SIP Routers Now let’s re-examine how to message gets from A to G. 1. At first it will be checked whether G is on the same subnet as A or not. A. subnet is simple a broadcast area 2, Station A determines that it ison a different subnet from station G by examining the addresses, 3. So station A cannot directly sends its IP packets to G. The network administrator sets up D as the A’s default router. A then sends an Ethernet frame from itself to D with following information © Source Ethemet address : 14 © Source IP address : 199.34.57.10 © Destination Ethernet address : 100 © Destination IP address ; 156,40, 10.50 4, Ethemet card D receives the frame and hands it to the IP software running in its PC. Examining the subnet, the PC sees that destination lies on the subnet that Ethemet adapter E is on, so it sends out frame from Ethernet adapter E, with ths information © Source Ethemet address : 100 © Source IP address : 199,34,57.10. © Destination Ethemet address : 115 © Destination IP address : 156.40.10.50, 5. G then gets the packet. It knows that it got this frame from E but the original message really came from another machine, the 199.34.57.10 machine. EFA, B, AND C-CLASS NETWORK : The NIC assigns a company a block of IP addresses according, to that ‘company’s size. Big, medium and others get A, B and C-class networks. EA class of network : A large network would have its first eight bits set by NIC and the networks intemal administrators could set the remaining, 24 bits. So it has 2° or 16 millions hosts. E.g. IBM(9), DEC(16) et. SB class of network : Medium sized networks have the left most 16 bits preassigned to them, leaving 16 bits for local use. Class B addresses is always have the values 128 through 191 in their first quad, then a value between 0 to 255 in their second quad. There are then 16,384 possible class-B network. Each of them can have up to 65,535 hosts. E.g. Microsoft and Exxon, &C class of network_; Small networks have the left most 24 bits preassigned to them, leaving only 8 bits for local administration, Class C addresses starts off with a value between 192 and 223, As the seconds and third quads can be ‘any value between 0 and 255 but that means there can potentially be 2,097,152 class C network. RESERVED ADDRESS: A number of addresses are reserved for multicast purposes and for experimental ‘purposes. So they can’t be assigned for networks. In particular, address 224.0.0.0 is set aside for multicasts, network transmissions to groups of computer. |THE LOOPBACK ADDRESS: The address 127.0.0.1 is reserved as a loop back. If you send a message to 127.0.0.1, then it should be returned to you, unless there’s something wrong on the network. An so no network has an address 127,xxXXXXXX,.XXXXXXAX,XXXXXXAX , an unfortunate waste of 24 million addresses.DOTSOL - Total Solution totsol inf DCN _- 55 & Subnetting a C-Class Network: Number of Remaining Desired Network Router Broadcast | Number of Subnets: Subnet Mask: | Number, | Address: | Address | IP Addresses: 1 255.255,255.0 xy.20 xy.z1 xy.2255, 253 2 255.255.255.128] xy.z0 xy.z1 xy.2127 125 255,255,255, xy.z128 | xy.2129| xy.2255 125 4 255,255.255.192| xy.z0 xy21 xy.263 61 255,255,255, xy.264 xy265 | xyz127 61 255.255.255, xy.z128 | xy.2129| xy.2191 61 255.255.255, xy.z192 | xy.2193| xy.2255 61 8 255.255.255,.224| xy.n0 xy21 xy231 29 255,255,255. xy.232 xy233 | xy.263 29 255,255.25. xy.Z64 xy265 | xy.295 29 255,255,255, xy.296 xy.297 | xyz127 29 255,255,255. xy.2128 | xy.2129| xy.2159 29 255,255,255. xy.z160 | xy.2161| xy.2191 29 255,255,255. xy.z192 | xy.z193| xy.z223 29 255,255.25. xy.2224 | xy.2225| xy.2255 29High-level data link control (HDLC) was developed by ISO. It offers high level of flexibility, adaptability, reliability and efficiency of operation for synchronous data communication needs. & GENERAL FEATURES: HDLC isa bit oriented data link control protocol supporting Point-to point and point-to-multipoint communications Two-way simultaneous communication over full duplex circuits ‘Two-way alternate communication over half duplex and full duplex circuits Synchronous and asynchronous communication ‘Communication between equal stations and between host and remote stations Full data transparency The ISO standards for HDLC are ISO 3309, 4335, 6159, 6256. Syeene & TYPES OF STATIONS : Three types of stations are defined — primary, secondary and combined stations. A primary station has responsibility of data link management (activating, maintaining, disconnecting the data link), A secondary station operates under the control of a primary station, The primary stations send “command” frames to secondary stations under it, The secondary stations then can send frames called “response”, A combined can aet both primary and secondary station, Communication can be between two logical equal status computers also, in which case they are designated as combined stations and can receive both command and response frames, & MODES OF OPERATION: The stations are generally disconnected. The primary station send mode setting command frame to activate them and to bring them under a certain data transfer mode. ‘There are three modes of operation to activate the stations- 1, Normal disconnected mode (NDM) - stations activate into normal response mode 2. Asynchronous disconnected mode (ADM) - stations activate into asynchronous response or asynchronous: balanced mode 3. Initialization mode (IM) - operational parameters are exchanged in this mode. It is invoked when either primary station finds abnormal functioning of a secondary station or a station itself can request one if unable to function properly. © There are three data transfer modes Modes Nontnal ‘Asynchronous ‘Asynchronous Properties | Response (NRM) Response (ARM) Balanced (ABM) Mode Synchronous ‘Asynchronous ‘Asynchronous Type of | Primary — secondary Primary — secondary Combined station (Logical unequal stations) (Logical unequal stations) (Logical equal or combined) Functioning | 1. Assecondary staton can send | 1. The secondary staoncan | 1. A station can senda a frame only as a result of send a frame without any | frame without any explicit receiving explicit permission explicit permission from | permission from the other from primary station pimary station station. 2. Primary station controls the 2. Both stations are overall link management 2. Primary station controls capable of data link functions the data link management | management functions. finetions ‘Applicable | Pomt-to-pont Point-t0-pomt Poittto-point to Point-to-multipoint Point-to-multipoint Polled multipointTOTSOL- Total Solution totsol inf DCN +5 & FLOW CONTROL: HDLC utilizes sliding window flow control mechanism. Window size can be either7 or 127. Error Control - error control based on retransmission of frames received with errors. Error detection is carried out using 16-bit CRC. & FRAMING: Two types of frame formats are used in HDLC. The frame formats are fixed. Alll the fields except the information field have fixed length. Two frame delimiters are required. The frame is transmitted from left to right and the lower bit is transmitted first Low-order bit Flag | Address | Control | Information | FCS | Flag | Information transfer frame Flag | Address [ Control | FCS | Flag | Supervisory / unnumbered frame FIELDS: 1. Flag : It is a unique 8-bit word pattern (01111110), which identifies the start and the end of each frame 2. Address : It contains the address of the secondary station whether frame is transmitted by primary or secondary. In command, a station gives the address of the destination, and in response it gives its own address. The address field consists of 8-bits, giving it a capability of 256 different addresses. 3. Control field : It carries the sequence number of the frame, acknowledgements, request for retransmission and other control commands and response. The field is of 8-bits 4, Information field: It contains the user data and is completely transparent. The length is variable. 5. Frame check sequence : FCS is a 16-bit CRC code for detection of errors in the address, control and information fields. & TYPES OF HDLC FRAMES. There are three types of HDLC frames: 1. Information transfer frame (I-frame) - The [frame is used for transporting user data, also carries acknowledgement of the received frames. The 1* bit is “O” which determines the frame as I-frame, The next 3 bits are the sequence number of the frame, 2. Supervisory fame (S-frame) - the S-frame is used to carry only acknowledgements and request for retransmission and does not have a data field. The I two bits of control field with value of “10” determines the frame as S-frame. Unnumbered frame (U-frame) - The U-frame is used for link establishment, termination, mode setting and other control functions. These do not have any sequence number and data field. The 1" two bits of control field with value of “11” determines the frame as U-frame.DOTSOL - Total Solution totsol info DCN _- 58 @ HDLC - High Level Data-Link Control: (COMMAND. RESPONSE. PRIMARY STATION Communication between Primary A B Secondary stations t 7 SECONDARY STATIONS ‘COMBINED a a COMBINED STATION ‘STATION RESPONSE }--» <--[ COMMAND (Communication between Combined stations ‘MODE SETTING COMMANDS: SABM: Set t Asynchronous Balanced Mode SIM; Set to Initialization Mode DISC: Set t» Dizconnectad Mode MODE SETTING COMMANDS, SINRM: Set to Normal Response Mode} SARM: Set to Asynchronous Response ModeTOTSOL- Total Solution totsol inf DCN 59 Integrated Services Digital Network (ISDN) ISDN Services: The key ISDN service will continue to be voice, although many enhanced features will be added. For example, ‘many corporate managers have an intercom button on their telephone that rings their secretaries instantly (no call setup time), One ISDN feature is telephones with multiple buttons for instant call setup to arbitrary telephones anywhere in the ‘world, Another feature is telephones that display the calle’ telephone number, name, and address on a display while ring- jing. A more sophisticated version of this feature allows the telephone to be connected to a computer, so that the caller's Customers oe. ISON Terminal 7D database record is displayed on the screen as the call comes in, Other advanced voice services include call forwarding and | _ ison 'sON ISON conference calls worldwide, | Telephone Terminal Alar Advanced —_non-voiee services are remote electricity meter reading, and on-line medical, burglar, and smoke alarms that automatically call the hospital, police, or fire department, respectively, and give their address to speed up response. ISDN System Architecture: It is now time to look at the ISDN architecture in detail, particularly the customer's equipment and the interface between the customer and the telephone company or PTT. The key idea behind ISDN is that of the digital bit pipe, a con- ceptual pipe between the customer and the carrier through which bits flow. Whether the bits originated from a digital telephone, a digital terminal; a digital facsimile machine, or some other device is imelevant, All that matters is that bits can flow through the pipe in both directions. ‘The digital bit pipe can, and normally does, support multiple independent channels by time division multiplexing of the bit stream. The exact format of the bitstream and its multiplexing is a carefully defined part of the interface specifi- cation forthe digital bt pipe, Two principal standards for the bit pipe have been developed, a low bandwidth standard for hhome use and a higher bandwidth standard for business use that supports multiple channels that are identical to the home use channel. Further more, businesses may have multiple bit pipes if they need additional capacity beyond what the standard business pipe can provide, In Fig. (a) we see the normal configuration for a home or small business. The carrier places a network terminating device, NTI, on the customer’s premises and connects it to the ISDN exchange in the carriers office, several kilometers away, using the twisted pair that was previously used to connect to the telephone, The NTI box has a connector on it into ‘which a passive bus cable can be inserted. Up to eight ISDN telephones, terminals, alarms, and other devices can be con- nected to the cable, similar to the way devices are connected to a LAN, From the customer’ point of view, the network boundary is the connector on NT 1 For large businesses, the model of Fig, (a) is inadequate because if is common to have more telephone conversations going on simultaneously than the bus can handle, Therefore, the model of Fig. (b) is used. In this model we find a device, NT2, called a PBX (Private Branch eXchange), connected to NTI and providing the real interface for telephones, terminals and other equipment, An ISDN PBX is not very different conceptually from an ISDN switch, although itis usually smaller and cannot handle as many consversations atthe same time, ‘Customer's equipment (a) Example ISDN system for home use. Carters equipment‘DOTSOL - Total Solution totsol inf DCN 60. Custom’ otfeo — —+l-—— canines tice —| CUT defined four reference points, called R, S, T, and U, between the various devices, These are smarked in the Fig. (b). The Gsriors | U reference point is the connection between the ISDN. exchange in the carrier's office and NTI. At Oigtei | Comers equipment vs : oueee present it is a two-wire copper twisted pair, but at some time in the future it [@ Esme SDN tn wise] ay be replaced by Aber —— optics. The T reference point is what the connector on NTI. provides to the Customer’ equipment pees ee point i the interface between the ISDN PBX and the ISDN terminals. The R reference point isthe connection between the terminal adapter and non-ISDN terminals. Many different kinds of interfaces will be used at R. The ISDN Interface: te, TEES ison COO Teton (a) Basic rate digital pipe ‘The ISDN Bic pipe supports mule (&) Primary rate dita pipe channels interleaved by time division = FD (18 bps) ‘multiplexing. Several channel types oe == o A-4-KHlz analog telephone channel B - 64-kbps digital PCM channel for voice or data C-8- oF 16 kbps digital channel D- 16-kbps digital channel for out-of band signaling, E.- 64-kbps digital channel for internal ISDN signaling H-384,, 1536-, of 1920-kbps digital channel It was not CCIIT intention o allow an arbitrary combination of channels on the digital bit pipe. Three combinations have been standardized so far: |. Basie rate: 2B + ID 2. Primary rate: 23B + 1D (U.S. and Japan) or 30B + ID (Europe) 3. Hybrid: A+ 1C ‘The basic rate and primary rate channels are illustrated in Fig. ‘The basic rate should be viewed as a replacement for POTS (Plain Old Tele, phone Service) for home or small ‘business use. Each of the 64-kbps B channels can handle a single PCM voice channel with 8-bit samples made 8000 times 1 second (note that 64 kbps means 64,000 here, not 65,536). Signaling is on a separate 16-kbps D channel, so the full 64 kbps are available to the user (asin the CCITT 2.048-Mbps system and unlike the U.S. and Japanese TI system). Because ISDN is so focused on 64-kbps channels, we refer to it as N-ISDN (Narrowband ISDN), to contrast it with broadband ISDN (ATM),TOTSOL- Total Solution totsol inf DCN 61 ‘Teetone fe por | ‘Soe ——— rex] eh Databases ‘Subscriber loop and ISDH ‘tna Saree ‘Shanna suchas: Concentual View of ISDN ‘Connection Feats Nema L oN & Local Area Networke (LAN, i } Digital Ciro | peitehed Béstebone — ‘Common ISDN Packet-rwitched Physical Central fietmorie Interface Office etre ® Subscriber Loop ISDN Channel Structures: © "Basic = 64 Kbps + 64 Kbps + 16 Kbps Primary = Multiplexed 64 Kbps Channels Network-based & B.Channel (64 Kbps]: D-Channel (16 Kbps]: Processing Services Digital Voice: @ Signaling: 64 Kbps PCM Basic Low Bit Rate Enhanced Block Diagram of ISDN_Functions: & High-Speed Data: © Low-Speed Data: Circuit-Switched Videotext Packet-Switched Teletext & Other Terminal Facsimile Telemetry Slow-scan Video Emergency Services Emergency management. ISDN Channel Functions:DOTSOL - Total Solution totsol inf ® ATM Protocol Reference Model: Control Plane User Plane Management Plane Higher Layer ATM Adaptation Layer (AAL} Layer Management ATM Layer Physical Layer + Management Plane Includes Plane management which performs management functions related to system as a whole and and provides coordination between all the planes, and Layer management, which performs management functions relating to resources and parameters residing in its protocol entities. Virtual Chennale ‘Virtual Pat Transmission Path Fig ATM Connection Relationships Plane Management The protocol reference Model makes reference to three separate planes * User Plane : Provides for user Information transfer, along with associated controls leg, Flow control, Error control). © Control Plane: Performs Call control and Connection control functions, ttTOTSOL- Total Solution totsol inf DCN 63 ® Call Establishment using Virtual Paths: Request for VCC originates Can quality of service be satisfied Yes ¥ No ¥ Establish a New VPC or VCC Block VCC Or request More capacity ¥ Request Yes J Make granted ? connection No , Reject VCC requestDOTSOL - Total Solution DCN 64 Virtual Path/Virtual Connection Terminology: ® Virtual Channel (VC A generic term used to describe unidirectional transport of ATM cells associated by common unique identifier value. & Virtual Channel Link: A means of unidirectional transport of ATM cells between a point where a VCI value is assigned and the point where the value is translated or terminated. ® Virtual Channel Identifier (VI Identifies a particular VC link for a given VPC. Virtual Channel Connection (VCC): ‘A concatenation of VC links that extends between two points where the adaptation layer is accessed. VCCs are provided for the purpose of user-user, user-network, or network-network information transfer. Cell sequence integrity is preserved for cells belonging to the same VCC & Virtual Path (VP) A generic term used to describe unidirectional transport of ATM cells belonging to virtual channels that are associated by a common unique identifier value. ® Virtual Path Linke: A group of VC links, identified by a common value of VPI, between a point where a VPI value is assigned and the point where that value is translated or terminated. ® Virtual Path Identifier (VPI): Identifies a particular VP link. ® Virtual Path Connection (VPC] A concatenation of VP links that extends between the point where the VCI values are assigned and the point where those values are translated or removed, i.e., extending the length of a bundle of VC links that share the same VPI. VPCs are provided for the purpose of user-user, user-network, or network-network information transfer.DOTSOL - Total Solution www totsol info DCN +65 ‘Twist © Separately insulated ——_t * Twisted together Length * Often “bundled” into cables. * Usually installed in building = POOOOOOCOKL (a) Twisted Pai Outer sheath Outer conductor (O) insulation )—— Inner conductor * Outer conductor is braided shielded (b} Coaxial cab! Inner conductor is solid metal. * Separated by insulating material. © Covered by padding. Core Jacket — ( >) aCladding LD] Sy * Glass or plastic core Light at less than Sere Beet * Laser or light-emitting diode Critical angle is * Specially designed jacket absorbed in jacket * Small size and weight (c] Optical cable Fig: Guided Transmission Media‘TOTSOL - Total Solution totsol info’ DCN _-66 Physical and transmission media The transmission modium is the physical path between transmitter and receiver in a data transmission system. ‘Classified into two categories- guided and unguided. With guided media, the waves are guided along a solid medium, such as copper tisisted pair, copper coaxial cable and optical fiber. The atmosphere and outer space are examples of ‘unguided media that provide a means of transmitting electromagnetic signals but do not guide them; this form of transmission is known as wireless transmission. ‘A number of design factors relating to the transmission medium and to the signal determine the data rate and distance: 1) Band width:- If ll other factors remaining constant, the greater the bandwidth ofa signal, the higher the data rate that can be achieved. 2) Transmission impairments:- Such as attenuation, limits the distance. For guided media twisted pair suffers more impairment than co-axial cable, which in tum, suffers more than optical fiber. 3) Interference:~ Interference fiom competing signals in overlapping frequency band can distort a signal. It can also hhappen by placing near by guided cables. Proper shielding of guided media minimizes this problem. 4) Number of receivers A guided media can be used to construct point-to-point link with multiple attachments Each attachment introduces some attenuation or distortion on the line limiting distance and/or data rate Guided transmission media: For guided transmission media, the transmission capacity, in terms of either data rate or bandwidth, depends critically (on the distance and on whether the medium is point-to-point or multipoint, such as LAN, Three commonly used guided media are twisted pair, coaxial cable, and optical fiber. ‘The point-to-point transmission characteristics of a guided modia ee Total Data Rate Band Width Repeater sqabing ‘Twisted pair 4 Mbps 3 Mz 210 10KM Conanial cable 500 Mbps 350 Miz Ito 10KM Optical fiber 2 Gops 2GHe 10 to 100KM. ‘Twisted pair © Physical description: (Fig—a) A twisted pair consists of two insulated copper wire aranged in a regular spiral pattem. A wire pair acts as @ single communication link. Typically a number of these pars are bundled together into cable by wrapping them in a tough protective sheath. Adjacent pairs in a bundle typically have somewhat different twist lengths to reduce the cross-talk interference, These twist lengths may vary from 2 t0 6 inch, It has thickness from 0,016 to 0,036 inches. Application: Analog, and digital signals can be transmitted through twisted pair. It is used for PBX (private branch exchange) system, Now itis also used for land connection. Its data transfer rate is 10 Mbps to 100 Mbps. © Transmission characteri For analog and digital signals amplifiers are required about every 5 to 6 km and 2 to 3 km respectively. Though it is cheapest but it has limitations in bandwidth, distance, and data rate. Attenuation is a very strong fimetion of fiequency and itis susceptible to interference and noise and can accommodate a number of channels, Impulse noise is also easily introduced in it. Twisting of wines reduces low frequency interference, Iis of nwo categories: (a) Unshiclded twisted pair (UTP) — consists of only the twisted pair without any extra insulation; it is used for telephone wire inside a building, least expensive transmission modia used for LAN, easy to work and install‘TOTSOL - Total Solution totsol info. DCN - 67 (a) Shiclded twisted pair (STP) — has an extra sheath outside the twisted pair for insulation. To improve transmission properties and to eliminate external electro-magnetic interference, twisted pair is surrounded with metallic braid or sheathing that reduces interference. It is expensive and more difficult to work and install but provide better performance at low data rate. & EIA-SOS_A recognizes three categories of UTP cabling: 1) Category 3 (CAT 3) — UTP cable and associated connected hardware whose transmission characteristics are specified up to 16 MHz 2) Category 4 (CAT 4) — UTP cable and associated connected hardwires whose transmission characteristics are specified up to 20MEz, 3) Category 5 (CAT 5) — UTP cable and associated connected hardwires whose transmission characteristics are specified up to 100MHz. © CAT 3 and CAT 5 are commonly used for LAN applications. CAT 5 is more tightly twisted (3 or 4 twists per inch), costlier and having better performance than CAT 3 or CAT 4 (3 or 4 twists per foot) Coaxial cable & Physical description: Coaxial cable like twisted pair consists of two conductors but it is constructed differently to permit it to operate over a wide range of frequencies. It consists of a hollow outer cylindrical conductor that surrounds a single inner wire conductor. Either regularly spaced insulating rings or a solid dielectric material holds the inner conductor in place. The outer conductor is covered with a jacket or shield. A single co-axial cable has a diameter of 0.4 to | inches. Because of its shielded, concentric construction, coaxial cable is much less susceptible to interference and cross talk than in twisted pair. Coaxial cable can be used over longer distances and supports more stations on a shared line on a twisted pair. & Application: Most popular applications are 1) Television distribution 3). Short run computer system links 2) Long distance telephone transmission 4) Local area network Co axial cable can cary over 10,000 voice channels simultaneously using FDM, It supports a large no of devices with a verity of data and traffic types. & Transmission characteristic Its frequency characteristics are superior than that of twisted pair. The principal constraints on the performance are attenuation, thermal noise and inter modulation noise. The latter is present only when several channels (FDM) of frequency bands are in use on the cable. Optical fiber & Physical description: Itis a thin (2-12 micrometer) flexible media capable of conducting an optical ray various glasses and plastics can be used to make optical fibers. The lowest losses have been obtained using fiber of ultra pure fused silica but it is difficult to manufacture so glass fiber are more economical and can provide good performance. = Cylindrical optical fiber cable has three concentric sections: 1) Cores the inner most section and consists of one or more very thin strands or fiber made of glass or plastic. 2) The cladding is of glass or plastic coating that has different optical property than that of core. 3) Jacket is the outer most layer surrounding one or a bunch of cladded fibers is the jacket. Jacket is composed of plastic and other materials layered to protect against moisture, crushing, and other environmental activities.TOTSOL - Total Solution sotsol info DCN 68 & Application: It s used for long distance telephone communication. 1), Greater capacity: - it has a data rate of 2 Gbps over tens of kilometers 2) Sal size and lighter in weight 3) Reduetion in its weight to reduces structural support requirements 4) Lower attenuations: - attenuation is lower than other guided media and it is constant over a wide range. 5) Electromagnetic isolation: - As itis not effected by external electromagnetic induction so it i free from impulse and cross talk noise 6) Itis also inherently difficult to tap. 7), Greater repeater spacing: - Fewer repeater means lower cost and fewer sources of error. ‘= Five basic categories of applications are: © Long haul trunks ‘© Subscriber loops ‘© Metropolitan trunk © Local area networks ‘© Rural-exchange trunk & Transmission characteristic: Rays of light can flow through optical fiber with frequency range 10" to 10" Hz in two different modes. 1), Multi mode: Where angle of incidences of different rays may have variety of values. 2) Single mode: By reducing the radius of the core to the order of a wavelength, only a single angle or mode can pass: the axial ray. ‘Single mode transmission will be less erroneous than multi mode transmission. Another type is multi mode graded index or multi mode step mode. In this mode the variable refraction has the effect of focusing the rays more efficiently than ordinary multimode. ‘Two different light sources are used in fiber optic system: ~ LED and injection lesser diode (ILD). LED is cheaper and has greater temperature rate and greater operational life but ILD is more efficient and can sustain greater data rate & Unguided transmission media: For unguided media, transmission and reception are achieved by means of an antenna. For transmission antenna radiates electromagnetic energy into the medium and during reception it picks up electromagnetic radiation from the surrounding medium, © There are two types of wireless transmission: 1) Directional: Transmitting antenna puts out a focused electromagnetic beam, so the transmitting and receiving antennas must therefore be carefully aligned. Higher frequency results more focusing/performance. 2) Onmi-directional: Transmitted signal spreads out in all directions and can be received by any antenna, Terrestrial Microwave ® Physical description The typical microwave antenna is a parabolic “dish” with typical 10 feet diameter. The antenna is rigidly fixed and focuses a narrow beam to achieve line of sight transmission of receiving antenna. The maximum distance between antennas conforms to 4=7.14 YR) where ‘The messages to be encrypted, known as the plaintext, are transformed by a function that is parameterized bya key. The output of the encryption process, known as the ciphertext, is then transmitted, often by messenger or radio, We assume that the enemy, or intruder, hears and accurately copies down the complete ciphertext. However, unlike the intended recipient, he does not know what the decryption key is and so cannot decrypt the ciphertext easily. Sometimes the intruder can not only listen to the communication channel (passive intruder) but can also record messages and play them back later, inject his own messages, or modify legitimate messages before they get to the receiver (active intruder). The art of breaking ciphers is called cryptanalysis. The art of devising ciphers (cryptography) and breaking them (cryptanalysis) is collectively known as cryptology. It will ofien be usefil to have a notation of relating plaintext, ciphertext, and keys. So we will use the expression C = Ex(P) to mean that the encryption of the plaintext P using key K gives the ciphertext C. Similarly, P= D(C) represents of decryption of C to get the plaintext again, It then follows that Dy (Ex (P)=P Encryption methods have historically been divided into two categories: substitution ciphers and transposition ciphers, We will now deal with each of these briefly as background information for modem cryptography. Substitution Ciphers: In a substitution cipher each letter or group of leters is replaced by another letter or group of letters to disguise it e.g. Caesar cipher. Here for example, attack becomes DWWDEN. st “plaintext: bedefghI nopa R L a jktm © uvwxyz Yciphertext: QWERTYUIOPASDFGHJK XCVBNM This general system is called a mono-alphabetic substitution. In English, for example, ¢ is the most common letter, followed by /, 0, a, 1, i, etc. The most common two letter combinations, or digrams, are #h, in, er, re, and an. The most common three letter combinations, or tigrams, are the, ing, and, and ion. A eryptanalyst trying to break a monoalphabetic cipher would start out by counting the relative frequencies of all letters in the ciphertext. We can also form the substitution table dynamically by finding a commonly occurred word (e.g. accounts) in the document by searching a word with 8 letters and with repeating second and third letters.DOTSOL - Total Solution totsol inf DCN -74 © Transposition Ciphers: Substitution ciphers preserve the order of the plaintext symbols but disguise them, Transposition ciphers, in contrast, reorder the letters but do not disguise them, The following figure depicts a common transposition cipher, the columnar transposition. The cipher is keyed by a word or phrase not containing any repeated letters, MEGABUCK Pleasetr ansferon Plain text emillion Pleasetransferonemilliondollarsto dollarst — myswissbankaccountsixtwotwo omyswiss bankacco Gipher text untsixtw AFLLSKSOSELAWAIATOOSSCTCLNMOMANT otwoabcd — ESILYNTWRNNTSOWDPAEDOBUOERIRICKB In this example, MEGABUCK js the key. The purpose of the key is to number the columns, column I being, under the key letter closest to the start of the alphabet, and so on. The plaintext is written horizontally, in rows The ciphertext is read out by columns, starting with the column whose Key letter is the lowest. Just the reverse ‘method will be applied at the receiving end to get back the original plain text Pseoret-Key Algorithms: ‘Modem cryptography uses the same basic ideas as traditional cryptography, transposition and substitution, ‘Transpositions and substitutions can be implemented with simple circuits. Figure (a) shows a device, known as a P-box (P stands for permutation), used to effect a transposition on an 8-bit input. Ifthe 8 bits are designated from top to bottom as 01234567, the output of this particular P-box is 36071245. By appropriate internal wiring, a Pbox can be made to perform any transposition, and do it at practically the speed of light. Pox Sto Product cer BS] 1S] [S] E_ Substitutions are performed by S- & Be ae boxes, as shown in Fig (b). In this PE Pete Mtg] PE example a 3-bit plaintext is entered shal and a 3-bit ciphertext is output. {s| [%] |e The 3-bit input selects one of the @ o @ eight lines exiting from the first Basic elements of product cipher (0 P-bx.(b bn. () Product. stage and sets it to 1; all the other lines are 0. The second stage is a box. The third stage encodes the selected input line in binary again. With the wiring shown, ifthe eight octal numbers 01234567 were input one after another, the output sequence would be 24506713. In other words, 0 has been replaced by 2, 1 has been replaced by 4, ete. Again, by appropriate wiring of the P-box inside the S-box, any substitution can be accomplished. The real power of these basic elements only becomes apparent when we cascade a whole series of boxes to form a product cipher, as shown in Fig, (c), In this example, 12 input lines are transposed by the first stage. Theoretically, it would be possible to have the second stage be an S-box that mapped a 12-bit number onto another 12-bit number. However, such a device would need 212 = 4096 crossed wires in its middle stage. Instead, the input is broken up into four groups of 3 bits, each of which is substituted independently of the others, Although this method is less general, it is still powerful, By including a sufficiently large number of stages in the product cipher, the output can be made to be an exceedingly complicated function of the input TTTDOTSOL - Total Solution totsol infor DCN 75 DES: ‘This cipher, DES (Data Encryption Standard), was widely adopted by the industry for use in security products. An outline of DES is shown in Fig. (a). Plaintext is encrypted in blocks of 64 bits, yielding 64 bits of ciphertext, The algorithm, which is parameterized by a S6-bit key, has 19 distinct stages. The first stage is a key independent transposition on the 64-bit plaintext. The last stage is the exact inverse of this transposition, The stage prior to the last one exchanges the lefimost 32 bits with the rightmost 32 bits. The remaining 16 stages are functionally identical but are parameterized by different functions of the key. The algorithm has been designed to allow decryption to be done with the same key as encryption. The steps are just run in the reverse order. ‘The operation of one of these intermediate stages is illustrated in Fig, (b). Each stage takes two 32-bit inputs and produces two 32-bit outputs, The left output is simply a copy of the right input. The right output is the bit-wise EXCLUSIVE OR of the left input and a function of the right input and the key for this stage, Ki All the complexity lies in this function. 4 WUT LU oe EES ——, —s— FU TL The function consists of four steps, carried out in sequence. First, a 48-bit number, E, is constructed by expanding the 32-bit R,. ; according to a fixed transposition and duplication rule. Second, E and K, are EXCLUSIVE ORed together. This output is then partitioned into eight groups of 6 bits each, each of which is fed into a different S-box. Each of the 64 possible inputs to an S-box is mapped onto a 4-bit output. Finally, these 8 x 4 bits are passed through a P-box. Stephene (t) In each of the 16 iterations, a different key is used. Before the algorithm starts, a 56-bit transposition is applied to the key. Just before each iteration, the key is partitioned into two 28-bit units, each of which is rotated left by a number of bits dependent on the iteration number. Ki is derived from this rotated key by applying yet another 56-bit transposition to it. A different 48-bit subset of the 56 bits is extracted and permuted on each round,