Chapter 2. The Physical Layer

Transcription

1 Chaper 2 The Physical Layer

2 The Physical Layer Defines he mechanical, elecrical and iming inerfaces o he nework Transmission media - guided (copper and fiber opics) - wireless (radio erresrial) - saellie

3 The Theoreical Basis for Daa Communicaion Fourier Analysis Bandwidh-Limied Signals Maximum Daa Rae of a Channel

4 Time- domain represenaion of signals x(); R x[n] ; n Z 2 1, n (a) (b) Fig.2-1 (a) coninuous ime signal (b) discree ime signal

5 Represenaion of he sine wave signal (fundamenal coninuous signal) x() [V] [ms] () x = sin π V [ ] (a) Ampliudine 1 Linia specrală a sinusoidei,5 1 1,5 2 f [khz] (b) Fig.2-2 (a) Time domain represenaion and (b) frequency domain represenaion of he sine wave signal

6 Frequency domain conceps x(), x 1 (), x 2 () Semnal sumă [ms] x x x 1( ) = 3 sin2π 2 () = sin2π 3 () = x () + x () (a) Ampliudine A A/3,5 1 1,5 2 2,5 3 3,5 f [khz] (b) Fig. 2-3 Time domain and frequency domain represenaion for periodic signals

7 Frequency domain conceps- periodic signals (b) (a) = = + + = ) ( k k k k T sin k b cos k a a x ω ω () ( ) = + = cos k k o k T k A x ϕ ω = ) ( 1 2 T T d x T a d cos k x T a T T k ) ( 2 ω = d sin k x T b T T k ) ( 2 ω = () = = k jk k T e c x ω () = 1 T jk T k d e x T c ω { } = = = 2 A c c arg A c k k k k ϕ (c)

8 x ( ) Frequency domain conceps- periodic signals 1-1 A k τ T /2 4 π T 1 a) x 4 π () = cos ( 2k + 1) A k A k k = 1 2k = π (2k + 1) 1 ω,1 A B ef = 9 f [ Hz] π 2 k ϕ k ω 3ω 5ω 7ω ω ω 3 ω 5 ω 7ω ω -π/2 b) Fig. 2-4 Time domain (a) and frequency domain (b) represenaion for square wave signal

9 Frequency domain conceps- aperiodic signals 1/τ x ( ) 2 τ 2 τ = R d e x X j ω ω ω ; ) ( ) ( > < = 2 ; 2 ; 1 ) ( τ τ τ penru penru x = = ) ( τ τ ω ωτ ω sinc d e X j X(ω) ω 1 τ 2 π τ 2 π τ 4 π τ 4 π τ 6 π τ 6 π ω 2π X(ω) 2 1 { } ) ( ) ( ) ( ω ω x X I = Fig. 2-5

10 x ( ) Relaionship beween daa rae and bandwidh 1 τ T /2 T T 1 τ= = 2 2f daa rae = 2 f bps - 1 1) Digial ransmission B C =4MHz A k 4 π 1 a) f = 1MHz B = (5 1) f = 4MHz daa rae = 2Mbps k 2) Digial ransmission B C =8MHz ω 3 ω 5 ω 7ω ω f = 2MHz = 2f ' B = (5 1) 2 f = 8MHz daa rae = 4Mbps b) Fig. 2-4 Time domain (a) and frequency domain (b) represenaion for square wave signal

11 Bandwidh-Limied Signals Fig. 2-6 Effec of bandwidh on a digial signal

12 Bandwidh-Limied Signals (2) (d) (e) Successive approximaions o he original signal.

13 birae = b bps Bandwidh-Limied Signals (3) 8 = sec b b 3 24 f = Hz BC = 3Hz = 8 b b 8 Fig. 2-7 Relaion beween daa rae and harmonics

14 Maximum daa rae of a channel Nyquis, Shannon max imum daa rae = 2H log 2 V bps The noise presence is appreciaed by he signal o noise raio = he raio of he signal power o he noise power S/N Usually he raio iself is no quoed; insead he quaniy is given The unis are called decibels (db) 1log 1 Shannon s major resul is ha he maximum daa rae of a noisy channel whose bandwidh is H Hz and whose signal - o - noise raio is S/N is given by: S N max imum number of bis / sec = H log (1 + S / N) 2

15 Maximum daa rae of a channel Expl: A channel of 31Hz bandwidh wih a signal o hermal noise raio of 3dB (ypical parameer of he analog par of he elephone sysem) can never ransmi much more han 3bps C = 31 log (1 + 1) = 3894bps 2 The capaciy indicaed is referred o as he error free capaciy NOTE if he acual informaion rae on a channel is less han he errorfree capaciy han i is heoreically possible o use a suiable signal code o achieve error-free ransmission hrough he channel - Shannon s heorem does no sugges a means for finding such codes The measure of efficiency of a digial ransmission is he raio: C/H [bps/hz] max imum number of bis / sec = H log (1 + S / N) 2

16 Bandwidh, bi rae, ransmission ime Many people confuse bandwidh and bi rae, bu you should keep he disincion The bi rae and he number of bis o ransmi deermine he ransmission ime.

17 Propagaion

18 Propagaion(2) Propagaion is he ime aken by he fron of a signal o reach he desinaion. I is independen of he bi rae. Propagaion of an elecromagneic signal is he speed ( also called celeriy) of ligh. I depends on he wavelengh and he elemen in which he signal is propagaing. Acousic waves move a 3m/s. Wha is he propagaion ime if we use an acousic phone sysem beween wo ciies which are 1 km apar?

19 Propagaion(3) Examples 1- way propagaion ime = 2km.1ms m s = 8 21 / ransmission ime = 1 8b / b s = 8ms recepion ime =.1ms + 8ms = 8.1ms

20 Throughpu Defines how much daa can be moved by ime uni. I is equal o he bi rae if here is no proocol. However in mos pracical cases he hroughpu is less han he bi rae for wo reasons: - proocol overhead: proocols like UDP use some byes o ransmi proocol informaion. This reduces hroughpu. If you send one-bye message wih UDP hen for every bye you creae an Eherne packe of size: = 53byes hus he maximum hroughpu you could ever ge a UDP service inerface if you use a 64kb/s channel would be 1.2 kb/s. - proocol waiing imes: some proocols may force you o wai for some even.

21 Analog and Digial Daa Transmission The naure of daa The acual physical means used o propagae he daa Wha processing or adjusmens may be required along he way o ensure ha he received daa are inelligible Analog and digial Daa Signaling Transmission

22 Daa Analog and Digial Daa Transmission - Expl: - Audio or acousic daa: frequency componens: 2Hz 2kHz analog daa -Video daa in erms of he viewer analog daa Tex or characer srings digial daa- codes used for represening each characer by a sequence of bis: ASCII( American Sandard Code for Informaion Inerchange)- 7 bis Signals - Acousic daa specrum of speech: 2Hz 2kHz sandard specrum for voice signal: 3Hz- 34Hz - Video daa Video signal: bandwidh: 4MHz

23 Analog and Digial Daa Transmission Binary digial daa - The bandwidh will depend in any specific case on he exac shape of he wave form and on he sequence of 1s and s.

24 Analog and Digial Daa Transmission Daa and signals Analog daa (voice sound waves) Telephone Analog signal Digial daa (binary volage pulses) Modem Analog signal (a) Analog signals; represen daa wih coninuously varying elecromagneic wave Analog daa Codec Digial signal Digial daa Digial ransmier Digial signal (b) Digial signals; represen daa wih sequence of volage pulses

25 Analog and Digial Daa Transmission Transmission (a) daa and signals Analog daa Analog signal Two alernaives: (1) signal occupies he same specrum as he analog daa; (2) analog daa are encoded o occupy a differen porion of specrum Digial signal Analog daa are encoded using a codec o produce a digial bi sream; Digial daa Digial daa are encoded using a modem o produce analog signal Two alernaives: (1) signal consiss of wo volage levels o represen he wo binary values; (2) digial daa are encoded o produce a digial signal wih desired properies

26 Analog and Digial Daa Transmission Transmission (b) Treamen of signals Analog signal Digial signal Analog ransmission Is propagaed hrough amplifiers; same reamen wheher signal is used o represen analog daa or digial daa. No used Digial ransmission Assumes ha he analog signal represens digial daa. Signal is propagaed hrough repeaers; a each repeaer, digial daa are recovered from inbound signal and used o generae a new analog oubound signal Digial signal represens a sream of 1s and s, which may represen digial daa or may be an encoding of analog daa; Signal is propag. hrough repeaers; sream of 1s and s is recovered from inbound signal and used o generae a new digial oubound signal

27 Analog or Digial Daa Transmission? Digial ransmission Digial echnology LSI, VLSI echnology Daa inegriy repeaers Capaciy uilizaion High degree of muliplexing ime division Securiy and privacy Encrypion echniques Inegraion Voice, video and digial daa

28 Transmission Impairmens Aenuaion and aenuaion disorion Delay disorion Noise

29 Transmission Impairmens (2) Aenuaion and aenuaion disorion - guided media: consan no of db/uni disance - unguided media: complex funcion of disance Consideraions for ransmission engineer: 1) - he received signal mus have sufficien srengh so ha he elecronic circuiry in he receiver can deec and inerpre he signal 2)- he signal mus mainain a level sufficienly higher han he noise, o be received wihou error 3) - aenuaion is an increasing funcion of frequency

30 Transmission Impairmens (3) Delay disorsion - he velociy of propagaion of a signal hrough a guided media varies wih frequency - criical for digial daa - inersymbol inerference limiaion o maximum bi rae over a ransmission conrol - equalizing echniques

31 Transmission Impairmens (4) Noise unwaned signals insered somewhere beween ransmission and recepion - Thermal noise - Inermodulaion noise - Crossalk (diafonia) - Impulse noise

32 Transmission Impairmens (5) Thermal noise - Noise power densiy N N = kt[ was / herz] k Bolzmann s consan T emperaure [K] - Thermal noise expressed in was presen in a bandwidh of W herz: N = ktw[ was/ herz]

33 Inermodulaion noise Transmission Impairmens (6) - effec: produces signals a a frequency ha is he sum or difference of he wo original frequencies or muliple of hose frequencies; - produced by some nonlineariy in he ransmier, receiver or ransmission sysem; Crossalk (diafonia) Impulse noise - unwaned coupling beween signals pah; - a minor annoyance for analog signal - he primary source of error in digial daa communicaion

34 Guided Transmission Daa Twised Pair Coaxial Cable Fiber Opics

35 Twised Pair (a) Caegory 3 UTP. (b) Caegory 5 UTP.

36 Coaxial Cable A coaxial cable.

37 Fiber Opics (a) Three examples of a ligh ray from inside a silica fiber impinging on he air/silica boundary a differen angles. (b) Ligh rapped by oal inernal reflecion.

38 Fiber Cables (a) Side view of a single fiber. (b) End view of a sheah wih hree fibers.

39 Fiber Cables (2) A comparison of semiconducor diodes and LEDs as ligh sources.

40 Wireless Transmission The Elecromagneic Specrum Radio Transmission Microwave Transmission Infrared and Millimeer Waves Lighwave Transmission

41 The Elecromagneic Specrum The elecromagneic specrum and is uses for communicaion.

42 Public Swiched Telephone Sysem The Local Loops The Trunks The Swiches

43 Srucure of he Telephone Sysem A ypical circui roue for a medium-disance call.

44 Major Componens of he Telephone Sysem Local loops Analog wised pairs going o houses and businesses Trunks Digial fiber opics connecing he swiching offices Swiching offices Where calls are moved from one runk o anoher

45 The Local Loop The use of boh analog and digial ransmissions for a compuer o compuer call. Conversion is done by he modems and codecs.

46 Circui Swiching (a) Circui swiching. (b) Packe swiching.

47 Packe Swiching PS is more faul oleran PS does no wase bandwidh and hus is more efficien PS uses sore and forward ransmission CS is compleely ransparen The sender and receiver can use any bi rae forma and framing mehod. The carrier does no know or care Wih PS he carrier deermines he basic parameers The charging algorihm: CS: charging based on disance and ime PS: he volume of raffic

48 Circui Swiching Packe Swiching (a) Circui swiching (b) Message swiching (c) Packe swiching

49 Differences A comparison of circui swiched and packe-swiched neworks.

50 Muiplexing Saisical and nonsaisical muliplexing Def: muliplexing means puing several sources (ogeher) on he same link Ways of muliplexing: - Sharing ime slos = emporal muliplexing - Sharing frequency bands = frequency muliplexing PS does no wase bandwidh and hus is more efficien Temporal muliplexing is used in many elecommunicaions neworks: - In CS each source is allocaed wih one ime slo; In saisical muliplexing daa unis are samped wih idenifiers so ha a source may send daa a will

51 Muliplexing Saisical muliplexing T 1 4 T 2 T 3 PS: The bi rae of he oupu (4) is ofen less hen he sum of he bi raes of all inpus (1 o 3); Special mechanisms, called congesion conrol are required o avoid ha packe losses happen o frequenly. CS: The bi rae of he ougoing circui (4) is a leas equal o he sum of he incoming circuis bi raes (1 o 3); There is no loss of daa.