Commuunieition | Systems 2!Communication Systems - | Third Revised Edition : Third Edition Reprint June 2003 August 2004 4° rh] Feet llrightsreserved. No part of this book should be reproduced in any form, Electronic, Mechanical, Photocopy or any information storage and retrieval system without prior permission in writing, from Technical Publications, Pune. Published by : ° Technical Publications rune Amit Residency, Office No. 1, Stilt floor, 412, Shaniwar Peth, Off N.C. kelkar road, Behind Mandar Lodge, Pune -411 030 Phone : (020) 24495496 , Tele/Fax : (020) 24495497 Email : [email protected] Distributed by : Pragati Books Pvt. Ltd. 119, Budhwar Peth, Pratibha Book Distributors Shop no. 3, 1st floor, Lokratna Commercial Complex, dihansi Rant square, Sltabuld), Nagpur - 440 012 Phone : 547129, Thakkar Enterprises Maner buiding, Near Kala Imam Talim, ‘Shaniwar Peth, Kolhapur- 416106 Phone : 520106 Retail Book Shops : Pragati Book Centre 197, Budhwar Peth, Pune - 411 002 Phone - (020) 24458887 Pragati Agencies 676B, Buchwar Peth, Opp. Jogeshwari Mandir Lane, Pune - 411 002 Ph. (020) 4017784 385, Printer : Alert DTPrinters Pune Sinnogad Hoag, seen tas Pragati Books Pvt. Ltd. SNP. Road Ramrata Co . Society, Phong" 23869976 1 29856999 Pragati Book House 1678-14th, Main Road, Prakash N: Bangalore - 5 Phone : 23324037, Fax : 23324848 jar, Pragati Books Pvt. Ltd. 34, V. V. Golani Market, Navi Peth, Jalgaon - 425 001 jone : 220395 Pragati Book Corner Indira Niwas, 111-A, Bhavani Shankar Road, Dadar (W), Mumbai - 400 028 Phone : (022) 24223526 Also available at other leading book shops ai)transmission bandwidth and signal to noise power ratio. Such transmission is also called bandpass transmission. Contents and Organization First chapter introduces the subject of communication in general. Chapter second is fully devoted for discussion of fourier transforms, properties and examples. Chapter third is based ‘on probability theory and random variables. The definition of probability, conditional and marginal probabilities are discussed. The cumulative distribution function, probability density function and their properties are also discussed. Autocorrelation and cross correlation functions and their properties are also presented in this chapter. A brief description of random processes is presented at the end of this chapter. Fourth chapter presents mathematical representation of noise. It explains about linear filtering, components of noise, performance of modulation systems in presence of noise etc. Chapter five explains sampling techniques. Various types of sampling, aliasing and sampling theorem are expiained in this chapter. It also explains about TDM in brief. ‘Sixth chapter is based on source coding techniques. It explains PCM, DM, ADM, DPCM etc techniques. The noise analysis of this techniques is also presented. Voice coders is also added in this chapter. Seventh chapter presents digital multiplexing. It explains various types of digital multiplexers, T1 system etc. This chapter also presents line codes, synchronization techniques, Intersymbol interference (ISI) and eye patterns. Chapter eight explains digital modulation techniques in detail. It explains transmitters, receivers, BW requirements, spectral properties, power requirements, constellation diagrams etc of almost all the types of modulation techniques. Ninth chapter presents detection of digital signals in presence of noise. Itexplains matched filter, optimum filter, error probabilities of various modulation techniques ete. in detail. Attempts are made to make this text as lucid as possible. Large number of solved examples and exercises are included for better understanding. However there is‘a change of errors due. to typing, calculations etc. Any suggestions or criticism in this regard will be highly appreciated. Acknowledgements| Itis a great pleasure for me to thank all the readers for their suggestions and encouragement. 1am also thankful to Shri. Avinash Wani & Shri. Ravindra Wani and entire team of Technical Publications for their efforts to publish this book. Author (iv)Table of Contents | Chapter-1 Introduction 1.4 Basic Communication Systems 1.2 Modulation and Demodulation in Communication System 1.3 Electromagnetic Waves : The Carriers of Electric Signals..... 1.4 Analog Communication .... 1.5 Digital Communication ..... 1.5.3 Digital Modulators and Demodulators . . meariaha he ecmmaem eee eaten 14 1.5.4 Communication Channel... 4.7 Multiplexing Techniques .......ssssesse -2 Signals and Systems 2.4.1 Deterministic and Random Signals . . 2.1.2 Continuous and Discrete or Analog and Distal Signals 2.1.3 Periodic and Non-periodic Signals. . . 2.2 Energy and Power of the Signals. 2.3. Phasors and Line Spectrums... 2.3.1 Double Sided Spectrums 24 Systems... 24.1 Linear and Nonlinear systems. . 2.4.2 Time Invariant and Time Varying Systems. 24.3 Causal and Noncausal Systems ......... 2.4.4 Stable and Unstable Systems ........ 2.5 Important Signals and Functions used in Mathematical Analysis of Communication Systems ... 2.6 Fourier Series . 2.6.1 Quadrature/Trigonometric Fourier Series 26.2 Polar Fourier Series . “ 26.3 Exponential Fourier Series (or Complex Exponential Fourier Series) we w2.64 Spectral Properties of Periodic Power Signals 2.6.5 Dirichlet Conditions 2.6.6 Exponential Fourier Spectrum and Concept of Negative Frequency . 2.7 Parseval’s Power Theorem 2.8 Fourier Transforms .. 2.8.1. Definition of Fourier Transform . . 2.8.2 Existence of Fourier Transform . . 2.8.3 Properties of Fourier Transform . 2.8.3.1 Linearity (Superposition) . . . 2.8.3.2 Time Scaling 2.8.3.3 Duality or Symmetry Proper 104 2.8.3.4 Time Shifting . 107 2.8.3.5 Frequency Shiting. 107 2.8.3.6 Area Under x(t)... » 108 2.8.3.7 Area Under X(F).. . 110 2.8.3.8 Differentiation in Time Domain 110 28.3.9 Integration in Time Domain. ©... ee ee . 110 2.83.10 Conjugate Functions . ; 2. MM 2.8.3.11 Multipication in Time Domain (Muliplication Theorem)... 2... . 12 2.8.3.42 Convolution in Time Domain (Convolution Theorern) 119 2.8.4 Fourier Transform of a Periodic Function 2.9 Rayleigh’s Energy Theorem . 2.10 Spectral Density Functions 2.10.1 Energy Spectral Density...... 2.10.2 Power Spectral Density bi i 2.40.3 Properties of Spectral Density Function . . 2.10.34 Properties of Energy Spectral Density Function. 2.10.3.2 Properties of Power Spectral Density Function _. 176 2.11 Autocorrelation FUMCtION .........cssscesessseceeceseceseesssecsecesssssesissesses BS 2.11.1 Autocorrelation of Energy Signals . .. 2.14.2 Autocorrelation Function of Periodic Signals . 2.12 Cross Correlation FUnctions ........csssssssssssssssessscccesssstsssstesssssssecsssessssssssssees 204 2.12.1 Cross Correlation of Energy Signals. 2.12.2 Cross Correlation of Periodic Signals. 2.13 Frequency Response of LT! Systems. 2.43.1 Frequency Response........... = 2.13.2 Distortionless Transmission Through System ... . . (vi)2.13.3 Ideal Low Pass Filters... 0.0.6.0... cece eee eect e eee ee eee eee eeetnee ee es 210 Chapter -3 _ Probability Theory and Random Processes 215 --3.1 Basic Terminologies Related to Probability 3.1.1 Experiment 3.1.2 Sample Space * 3.1.3 Event. 3.2 Probability... 3.2.1 Definition of Probability 3.2.2 Properties of Probability 3.2.3 Conditional Probability 3.2.4 Bayes' Rule Or Bayesian Policy . 3.2.5 Probabilities of Statistically Independent eat, 3.3.3 Cumulative Distribution Function (CDF)... . 3.3.3.1 Properties of CDF 3.3.4 Probability Density Function (PDF). .....sssssssssseceeseseeseesneeeeseseneees 248 Cha Popatbed POF: Lora ye ce EN IS ea ess 248 Lis uneae ‘i 3.3.5.1. Properties of Joint Distribution Function. 3.38 Joint Probabilty Density Function. 3.3.6.1 Propertios of Joint PDF re 3.3.6.2 Relationship Between Joint PDF and Probably a 3.3.7_ Marginal Densities . 3.38 Conditional Probability Density Funciion. .. 3.3.8.1 Properties of Conditional PDF... 3.3.9 Statistical Averages of Random Variables . . 3.3.9.1 Mean/Average or Expected Value soecmcane : 3.4. Probability Models.. 3.4.2 Poisson Distribution . . 3.4.3 Uniform Distribution3.4.4 Gaussian or Normal Distribution . . 3.4.5 Rayleigh Distribution ........ 3.4.6 Comparison of Probability Models 3.6 Random Processes ..... 3.6.1 Ensemble and Time Averages . 3.6.1.1 Ensemble Averages © 3.6.1.2 Time Averages . 3.6.2 Stationary and Nonstationary Random Processes... swe se eee sees sees eee 3.6.3 Ergodic Process... see eee eee ss 3.7 Error Function 3.8 The Function Q(u) ... Chapter-4 Mathematical Representation of Noise 354 4.1 Sources of Noise... 4.1.1 ExteralNoise 2.2.0... ...ce cee 4.1.1.1 AtmosphericNoise 2 ee .. 1 4.1.1.2 Extraterrestrial Noise . 3% ¥ 351 444.3 Industrial Nose... ee . 352 4.1.2 intemal Noise .. 362 4.1.2.1 Thermal or Thermal Agtaton ‘or Johnson Noise 353 44.22 ShotNose 4.1.2.3 Partition Noise 4.1.2.4 Low Frequency, or Flicker Noise 4.4.25 High Frequency or Transit Time Noise 4.2 Noise Due to Several Amplifiers in Cascade . 4.3 Noise Factor ABA Noise Figure... eeceeee eee 43.2 Cascade Connection of Amplifiers (iss Formula). 4.4 Noise Temperature ...... 4.5 White Noise .. 4.6 Frequency Domain Representation of Noise 4.7 Spectral Components of Noise ..... 4.8 Superposition of Noises .. 4.9 Linear Filtering .. 434 RO Lowpete FINO sscoccccrc ne (wit)4.0.2 deal Low-pass Filet seccsemeravesrreenreeenreeaneesceess reas ID 49.3 Ideal Bandpass Filter .. 4.10 Noise Bandwidth .. 4.11 Quadrature Components of Noise 4.12 Performance of AM in Presence of Noise 4.12.1 Performance of SSB-SC in Presence of Noise Chapter-5 Sampling of Analog Signals 398 5.1 Sampling Theorem... 5.1.1 Sampling Then Fur Eow Piss Signals in Time eat 5.4.2 Aliasing 5.1.3 Reconstruction iter (lowpass fier or interpolation filter) 5.4.4 Sampling Theorem in Frequency Domain 5.4.5 Sampling of Bandpass Signals. §.2 Sampling Techniques ... 5.2.1 Ideal Sampling or instantaneous Sampling or Impulse Sampling . 5.2.2 Natural Sampling or Chopper Sampling.............606+ . 5.2.9 Flat Top Sampling or Rectangular Pulse Sampling 9.2.3.1 Aperture Effect. . 2 5.2.4 Comparison of Various Sampling oatusa: 5.3 Analog Pulse Modulation Methods §.3.1 Pulse Amplitude Modulation .. . 5.3.1.1 Flat Top Sampled PAM aime oe ee om 5.3.1.2Naturaly Sampled PAM . Lee 5.3.1.3 Ideally or Instantaneously Sampled PAM... . 5.3.1.4 Transmission Bandwidth of PAM Signal 5.3.1.5 Disadvantages ofPAM .. . se eer e 4.3.2 Pulse Duration and Pulse Position Modulation 00 8G 5.4 Time Division Multiplexing (PAM/TDM System) ........... 5.4.4 Synchronization in TDM System ... §.4.2 Crosstalk and Guard Times 0. Chapter -6 Waveform / Source Coding Techniques 474 6.1 Pulse Code Modulation... 6.1.1 PCM Generator. 6.1.2 Uniform Quantizer ()6.1.3 Transmission Bandwidth in PCM ... . . oe 475 6.14 PCM ROCONOE ieee AID, 6.1.5 Quantization Noise and Signal to Noise Ratio in PCM. 6.1.6 Nonuniform Quantization and Companding in PCM. wi 6.1.6.1 Necessity of Nonuniform Quantization for Speech Signal_. E rere) 6.1.6.2 Companding in PCM. 6.2 Delta Modulation 6.2.1 Advantages of Delta Modulation. 6.2.2 Disadvantages of Delta Modulation 6.2.2.1 Siope overload distortion (startup error)... 6.2.22 Granular Noise (Hunting). _. 6.3 Adaptive Delta Modulation .. 6.3.1 Advantages of Adaptive Delta Modulation . 6.4 Differential Pulse Code Modulation 6.5 Comparison of Digital Pulse Modulation Methods .. 6.6 Speech Coders (VOCODERS).. 528, 6.6.1 Voice Model ........... 528 6.6.2 Block Diagram of VOCODER e 529 6.63 Chanel VOCODER. 0. 6.6.4 Linear Predictive Coding (LPC)... 82 Chapter-7 Digital Multiplexing 534 7.1 Digital Multiplexers... 1.4.1 Types of Digital Multiplexers. 7.1.2 Multiplexing Hierarchies 7.1.3 PCM/TDM System... .. was L . 7.1.3.4 Multiplexing Hierarchy... se ratech Tabac OO 7.4.32 Multiple Channel Frame Aignment For TOM/PCM(TSysiem) ...... . . 536 7.1.4 Quasi Synchronous Mulfiplexing . 7.2 Digital Formats or Line Coding .. 7.2.1 Unipolar RZ and NRZ 7.2.2 Polar RZ and NRZ . : 7.233 Bipolar NRZ [Pseudo-trinary or Altemate Mark Inversion (AMI)} 7.24 Split Phase Manchester 7.25 Polat Quatemary NRZ 1.2.6 Gray Codin. 7.2.7 Meary Coding . 0)7.2.8 High Density Bipolar (HDB) Signaling . 7.2.9 Desirable Properties of Line Codes .. . 7.2.40 Spectral Features of Line Codes 7.2.14 Comparison of Line Coding Formats . . 7.3 Synchronization Techniques... 7.3.1 Carrier Synchronization (Carrier Recovery) 7.3.1.1 Carrier Synchronization Using M™ Power Loop 56: 7.3.1.2 Costas Loop for Carrier Synchronization. es ee 563 7.3.2 Bit and Symbol Synchronization (Clock Recovery) . . 7.3.2.1 Closed Loop Bit Synchronization sbetes: 7.3.2.2 Early Late Synchronizer... ss 7.3.3 Frame Synchronization 7.4 Scramblers and Unscramblers ..... 7.4.4 Scrambler Operation... 7.4.2 Unscrambler Operation. . 7.5 Intersymbol Interference .. sire 7.5.1 Baseband Transmission of Binary Data 7.5.2 Intersymbol Interference (ISI) Problem. 7.5.3 Nyquist Pulse Shaping Criterion............ EBA Wal St SS ncn ctr npn nnn tcc EE 7.6 Eye Pattem 7.7 Equilization. 7.7.4 Tapped Delay Line Fitter. 7.1.2:Adaptve Equalization Chapter-8 Digital CW Modulation Techniques 585 8.1 Binary Phase Shift Keying (BPSK) . 8.1.1 Generation and Reception of BPSK Signal . 8.1.1.1 Generator of BPSK Signal 8.1.1.2 Reception of BPSK Signal... 8.1.2 Spectrum of BPSK Signals. 8.1.3 Geometrical Representation of BPSK Signals 8.1.4 Bandwidth of BPSK Signal . . 8.1.5 Drawbacks of BPSK : Ambiguity in Output Signal . 8.2 Differential Phase Shift Keying (DPSK) 8.2.1 DPSK Transmitter and Receiver. 8.2.1.1 Transmitter / Generator of DPSK Signa. ©. 2 ee 5o7 (xi)G2AZOPSKRecover ets 599 8.2.2 Bandwidth of DPSK Signal . 601 8.2.3 Advantages and Disadvantages of DPSK - 601 8.3 Differentially Encoded PSK (DEPSK) .. 8.3.1 Transmitter and Receiver of DEPSK Signal 8.3.2 Advantages and Disadvantages of DEPSK . é wee 604 8.4 Quadrature Phase Shift Keying .... 8.4.1 OPSK Transmitter and Receiver 2 BG 8.4.1.1 Offset QPSK (OQPSK) or Staggered QPSK Transmitter 8.4.1.4 Carrier Synchronization in QPSK # wis b i‘ 610 8.4.2 Signal Space Representation of QPSK Signals 641 8.4.3 Spectrum of QPSK Signal_. 8.4.4 Bandwidth of QPSK Signal. 8.4.5 Advantages of QPSK . 8.5 M-ary PSK..... 8.5.1 Signal Space Diagram . . 8.5.2 Power Spectral Density of M-ary PSK .. 8.5.3 Bandwidth of M-ary PSK... . 8.5.4 Distance Between Signal Points (Euclidean Distance) 8.5.5 Transmitter and Receiver of M-ary PSK... . 8.5.5.1 M-ary PSK Transmitter... 85.5.2 M-aryPSK Receiver... . 8.6 Quadrature Amplitude Shift Keying (QASK) [or Quadrature Amplitude Modulation (QAM)... 8.6.1 Geometrical Representation and Euclidean Distance of QASK Signals (or Signal Spat Representation or Signal Space Constellation) ..........c.c:eceeeeeeeeeeee eee 627 8.6.2 Transmitter and Receiver of QASK .. . .. 628 8.6.2.1 Transmitter of QASK Signal for 4 8.6.2.2 Receiver of QASK Signal . . . 8.6.3 Power Spectral Density and Bandwidth of QASK. Signal 8.6.4 Comparison Between QASK and QPSK . 8.7 Binary Frequency Shift Keying (BFSK) 8.7.1 BFSK Transmitter. 7 8.7.2 Spectrum and Bandwidth of BFSK. B23 BFSK Recetas seri cra ae ea BSB. Symbol (xii)8.7.4 Geometrical Representation of Orthogonal BFSK or Signal Space Representation of Orthogonal BFSK ... 637 87.5 Geometrical Representation of Non Orhogonel BFSK Signals. 638 8.7.6 Advantages and Disadvantages of BFSK. ... 639 8.8 M-ary FSK.. 8.8.1 Transmitter and Receiver of FSK BBS A Treinen oe a Ba 8.8.1.2 Receiver 840 8.8.2 Power Spectral Density and Bandwidth of M-ary FSK........... oo 641 88.3 Geometrical Representation of M-ery FSK or Signal Space Representation 8.9 Minimum Shift Keying (MSK)... 8.9.1 Signal Space Bapesseninbwr of MSK | ira Distance between the Signal Points (or Geometrical representation of MSK)...........0seeeseeeeu te 8.9.2 Power Spectral Density and Bandwidth of MSK. 8.9.3 Phase Continuity in MSK reset eens 8.9.4 MSK Transmitter and Receiver........ 4. nsmitter 8.9.4.2 MSK Receiver . Lan 8.9.5 Advantages and Disadvantages of MSK as Compered to 0 PSK... iM eacesenebaesivaineas 8.9.6 Gaussian MSK .. 8.10 Amplitude Shift Keying or ON. OFF ‘Keying . 8.10.1 Signal Space Diagram of ASK.......... severe 8.10.2 Generator and Detector of ASK 659 8.11 Comparison of Digital Modulation Techniques... Chapter-9 Detection of Binary Signals in Presence of Noise 663 9.1 Integrate and Dump Filter (Receiver) .. 9.1.1 Signal to Noise Ratio of the Integrator and Dump Filter 9.1.2 Probability of Error in Integrate and Dump Filter Receiver 9.2 Optimum Receiver (or Optimum Filter)... 9.2.1 Probability of Error of Optimum Filter. 9.2.2 Transfer Function of the Optimum Filter. 9.3 Matched Filter .. 9.3.1 Impulse Response of the Matched Filter . 9.3.2 Probability of Error of the Matched Filter . 9.3.3 Properties of Matched Filter 9.4 Correlator .. 9.5 Error Probabilities of Baseband Signaling Schemes (xiii)9.5.1 Detection of PCM signal........ 9.5.2 Detection of ASK Signal . 9.5.3 Detection of PSK Signal. . si 9.5.4 Detection of FSK Signal... .. 0.22... 2.. cece eee 915.5 Error Probability of DPSK Appendix-A_Trigonometric Relations 732 Appendix-B Derivatives 733 Appendix-C Integration 734 Appendix-D Series Expansions 737 Appendix-E Schwarz’s Inequality : 738 Appendix-F Fourier Transform Relations 740 Appendix -G_Error Function 742 Appendix-H_Furictions 743 Appendix -1__ Probability Density Functions iv)ac AM ANSI Asc ASK AD AWGN BER BPF BSC BW BPSK BFSK ccrrt cw CDM CDMA CDF de DFT DM DPCM DPSK DSB Abbreviations Alternating cycle voltage or current Amplitude Modulation American National Standards Institute American Standard Code for Information Interchange. Adaptive Delta Modulation Amplitude Shift Keying ‘Analog to Digital Converter Additive White Gaussian Noise Bit Error Rate Bandpass Filter Binary Symmetric Channel Bondwidth Binary Phase Shift Keying Binary Frequency Shift Keying Consultative Committee for International Telegraph & Telephone Continuous Wave Code Division Multiplexing Code Division Multiple Access Cumulative Distribution Function Decibel direct current or voltage Discrete Fourier Transform Delta Modulation Differential Pulse Code Modulation Differential Phase Shift Keying Double Sided Modulation error function ‘Complementary error function exponential (¢) energy spectral density Frequency Modulation Frequency Division Multiplexing Frequency Division Multiple Access Fast Fourier Transform Frequency Shift Keying ov)Modem MSK NRZ PAM PDM PM PDF psd PSK PLL PPM PWM QAM QPSK RE S or sec SNR zla STFT TDM TDMA vco vec Fourier Transform Hertz High Frequency Inter Symbol Interference Integrated Services Digital Network Inverse Fourier Transform Low-Pass Filter natural logarithm / log to the base e’. Jog to the base 10. millisecond microsecond Modulator-demodulator Minimum Shift Keying Nonreturn to Zero On-off Keying Pulse Amplitude Modulation Pulse Duration Modulation Phase Modulation Probability Density Function power spectral Density Phase Shift Keying Phase Locked Loop Pulse Position Modulation Pulse Width Modulation Quadrature Amplitude Modulation Quadrature Phase Shift Keying Radio Frequency Return to Zero root mean square Second Signal to Noise Ratio Signal to Noise Ratio Short Time Fourier Transform Time Division Multiplexing Time Division Multiple Access Voltage Controlled Oscillator Voltage Controlled Clock (Gvi)I o xe xO *y@ arg () Rel | Im[ | Fix FTX IFT [X (9) @ a {a} x Et] P, or P(e} zor P (x/y) "C, or () or i c R, Ry (Dor Ra, (1) fy @) fy y) P(X <») Fy m, M% oF vo so Symbols Magnitude of the complex quantity contained within Fourier transform pair Time average of x(t) * denotes complex conjugate function. Here X * is complex conjugate of X. Convolution of x (t) and y (t) Phase angle of the complex quantity contained within Real part of quantity contained within Imaginary part of the quantity contained within Fourier transform of x () Inverse fourier transform of X (f) Inverse fourier transform of X (f) Modulo - 2 Addition i.e. Exclusive-OR operation Symbol * represents that 4 (1) is the reconstructed value of x (t) in receiver. Represents sequence whose k“" value is a,. Sum over index n. Expected value or mean value of the random variable contained within. Probability of error of symbol or bit Approximately equals to. Conditional probability of x given y. Binomial coefficient (o- iti! The fourier series coefficients Autocorrelation or crosscorrelation function of the signal. Crosscorrelation function of the signal Probahility density function of a continuous random variable at X = x. Probability density function of random variables X & Y at X =x & Y=y Probability of a random variable X for X < x Cumulative Distribution function of a random variable X at X < x Mean or average value of a random variable X. Standard deviation of a random variable X. Variance of a random variable X. Energy spectral density of a signal x (1). Power spectral density of a signal x (t). Govt)5 ow) ¥@ 3) rect (0) sgn (0) sinc (t) G@ x@ Mp) Ce) Cc R I H Impulse response of the linear system Transfer function of the linear system ‘Sampling frequency of the signal ‘Sampling period ( Duration of one bit Transmission channel bandwidth in digital transmission. Signaling rate in digital transmission. It indicates bit rate or code rate. Number of binary bits used for encoding a sample value. Number of digital levels used to encode a sample value. Step size of the quantizer used in Digital Modulation Methods i.e, PCM, DPCM, DM, ADM etc, Symbol power Average interference power Quantization error in digital modulation methods Coherent / noncoherent carrier & reference’ signal used in digital passband transmission Unit step function at t= 0 Delta function at t= 0 Rectangular function Signum function Sine function Generator polynomial in cyclic codes Code vector polynomial Message bits polynomial Check bits polynomial Channel capacity Information rate Amount of Information Entropy or average information per message. (xvity1 Introduction | We are going to start the study of Communication Systems. Before that, we will revise some of the important concepts of communication. What is Communication ? Communication is the process of establishing connection (or tink) between too points for information exchange. In our day to day life we come across many ways of communication. For communication to take place, three essential things must be present ie., 1) Sender or Transmitter : It sends information. For example TV transmitting station or radio transmitting stations are ~senders’, since they transmit information. 2) Receiver : It receives information. For example all TV sets and radios are receivers. They get information from transmitter. 3) Communication Channel : This is the path through which the signal propagates from transmitter to receiver. Fig. 1.1 shows an example of communication system. ‘Communication channel ‘Sender or Transmitter J f Receiver Transmitter Fig. 1.1 An example of communication system Now let us consider different examples of communication that we use in our day to day life. a)Communication Systems - | 2 Introduction 1) Telephone : All of us use telephones now a days. We know that we talk and listen through a telephone. The telephone set is connected to telephone exchange through wires. The other telephones are connected in the same way. Transmitter / Receiver or Talker / Listener Transmitter / Receiver or Talker / Listener Telephone exchange Wires connecting the teleph ‘Communication channel is the wire joining two ssibscribes 1 the. telephone subscribers through exchange telephone exchange Fig. 1.2 Communication on telephones; Line communication through wire As shown in Fig. 1.2 the transmission is bidirectional. Every telephone subscriber can talk as well as listen. Talking is equivalent to transmission of data (voice) and listening is equivalent to receiving of data. The data (ie. voice signal) is passed through wires. Thus the communication channel contains wires which are routed through telephone exchange. 2) Mobile phones (Cellular phones): As we know, the mobile phones are becoming very much popular now a days. In the mobile phones also we can talk and listen. Thus mobile subscribers are essentially talkers (i.e transmitters) and listeners, (i.e. receivers). Signate propagate through air Talker / listener Talker /listener ‘transmitter | receiver 5 L A 5 transmitter / receiver Mobile phone Mobile phone subscriber Base station Base station supsoriber antenna antenna ‘Communication channel does not contain any connection between two mobile subscribers. But the signals propagate through air. This is also called wireless communication, Fig. 1.3 Mobile communication. An example of wireless communicationCommunication Systems - | 3 Introduction As shown in the Fig.1.3, the mobile phone subscribers communicate to the base stations in their cell. Therefore two mobile subscribers communicate with each other through the base stations. Base station is similar to telephone exchange which work in particular area. But there are no wires joining the mobile phone subscribers and base stations. Thus the voice signals pass through air. Air acts as the conducting medium. This is also called wireless communication. The communication 0 channel is the air (ie. wireless). 1.1 Basic Communication Systems From the examples of communication discussed so far, now we can prepare the model of basic communication system. We know that for any communication to take place, three things are essential. They are, 1) Transmitter (Talker) 2) Receiver (listener) and 3) Channel or transmission medium. The above three things are present in any communication system. Whenever two people are talking with each other, it becomes a communication system. In that case the voice signal is exchanged and the transmission medium is air. Fig. 1.1.1 shows the block diagram of basic communication system. Transmission] iis Source Giaaial Rociver Destination Noise, interference and distortion Fig. 1.1.1 Block diagram of basic communication system As shown in the figure, the source generates the message to be transmitted. The transmitter or talker sends that message over the transmission channel. The transmission channel can be the medium such as electric conductors, air. or light (in case of optical fibré), The receiver or listener receives the message from transmission channel. It is then given to the destination by the receiver. During the transmission over the channel, the message is distorted and it becomes noisy. What is the necessity of transmitter and receiver? Why source does not speak to destination directly? This question is obvious. Source cannot speak to destination directly. For example consider a very simple example of Television System. The programs are generated by shooting the objects through camera and they are stored on-the video tape. Thus the program to be broadcasted is stored on the video tape initially. This means the source of information is “Video tape”. The people watch this program on television. ThisCommunication Systems - | 4 Introduction means the destination of the program is the ‘people’. But the people (Destination) cannot watch the program recorded on the video tape (Source) directly. The tape is broadcasted on some TV channel by the TV broadcasting station or satellite. The Television set then receives this channel and people can watch the program on their TV set. Thus for the signal to be transmitted over long distance, the TV broadcasting station is essential. And to receive the signals which are present in the air, TV set is essential. In this communication system, the TV broadcasting station is the transmitter and TV set is the receiver. Thus we can say that, The transmitter is required to make the signal suitable for conduction over the channel. And the receiver is required to convert the signal from transmission channel and make suitable for destination. 1.2 Modulation and Demodulation in Communication System Before discussing about modulation and demodulation directly, we will consider an interesting example first. Let us consider that you want to send some “parcel” from Pune to Nasik. It is obvious that the parcel cannot move from Pune to Nasik on its own. What all you need to do is, to contact some courier company. The courier company will send the parcel by some Pune-Nasik “Bus”. The courier agent at Nasik will recognize the ‘Bus’ and receive the parcel. He will then deliver the parcel to the concerned person.. This example basically represents the communication system. The parcel becomes an information to be transmitted. The courier agency office at Pune is the transmitter and its office at Nasik is the receiver. The parcel is sent through the ‘Bus’. This ‘Bus’ is the carrier which carries the parcel (ie. message). Fig. 1.2.1 shows this illustration in diagram form. . Carrier of the parcel i.e. "Bus" ‘Source or End user or ‘Sender Transmitter Receiver Destination — —+ —+agency} — Communication channel from Pune to Nasik i.e. "Road" Fig. 1.2.1 An example representing communication system As shown in Fig. 1.2.1, the parcel rides over the bus and is transmitted from source to the destination. Similarly in electrical communication, the message signal is converted to an electrical signal and given to the transmitter. The transmitter allows the signal to ride over some electrical carrier. At the receiver, the information signal isCommunication Systems - | 5 Introduction separated from the carrier. The information signal is then converted to the required format. We will now introduce modulation and demodulation. Modulation : Whenever the message or information signal rides over the carrier, it is called modulation. Consider for example a sinusoidal carrier signal. Then message signal can ride over this carrier (sine wave) in three ways - 1) It can ride over the amplitude of the carrier. 2) It can ride over the phase of the carrier. or 3) If can ride over the frequency of the carrier. In electrical sense, the operation of riding over the amplitude of carrier means to alter (or modulate) the amplitude of carrier. This is called Amplitude Modulation of the carrier. Thus the message signal becomes the modulating signal and it is transmitted by variations in the amplitude of the carrier. Similarly when message (or modulating) signal rides over the phase of the carrier it is called Phase Modulation. ‘And in case frequency Modulation, the frequency of the carrier is altered by message ig. 12.2 shows an example of Amplitude modulation. (See Fig. Modulation is the process of placing the message signal over some carrier to make it suitable for transmission over long distance. The carrier signal is basically of higher frequency than that of message. And it has the ability to travel over long distance. We will come back to the discussion of modulation later on. Now we will first see what is demodulation. Demodulation : The modulated carrier signal is transmitted by the transmitter. The receiver receives this carrier. It separates the message signal from the modulated carrier. This process is exactly 0} ite to that of modulation at the transmitter. Demodulation is the process of separating message signal from the modulated carrier signal. Continuous Wave (CW) Modulation : The sinusoidal carrier modulation discussed above is called Continuous Wave or CW modulation. The CW modulation is used by most long distance communication. The CW modulation produces frequency translation. We will not go into much details of this concept here. For example consider the case of Amplitude modulation. The message signal is of frequency 10 kHz and the carrier is of frequency 50 kHz. Then mathematically we can prove that the amplitude modulated signal will have carrier of 50 kHz, and two sidebands of (50 - 10) = 40 kHz and (50 + 10) = 60 kHz. These two side bands carry the message signal actually. But at the same time their frequency is much higher than that of message. This is nothing but translation of message frequency to high frequency side.Communication Systems - | 6 Introduction cartier is higher than that of message signal Amplitude modulated 4 signal Fig. 1.2.2 a) The message signal to be sent to destination b) The carrier signal generated by the transmitter c) Message signal rides over the amplitude of carrier signal. That is the carrier is Amplitude modulated by the message 1.3 Electromagnetic Waves : The Carriers of Electric Signals We know that electromagnetic waves have two fields. One field is electric field and other is magnetic~at 90°. The direction of propagation of those waves is perpendicular to both electric and magnetic fields. The electromagnetic waves travel at the speed of light (3 x10° m/sec). They can travel on electric conductors, in air, in light or in vacuum. Therefore high frequency electromagnetic waves are used as carriers for electric signals. (Actually speaking, electric signals and light are also electromagnetic waves). The electromagnetic spectrum ranges from very very lowCommunication Systems - | 7 Introduction frequencies to very very high frequen spectrum and its applications. cies. Fig. 1.3.1 shows the electromagnetic Frequency band Use Ultraviolet 10° He Experimental Wideband data 10m a 10" Hz Millimeter waves (EHF) Experimental 100 GHz 1 Navigation om Satelite to satelite ‘Super high frequency Microwave relay 10Gtt (SHF) Earth-sataltie z Radar ‘tem Utra high frequency Mobile radio. 6 (UHF 1 GHz 5 am SMF, UHF TV and mobile radio z 2 , ‘Mobile, aeronautical 3 s veo VHF TV and FM broadcast too mn: & 10 Mobile radio “ m High frequency Business (HF) ‘Amateur radio 40 MHz Intemational radio 100 m Citizen's band Mesure | ____AifBreadeast (MF) ‘AM Broadcasi 4 MHz + km ore Aeronautical owe Navigation 100 kHz ) Radio teletype 10 km Very low frequency (VLF) 10 KHz 100 km Teleshone Telegraph 1 kHz Fig. 1.3.1 Electromagnetic spectrum. Fiber optic communication : As shown in Fig 1.3.1 as carrier frequency goes on increasing the transmission medium, propagation mode and applications change. Communication through optical fibers takes place at Ultraviolet, Infrared and Visible light frequenies. This is calledCommunication Systems - | 8 Introduction optical communication and it is used for wide band data. Normally the transmitter or modulator in fibre optic communication system is the light source such as light emitting diode (LED) or Laser beam. The intensity of the light beam is modulated by the message signal. At the receiver, the photodiode (ie. detector) detects the varying light intensity and generates the message signal. The fiber optic communication has advantages such as wide bandwidth, low signal attenuation and reduced noise interference. Wireline Communication : The telephone network which we are using presently is an example of wireline communication. Normally wire pairs or coaxial cables are used for this purpose. The signals in audio, very low and low frequency range are carried on wire pair. The bandwidth of wire pairs is limited upto 300 kHz. The coaxial cables give comparatively larger bandwidth ie. upto 14Hz. The cable television which we see today is on the coaxial cable only, and they use TV channels in the UHF and VHF range. Fig. 1.3.2 illustrates the frequency ranges of wireline, wireless and fiber optic communications. Optica Laser ‘fers beams 100 Giz s0GHe 1GH 100 MHz, Frequency 10 MH MHz 100 kite t0KHz Fig. 1.3.2 Electromagnetic spectrum. Propagation modes and transmission mediaCommunication Systems - | 9 Introduction Wireless communication : As the name indicates, there is no wire or any such medium for conduction of electromagnetic waves. Wireless communication takes place through air or vacuum. The frequencies right from 10 kHz upto 100 GHz are transmitted using wireless communication. Electromagnetic waves can be transmitted either by ground wave propagation, sky wave propagation or line of sight propagation depending upon the frequency. Fig. 1.3.2 shows different frequency. ranges and their modes of propagation. All the radio, TV and satellite broadcasting is wireless communication. In wireless communication, the electromagnetic waves are transmitted and received with the help of radiators or antennas. The size of such radiations or antennas depend upon the frequency being transmitted/received. Normally for efficient radiation of the electromagnetic energy, the length of the antenna must be at least \ of the wavelength of frequency being transmitted. For example consider the case of AM radio station transmitting IMHz frequency. The wavelength of this frequency is, velocity of light (c) le Le wavelength ay _ 3x08 1 «10° = 300m Length of the transmitting antenna = a sen = 30 m at least. The size of the antenna will reduce if the frequency of transmission is increased. Because of this reason, wireless communication is not used for low frequencies. The advantages of wireless communications are cost effectiveness, possible long distance communication and simplicity. But the drawbacks are noise interference, environmental effects and attenuation of the signal. Wireline communication and wireless communication can both be used for some frequency range. The choice is made upon the required distance and coverage area. 1.4 Analog Communication With the introductory discussion of basic communication systems in previous sections, we will now tum towards the classification of analog and digital communication. In case of analog communication, the message signal to be transmitted is analog. This analog message can be obtained from sources such as speech, video shooting etc. The analog signal varies smoothly and continuously with time. The message signal is then modulated on some carrier frequency by the modulator. The amplifier then gives this signal to the transmitting antenna. Fig, 1.4.1 shows the basic, block diagram of analog communication system.Communication Systems - 10 Introduction Transmitter Modulator Carrier Noise Receiver Demodulator} Analog message Destination signal Fig. 1.4.1 Basic analog communication system Presently all the AM, FM radio transmission and TV transmission is analog communication. The analog communication needs lower bandwidth compared to digital communication as we will see. But the effect of noise interference is more in case of analog communication. 4.5 Digital Communication Fig, 1.5.1 shows the basic elements of a digital communication system. The source and the destination are the two physically separate points. When the signal travels in the communication channel, noise interferes with it. Because of this-interference, the smeared or disturbed version of the input signal is received at the receiver. Therefore the signal received may not be correct. That is errors are introduced in the received signal. Thus the effects of disturbance and noise due to the communication channel limit the rate at which signal can be transmitted, The probability of error in the received signal and transmission rate are normally used as performance measures of the digital communication system. infrtion source Channel source encoder encoder Noise ‘Source ‘Channel pon Demodulator Fig. 1.5.1 Basic digital communication system The information source generates the message signal to be transmitted. In case of analog communication, the information source is analog. In case of digital communication, the information source produces a message signal which is not continuously varying with time. Rather the message signal is intermittent with respect to time. The examples of discrete information sources are data from computers,Communication Systems - | 1 Introduction teletype etc. Even the message containing text is also discrete. The analog signal can be transmitted to discrete signal by sampling and quantization. In sampling, the analog signal is chopped off at regular time intervals. Those chopped samples form a discrete signal. The discrete information sources have following important parameters : a) Source alphabet : These are the letters, digits or special characters available from the information source. b) Symbol rate : It is the rate at which the information source generates source alphabets. It is normally represented in symbols/sec unit. c) Source alphabet probabilities : Each source alphabet from the source has independent occurance rate in the sequence. For example,letters A, E, I etc. occur frequently in the sequence. Thus probability of the occurrence of each source alphabet can become one of the important property which is useful in digital communication. d) Probabilistic dependence of symbols in a sequence : The information carrying capacity of each source alphabet is different in a particular sequence. This parameter defines average information content of the symbols. The entropy of a source, refers to the average information content per symbol in long messages. Entropy is defined in terms of bits per symbol. Bit is the abbreviation for binary digit. The source information rate is thus the product of symbol rate and source entropy. i.e. Information rate = Symbol rate x Source entropy (Bits/sec) (Symbols/sec) (Bits /symbol) The information rate represents minimum average data rate required to transmit information from source to the destination. 1.5.1 Source Encoder and Decoder The symbols produced by the information source are given to the source encoder. These symbols cannot be transmitted directly. They are first converted in to digital form (i.e. Binary sequence of 1’s and 0's) by the source encoder. Every binary ‘1’ or ‘0’ is called a bit. The group. of bits is called a codeword. The source encoder assigns codewords to the symbols. For every distinct symibol there is unique codeword. The codeword can be of 4, 8, 16 or 32 bits length. As the number of bits are increased in each codeword. The symbols that can be represented are increased. For example, 8 bits will have 28 = 256 distinct codewords. Therefore 8 bits can be used to represent 256 symbols. 16 bits can represent 2!6 = 65536 symbols and so on. In both of the above examples the bits in every codeword are same throughout. That is 8 in first case and 16 in next case respectively. This is called fixed length coding. Fixed length coding is efficient only if all the symbols occur with equal probabilities in a statistically independent sequence. In the practical situations, the symbols in the sequence are statistically dependent and they have unequal probabilities of occurance. For example, let us assume that the symbol sequence represents the percentage marks of the students. The 02%, 08%, 20%, 98%, 99% ete symbols will have minimum probability of occurrence. But 60%, 55%, 70%, 75% will have more probability. For such symbols normally variable length codewords are assigned. More bits (MoreCommunication Systems - | 12 Introduction length) are assigned to frequently occuring symbols and less bits are assigned to rarely occuring symbols. Typical source encoders are pulse code modulators, Delta modulators, vector quantizers etc. We will come across these codewords in detail in the subsequent chapters. Source encoders have following important parameters. a) Block size : This gives the maximum number of distinct codewords that can be represented by the source encoder. It depends upon number of bits in the codeword. For example, the block size of 8 bits source encoder will be 2°=256 codewords. b) Codeword length: This is the number of bits used to represent each codeword. For example, if 8 bits are assigned to every codeword, then codeword length is 8 bits. c) Average data rate : It is the output bits per second from the source encoder. The source encoder assigns multiple number of bits to every input symbol. Therefore the data rate is normally higher than the symbol rate. For example Let us consider that then symbols are given to the source encoder and the length of codeword is 8 bits. Then the output data rate from the source encoder will be, Data rate = Symbol rate x Codeword length = 10 x8 = 80 bits/sec Information rate is the minimum number of bits per second needed to convey information from source to destination as stated earlier. Therefore optimum data rate is equal to information rate. But because of practical limitations, designing such source encoder is difficult. Hence average data rate is higher than information rate and hence symbol rate also. d) Efficiency of the encoder : This is the ratio of minimum source information rate to the actual output data rate of the source encoder. At the receiver, some decoder is used to perform the reverse operation to that of source encoder: It converts the binary output of the channel decoder into a symbol sequence. Both variable length and fixed length decoders are possible. Some decoders use memory to store codewords. The decoders and encoders can be synchronous\or asynchronous. 1.5.2 Channel Encoder and Decoder At this stage we know that the message or information signal is converted in the form of binary sequence (i.e. 1’s and 0's). The communication channel adds noise and interference to the signal being transmitted Therefore errors are introduced in the binary sequence received at the receiver. Hence errors are also introduced in the symbols generated from these binary codewords. To avoid these errors, channel coding is done. The channel encoder adds some redundant binary bits to the input sequence. These redundant bits are added with some properly defined logic. For example consider that the codeword from the source encoder is of 3-bits in length and the channel encoder adds one bit to the codeword to make it 4-bit long. This 4! bit is added (either 1 or 0) such that numberaa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.Communication Systems - | 4 Introduction 1.5.3 Digital Modulators and Demodulators In the previous sections we studied why modulators and demodulators are required. Whenever the modulating signal is discrete (ie. binary codewords), then digital modulation techniques are used. The carrier signal used by digital modulators is always continuous sinusoidal wave of high frequency. The digital modulator maps the input binary sequence of 1’s and 0's to analog signal waveforms. If one bit at a time is transmitted, then digital modulator signal is s,(t) to transmit binary *0’ and 2s( to teanemix binary °Y', For exariple consider the output of digital modulator shown in Fig, 1.5. soit) s2(l) S(t) Sol) s(t) S(t) Fig. 1.5.2 Frequency modulated output of a digital modulator The signal s)() has low frequency compared to signal s(t). It is frequency modulation (FM) in two steps corresponding to binary symbols °0’ and *1’. Thus even though the modulated signal appears to be continuous, the modulation is discrete (or in steps). Single carrier is converted into two waveforms s,(f) and s(t) because of digital modulation. If the codeword contains two bits and they are to be transmitted at a time, then there will be M =2? =4 distinct symbols (or codewords). These four codewords will require four distinct waveforms for transmission. Such modulators are called M-ary modulators. Frequency shift keying (FSK), phase shift keying (PSK), Amplitude shift keying (ASK), Differential phase shift keying (DPSK), Minimum shift keying (MSK) are the examples of various digital modulators. Since these modulators use continuous carrier wave, they are also called digital CW modulators. In the receiver, the digital demodulator converts the input modulated signal to the sequence of binary bits. The most important parameter for the demodulator is the method of demodulation. The other parameters for the selection of digital modulation method are, a) Probability of symbol or bit error. b) Bandwidth needed to transmit the signal -c) Synchronous or asynchronous method of detection and d) Complexity of implementation.Communication Systems - | 15 Introduction 1.5.4 Communication Channel ‘As we have seen in the preceding sections, the connection between transmitter and receiver is established through communication channel, We have seen that the communication can take place through wirelines, wireless or fibre optic channels. The other media such as optical disks, magnetic tapes and disks etc. can also be called as communication channel, because they can also carry data through them. Every communication channel has got some problems. Following are the common problems associated with the channels : a) Additive noise interference : This noise is generated due to internal solid state devices and resistors etc. used to implement the communication system. b) Signal attenuation : It occurs due to internal resistance of the channel and fading of the signal. ©) Amplitude and phase distortion : The signal is distorted in amplitude and phase because of nonlinear characteristics of the channel. d) Multipath distortion : This distortion occurs mostly in wireless communication channels. Signals comming from different paths tend to interfere with each other. There are two main resources available with the communication channels. These two resources are - a) Channel bandwidth : This is the maximum possible range of frequencies that can be used for transmission. For example, the bandwidth offered by wireline channels is less compared to fibre optic channels. b) Power in the transmitted signal : This is the power that can be put in the signal being transmitted. The effecte of noise can be minimized by increasing the power. But this cannot be increased to very high value because of the equipment and other constraints. For example, the power in the wireline channel is limited because of the cables. The power and bandwidth limit the data rate of the communication channel. As we know, the fiber optic channel transports light signals from one place to another just like a metallic wire carriers an electric signal. There is no current or metallic conductor in optical fiber. The optical fibre has following advantages : a) Very large bandwidths are possible : b) Transmission losses are very small ) Electromagnetic interference is absent d) They have small size and weight e) They offer ruggedness and flexibility £) Optical fibres are low cost and cheap. Satellites essentially perform wireless communication. Mainly satellites are repeaters. Broad area coverage is the main advantage of satellites. The power requirement is also less, since solar energy is used by satellites. Global communicationCommunication Systems - | 16 Introduction is very easily possible due to satellite channels. The interference on satellite channels is present but it is minimum. 1.6 Baseband and Passband Transmission The message signal generated from the information source is also called Baseband Signal. The baseband signal can be a combination of two or more message signals. If the baseband signal is transmitted directly, then it is called baseband transmission. Thus baseband transmission does not use modulators and demodulators. The baseband signal can be both analog as well as digital. The analog baseband signal varies continuously with time and has continuous amplitude. The digital baseband signal is discrete in both time and amplitude. The baseband transmission is preferred at low frequencies and for short distances. Intersymbol Interference (ISI) is the major problem associated with baseband transmission. If modulated signal is transmitted over the channel, it is called bandpass (or passband) transmission. Whenever the modulating signal is impressed upon a carrier, the modulated signal is generated. This modulated signal has fixed band of frequencies around carrier frequency. The nature of such a signal is bandpass type. Hence modulated signals are called passband (or bandpass) signals. For example if kHz modulating signal is impressed upon 100 kHz carrier by using amplitude modulation, then two side bands will be generated. Lower sideband will be at (100 - 1) = 99 kHz and upper sideband will be at (100 + 1) = 101 kHz. Thus the Amplitude modulated signal will have frequencies from 99 to 101 kHz. These frequencies are bandpass type. Hence modulated signals are also passband (or bandpass) type. Bandpass transmission is normally used at high frequencies and for long distances. All television and satellite transmission is passband type. The passband transmission can be both analog and digital. If the digital signal modulates the carrier, then it is called digital passband transmission. We will see baseband and passband (or bandpass) transmission in forth-coming chapters. 1.7 Multiplexing Techniques Multiplexing techniques allow many users to share a common communication channel simultaneously. There are two major types of multiplexing techniques. They are - a) Frequency Division Multiplexing : This technique allots a fixed frequency band to every user in the complete channel bandwidth. Such frequency slot is allotted continuously to that user. For example consider that the channel bandwidth is 1MHz. Let there be ten users, each requiring upto 100kEiz bandwidth. Then the complete channel bandwidth of IMHz can be divided into ten frequency bands; each of 100 kHz and every user can be allotted one independent frequency band. This technique is frequency division multiplexing (FDM). It is mainly used for modulated signals. This is because modulated signal can be placed in any frequency band by just changing the carrierCommunication Systems - | 47 Introduction frequency. At the receiver, these frequency multiplexed signals can be separated by the use of tuned filters (bandpass filters) of their respective frequency band. And for every band, there are independent tuned filters and demodulators. b) Time Division Multiplexing : In case of Time Division Multiplexing (TDM), the complete channel bandwidth is allotted to one user for fixed time slot. For example, if there are ten users, then every user can be given the time slot of one second. thus complete channel can be used by each user for one second time in every ten seconds. This technique is suitable for digital signals. Because digital signals are transmitted intermittently and the time speacing between two successive digital codewords can be utilized by other signals. There is possibility of crosstalk in FDM where as Intersymbol Interference is possible in TDM. These problems can be overcome by some special cares. Exercise What is communication ? Explain with the help of examples, What is the function of communication channel ? Explain basic communication system with the help of block diagram. Why transmitters and receivers are required for communication ? What is the use of modulation and demodulation ? What is wireless and wireline communication ? In which range of electromagnetic waves it is used ? 7. What is fiber optic communication ? What are its advantages over other methods of communication ? 8.. Explain how modulation helps in reducing antenna size in wireless communication with the help of example. 9. What is Digital Communication ? Explain with the help of block diagram. 10. Compare analog and Digital Communication 11. What are the problems associated with communication channel ? 12. What are the resources of communication channels ? 13. What are the advantages and disadvantages of Digital Communication ? 14. What are baseband and passband signals ? How their transmission differs from each other ? 15. What are multiplexing techniques ?. Explain them briefly. aa Pen 0002 Signals and Systems a Communication systems consists of various blocks performing specific functions. For example modulator, amplifiers, detectors, mixers, various transmission channels are blocks. The signal is entered at one end (input) and depending on the transfer characteristic of the block, the signal is available at the output. Every block is treated as a system and has specific transfer function. Communication theory deals with the mathematical models of these systems and the mathematical representation of the signals being applied. Complete mathematical analysis of communication systems and signals can be performed. Signals are electrical quantities like voltage, current or power varying with time. Using mathematical model of the system, the response of the system can be obtained that vary with time. Here time is an independent variable and electrical quantity is dependent variable. This is also called time domain analysis, If the signal is represented in frequency domain, that is variation of electrical quantity with signal frequency, then output system response is also in frequency domain. The signal frequency is independent variable and electrical quantity is dependent variable. ' ‘The transfer function of the communication system is expressed in frequency domain. When dealing with communication system, frequency domain analysis has specific advantages. Because of frequency domain analysis, spectrum of various electrical signals and their bandwidth are directly obtained. The mathematical tools like fourier transforms give conversion between time and frequency domains. Various important quantities like power spectral densities, energy density functions different error probabilities are obtained in frequency and time domains. The signals and systems represented hence forth are mathematical models. In this chapter we will see how fourier series and fourier transforms are used to represent signals in time and frequency domains. Then it becomes easy to estimate the spectrum of these signals. Some signals appear random when observed in time domain. But when they are converted to frequency domain, then they exhibit specific properties which can be easily studied. In communication systems, analysis and effects of noise is as important as that of wanted signal itself. Frequency domain representation of noise give much more information compared to time domain. In the present day every analysis is getting computerised. Advanced softwares like MATLAB and MathCAD are available. The spectral properties of signals can be studied using these software standard functions such as plot, abs. freqs, psd, angle, erf, erfc, corrcoef and xcorr etc. are available directly to calculate magnitude and phase plots, power spectral density, correlation functions etc. A software program (normally called M file) can be written with the help of MATLAB functions to perform some specific analysis. (18)Communication Systems - 1 19 Signals and Systems 2.1 Signals A function of one or more independent variables which contain some information is called signal. In electrical sense the signal can be voltage or current. The voltage and current is the function of time as an independent variable. In our normal life we come across many electric signals such as radio signal, TV signal, telephone signal, computer signals etc. The signals are classified as follows : (Deterministic and random signals. (i) Continuous and discrete or analog and digital signals. (ii) Periodic and non-periodic signals. (iv) Noise signals. (v) Physically realizable and non-tealizable signals. (vi) Energy and power signals. 2.1.1. Deterministic and Random Signals A deterministic signal is a signal about which there is no uncertainty with respect to its value at any time. A deterministic signal can be completely represented by mathematical equation at any time. For example sine wave, cosine wave, triangular wave, exponential pulse etc. are deterministic signals, since they have unique mathematical equations. ie. x(Q = 100sin 50t represents a deterministic signal at any time. A random signal has some degree of uncertainty before it actually occurs. That means random signal cannot be defined by some mathematical expression. Thus the value of random signal is not predefined or it cannot be calculated from previous values of the signal, Fig. 21.1{a) shows an example of deterministic signal and Fig. 2.1.1(b) shows random signal. x(t) Dote istic signal chee Fig. 2.1.1 (a) : Cosine wave is an example of deterministic signal. It can = be defined by mathematical expression. x(t) = Acos atCommunication Systems - | 20 Signals and Systems x(t) Fig. 2.1.1 (b) : Random signal cannot be mathematically t modelled. Its value is unpredictable at any time Random signal Random signals are broadly called as Noise signals. 2.1.2 Continuous and Discrete or Analog and Digital Signals When the amplitude and time of the signal is continuous, it is called continuous or analog signal. In other words continuous signals have some value at every instant of time. Signals arising from converting a physical parameters such as sound, temperature, light intensity etc. result in continuous signals. For example consider that the temperature is converted into an electric signals. We know that temperature never changes in steps, rather the change is continuous and smooth. Therefore an equivalent electric signal will also be continuous and smooth. When the signal is represented only at discrete instants of time, it is called Discrete signal. For example, if we record the temperature only at one hour intervals, then we get discrete time representation of temperature. Fig. 2.1.2 (a) shows an example of continuous signal and Fig. 2.1.2(b) shows its equivalent discrete time signal. Fig. 2.1.2 (a) : Analog ‘or continuous signal : This signal is continuous in amplitude and time. tt can be represented at any time instant. x(t) ‘Samples take continuous Fig, 2.1.2 (b) : Discrete values of amplitude time signal : This eos signal has values only at fixed time intervals. The amptitude values may be continuous and discontinuous pneeniiereena-oo-cediannen=)Communication Systems 2 Signals and Systems The discrete signal of Fig. 2.1.2 (b) appears to be a sampled version of continuous signal of Fig. 2.1.2 (a). When the signal is discrete in both time as well as amplitude, it is called digital signal. This means the amplitudes of digital signals are represented only at some fixed number of values. The time representation is also discrete. Fig. 2.1.3 shows the signal of Fig.2.1.2 (b) converted to discrete amplitude levels. As shown in Fig. 2.13, the complete range of 1 volt is represented into 5 discrete levels. Each sample amplitude is represented by the nearest discrete amplitude level. This process is called quantization. Thus there is sometimes difference in actual sample amplitude and quantized (or digital) sample amplitude. Almost all the digital signals are obtained from analog signals. The analog signal is first sampled at regular intervals. This sampled signal is then quantized to give digital signal. The digital signal is further coded into binary, hex or octal number representation for further processing. This process of obtaining digital signal from analog signal, is called Analog to Digital conversion. The signals handled by computers are all digital. The sample is quantized xq('T) to the nearest voltage level of 1.0 votts| Fig. 2.1.3 : An example of digital signal. The signal is sampled at multiples of ‘T instants, The amplitude is represented in 5 levels. The sample is quantized t to nearest amplitude level. Discrete amplitudes Discrete time instants 2.1.3. Periodic and Non-periodic Signals A signal is said to be periodic, if it repeats after a fixed time period. This can be defined by mathematical expression as, x() = x(t +7) ww (2.1.1) Here T, > 0; is called the period of the signal x(t). Fig. 2.1.4 (a) shows a waveform of sine wave. This waveform can be represented by the following mathematical equation i.e., x(@) = A sin(2nfy ) ww (2.1.2) A Here f, = z = frequency of the signal. 0 The signal x(¢ +T,) will be, x(¢+Tp) = A sin[2nf, ¢ +T,)] =A wna al since fy = foCommunication Systems - | 22 Signals and Systems A sin[2nf, t +2r] = A sin(2nf,f) since sin (8 +2n) = sind by trigonometric relation. " x(t) which is same as equation 2.1.2 A signal is said to be non-periodic, if it does not repeat. Some times non-periodic signals are said to have a period T, equal to infinity. Fig, 2.1.4 (b) shows an example of non-periodic signal. This is decaying exponential pulse. Consider, -at + Ty) xt +T,) ze = hte) -at zene = 0 which is not equal to x(t) This shows that the signal having period.T, = © is nothing but non-periodic signal. x(t) x(t) = Asin (2nfyt) Fig. 2.1.4 (a) : An example ‘ of periodic signal. The : period of this signal is To. y The signal repeats at multiples of Ty , Le. Ty, 27g, 37g, ° Fig. 2.1.4 (b) : An example of non-periodic signal. The signal does not repeat (I.e. Ty =~) but can be represented by mathematical expression, 2.2 Energy and Power of the Signals The signals are classified as energy signals and power signals. Before defining these classification, we will first define energy and power concepts. 1, Average value (Time mean or DC value) The average value of the signal x(t) over all the time is defined as,aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.Communication Systems - 1 2 Signals and Systems In the above equation, x"(0) is integrated from - 74107 and limit is taken as T -+ ». This makes the upper and lower limits approaching infinity. Thus equation 2.2.11 and equation 2.2.10 have the same meaning. 5. Energy signals The signal x(t) is said to be Energy Signal if and only if the total normalized energy is finite and nonzero (ie. 0 < E < ©). ie, Almost all the practical nonperiodic signals which are defined over finite time (also called time limited) are Energy Signals. 6. Power of the energy signal Now let us see what will be the normalized average power of the energy signal consider that the signal x(?) has finite energy “E’ and hence it is an energy signal. Then by equation 2.2.6 we can calculate the normalized average power of this signal as, lim aT laa ri " Pe 1 fT Seat Here x(() is energy signal. Ta0T dy am lim 1 Jim 77? 2 Wy at “T30 T Too J xo -T/2 lim 1 @ Log Tox T from equation 2.2.11, since energy is finite. =0 Byletting T + © This shows that, 7. Energy of the power signal Now let us consider that xf) is a power signal which has finite power. Then normalized energy of x(t) can be obtained from equation 22.11 as, lim T/2 2 Es J x Qadt Here x(t) is the power signal. Toe 4) ; Piz = lim Jp JP @at [By rearranging the equation. Tox| T 4, ’ : i 12 tin pm TP ey T30° ToT J,Communication Systems - | 26 Signals and Systems = _1™ F.p From equation 2.26, since power is finite Toe =o By letting T + This shows that, he energy of the power signal is infinite over infinite time. Table 2.2.1 shows the comparison of Energy and Power Signals. Table 2.2.1 Comparison of energy and power signal Power signal . Energy sigral The normalized average power is finite and . | Total normalized energy is finite and nonzero. nonzero. The power is obiained by Total normaized energy is obtained by im 172 Ti ee LT Rae inf Aa To2T tp 72 ty Practical periodic signals are power signals. ._|_Nonperiodic signals are energy signals. ‘These signals can exist over infinite time. ._| These signals ate time imited. Energy of the power signal is infinite over Power of the enargy signal is zero over infinite time. infinte time. For example a pertodic pulse train. . | For example a single rectangular pulse x x(t) Ex.2.2.1: Determine whether the following signals are energy signals or power signals and evaluate their normalized energy or power. i) x(t) = rect (<) q% t ii) x(2) = rect [z) COS yt 1 ‘it a iii) x8) = cos? wy! Sol. : Let us first define rect (<) 0Communication Systems - | ar Signals and Systems T, T, =1 For -—2 to oI The energy of the given signal is T, ; which is finite and nonzero. Hence this is energy signal. j ii) The given signal is xt) = rect (<] cos, t. Here let us assume that there ° T , T, are integer number of cycles of cosine wave from = to > It is shown in Fig. 2.2.2. The total normalized energy of this signal can be obtained from equation 2.2.10 ie., 2 Ty/2 fod = f[ [-coso, Hat -2 ~Ty/2 ECommunication Systems - 1 28 Signals and Systems Ty/2 = f [+Leos2og|a -ty/2 b2 2 0 To/2 Ty/2 = J fated J coszagtat -Ty/2 2 ~T/2 T Ty /2 = 240 since [ cos2aptdt 2 ~Ty/2 is the integration of “Full cycles” of cosine wave, which is zero. Thus the energy of the given signal is finite and nonzero, hence it is energy signal. x(t) = rect (Joos et Fig. 2.2.2 The function x(t) appears to be multiplication of rectangular pulse and cosine wave. There are t_ Integer number of full cycles of cosine wave in the period -70 to 70 2 #0 2 iii) The given function is x(t) = cos” wgt, which is periodic cosine wave. We can say that this is power signal. Let us prove it. From equation 2.2.6 the average normalized power of x(!) is given as, li 1/2 P= i ; xq) dt as 7 ‘ 2 jim o, =; et J [e087 a at 0 0 -19/2 lim 7 7? , = = J cost aot at T > @T, =Ty/2 lim 1 a 1 = = 5(3+ 4 cos 2w ot + cos4upi]dt Ty 2Ty yr 8Communication Systems - | 2» Signals and Systems j Ty/2 : Tl? ‘ Ty/2 lim 9 lim lim 7 2 J Gate “F cosaugt dt J cosdogt at “7, +0T, The $ 7? 0 agai * 0 ~T9/2 ~Ty/2 ee This term will be zero. This term willbe also zero since it is integration of since iis integration of cosine wave over "full cycles" cosine wave over "full cycles” lim 3.7, +0+0 “1, 0 = ‘8 This proves that the power of the given signal is finite and nonzero, hence it is power signal. Ex.2.2.2: The two signals x(t) and y(t) have energies E, and E, respectively. Under what condition energy of x(t) + y(t) will be equal to E, +E,”? INov.-97, 8-Marks] Sol. : Here we have, Energy of x(t) = fe (tat and energy of y(t) = Ivo Then energy of the sum of x(t) ana y(t) will be, feorvore = f[%o+o+20M0 Jat feo + j y(Oat s2fanpat E, +E, +2 foe at If third term in above equation is zero, then energy of x(t) + y(t) will be equal to E, +E, ie, x ty JO-vOat = 0 Above equation will be true if x(t) and y(t) are orthogonal to each other. Exercise 1. Define average value or time mean of the signal. 2. Define average normalized power of the signal. 3. What is normalization ? How it is advantageous in mathematical modelling ?Communication Systems - | 30 Signals and Systems 4. Define power and energy signals and differentiate between them. 5. What is total normalized energy of the signal ? Unsolved Examples 1. _ Determine whether the following signals are energy signals or power signals and why ? i) ae” {Ans. : Energy signal] i) x)= A+ for -1(¢) =2n+, Ty is one cycle period of x() or fy =Communication Systems - | 31 Signals and Systems. The sine and cosine waves can be represented by a complex exponential phasor. From Euler's theorem we know that, e*!° = cos 0 +] sin 0 ww (2.3.2) Here cos® = Real part of e/° and sin® = Imaginary part of e/° If we want to represent cosine wave of equation 2.3.1 by complex exponential phasor, then with @ = agt + we have, A cos(ogt +) = A -Refe! a! * ] *Re’ means real pargof = RefAe!* ef °0'] un (2.33) Fig, 2.3.1 shows the line spectrum representing above equation. Fig. 2.3.1 (a) : Amplitude or magnitude spectrum of x(t). It shows that x(t) contains ‘one frequency component at f, and its amplitude is A. Amplitude Fig. 2.3.1 (b) Phase spectrum of x(t). It shows that the phase shift of frequency component f, is The line spectrum shown in above figure are very useful in frequency domain analysis. It is compact representation of the signal in pictorial form. Normally following conventions are followed when dealing with line spectrums. 1. The phase angles are measured with respect to cosine waves most of the times. For example if line spectrum sine wave is required, then it is first converted to cosine as follows. sin (wf) = cos(wt — 90°) we (2.3.4)Communication Systems - | 32 Signals and Systems 2. The negative amplitudes can be converted to positive value by phase changes. ie. consider the following example, -A coswt = A cos(t £180) o» (23.5) In the above equation, + 180 or — 180° phase shift have same effect. Ex. 2.3.1: Plot the line spectrum of the signal, x(t) = 7 10 cos (40 xt — 60°) + 4sin(120 nt) Sol. : In the above equation first term represents de or zero frequency and the last term need to be converted to cosine wave. Then the equation will be, x) = 7cos(2n04) + 10 cos(40mt— 60° 180°) + 4 cos(120nt ~ 90°) Cosine wave with zero Negative tmplitude is Sine wave is converted t0 frequency is written here cosine wave by phase shift punt by y ihare a shift = Tcos(2r0t) + 10cos(40°xt -60°+180°) + 4cos(120nt - 90°)... (2.3.6) f,=0Hz a= 60HE = 0° 83 =~ 9° In the above equation there is de component of amplitude 7 and other two frequencies. Their frequencies and phase shifts are as shown above Fig. 2.3.2 shows the line spectrum. Fig, 2.3.2 (a) : Amplitude spectrum All amplitudes are positive. This Is single sided ‘spectrum. Amplitude Phase Fig. 2.3.2 (b) Phase spectrum phase can be altered to make araplitudes positive. The magnitude and phase spectrums as shown above are drawn only for positive frequencies.Communication Systems - 33 Signals and Systems Ex. 2.3.2 :-Plot the various possible spectra of the signal x(0) = Scos(2m10¢ + 30°)-10 cos(2n201 + 60°) Sol. : Since both the terms in the above equation are of same type (ie. cosine), we can plot the spectrum of the signal directly. 1) aff) = Scos(2n10t + 30°) -10cos(2n20t + 60°) (23.7) eye A= fr=io = 30° 4 = 60" Fig, 2.3.3 shows the amplitude and phase spectrum represented by above equation 3 Fig. 2.3.3 (a) : Amplitude E spectrum of equation 2.3.7 i a Fig. 2.3.3 (b) : Phase ‘spectrum of equation 2.3.7 2) Now let us write equation 2.3.7 as Mf) = 5cos(2n10F + 30°) + 10cos(2n20t + 60 ~ 180°) cos(2n10# + 30°) + 10.cos(2m20t - 120°)... (2.3.8) ee SS ee ae3 ‘Negative ainplitude is f=lo absorbed in phase shift ‘ast 6, = 30"Communication Systems - | 4 Signals and Systems Fig. 23.4 shows the amplitude and phase spectrum of the above equation. Fig. 2.3.4 (a) : Amplitude spectrum of equation 2.3.8 ‘Amplitude Fig. 2.3.4 (b) : Phase spectrum of equation 2.3.8 Phase Still other spectrums are possible. Reader can verify these. 2.34 Double Sided Spectrums So far we have discussed line spectrums which were functions of only positive frequencies. In this subsection we will study the method to represent spectrums of positive as well as negative frequencies. These are called double sided spectrums. This type of representation is very useful in mathematical analysis. From Euler's theorem we know that, z =e)? =cos0+j sind w= (23.9) The complex conjugate of the above phasor z, is obtained by changing sign of imaginary part j.e., Complex conjugate of z:2* = e !° = cos@—j sin® (23.10) From above two equations (equation 2.3.9 and equation 23.10) we can write, z 42° = 2cos0 =2-Re[z] + cos® is Real part of z and z* both. Re[z] ws (2.3.11) Now consider the equation, a) = A cos(agt + 6) Lett z= A ef(vot*® (2.3.12)Communication Systems - | Kb) Signals and Systems then zt = Ae hols (2.3.13) Putting those values in equation 2.3.11 we get, ie : =A ellot "| 7 [- pilots 0, 4 iloot + | 2 Re (a gilt + “| CA gilts O 1, A Silos e aaa $ : 23. From equation 2.3.3 earlier we know that Re [4 e/°0'* 9) = A cos(a,t +9) then the above equation becomes, A exe ad} -4 pilots A grilogt + 9) aA sleet ih pA cl totarls vw (2.3.15) 2 2 Amplitude: 4 Amplitute=4 frequency =— fo Prase== 6 Thus the above equation represents a magnitude and phase spectrums which has frequencies of f, and -f,. At f,, amplitude and phase both are positive and at ~f, , amplitude is positive but phase is negative. The spectrum is shown in Fig. 23.5. Amplitude Fig. 2.3.5 (a) : Double sided magnitude spectrum. It shows two components of amplitude and both have same polarity Phase Fig. 2.3.5 (b) : Double sided phase spectrum. It shows two t components of phase and they have opposite polarities. dh -nan----==--=- ht The above figure and Fig, 23.1 shows the line spectrums of x() = A 05 (a, t + 4).Communication Systems 36 Signals and Systems From thesé two spectrums we can draw following important conclusions - 1. In single sided amplitude spectrum there is only one component at f,, and its amplitude is A. In double sided amplitude spectrum there are two components at f, and - f,, and their amplitudes are 44. Thus the amplitude in double sided spectrum reduces by half but its polarity is unchanged 2. In single sided phase spectrum there is only one component at f, and its phase is . In double sided phase spectrum there is one component at f, with phase $ , and second component at - f, with phase -}. Thus the phase shift remains same in double sided spectrum but their signs are opposite. 3. The amplitude spectrum has even symmetry whereas phase spectrum has odd symmetry in double sided spectrums. Ex. 2.3.3. Draw the double sided line spectrum of the signal given in example 2.3.1. Sol. : The single sided spectrum is represented by Fig. 2.3.2 and it is given by equation 2.3.6. The equation is reproduced below. xt) = 7cos (2n0#) + 10cos (2n20t+ 120°) + 4 cos (2n60t - 90°) From equation 2.3.15 we know that, A cos(w, t +4) = Zeleo! elt 4 Aid el* Applying this result to equation 2.3.16 we obtain, a= 7Zcos(2n 0) + Wel?R2F i120", 10 y-j2n201 5-127 ———- 2 Z “This is de signal therefore itis unchanged Saat ei 2m 60 =] 90°, 4 e128 608 7 0 = 7oos(2n Ot) + 5 ef28201 yf 120", g yf 2n201 ,-j 120° dry) oe, w+ (23.16)Communication Systems - | 7 Signals and Systems Fig. 2.3.6 shows the double sided amplitude and phase spectrum represented by above equation. Fig, 2.3.6 (a) : Double sided amplitude spectrum of Fig. 2.3.2 (a). The amplitude except at de are reduced to half. Even spectrum. Fig. 2.3.6 (b) : Double sided phase spectrum of Fig. 2.3.2 (b). The phase shift are same but the spectrum Is ‘odd symmetric. Exercise 1. Explain what are phase and amplitude spectrums. 2. Explain how sinusoidal signals can be represented with the help of complex exponential phasor e*!®, 3. What is the advantage of phasor notation ? 4, What are single sided and double sided spectrums ? Compare them. 5. Explain how positive and negative amplitude signs are changed by changing phase of the signal. 6. How complex conjugate phasors produce double sided spectrums ? 7. Explain how the frequencies becomes positive as well as negative in complex conjugate phasors representation with the help of example. 8. What is the advantage of double sided spectrum ? Unsolved Examples 1. Construct one sided and two sided spectrum of at) = —3-4 sin 30 nt [Hint : First convert all the terms in cosine terms by phase changes, Then the equation will be, 2(f) = 3cos (2m 0f +180°) + 4 cos(2n15t+90°)] 2. Represent the signal x{f) = 12sin(@, t-25°) where w, = 20007 in the double and single sided from construct the respective spectrums.Communication Systems 38 Signals and Systems 2.4 Systems A. system is a set of elements or functional blocks that are connected together and produces an System output in response to an input signal. The response of the system depends on transfer function of the system. The functional relationship between input and output can be written mathematically as, Fig. 2.4.1 Block diagram of a system Wig) = fe 0 -1fort<0 a (2.5.3) Fig. 2.5.5 shows the signum function. It is positive (ie. + 1) for positive time and negative (ie. -1) for negative time. The signum function is odd function. Fig. 2.5.1 Signum function 3. Rectangular pulse x The rectangular pulse centered at t = 0 represents the double sided 1 frequency response of low pass filter. Fig. 2.5.2 shows the rectangular pulse of unit amplitude and unit duration. The -2- 0 12 pulse is centered around t = 0. Fig. 2.5.2 Rectangular pulse of unit amplitude and unit duration The rectangular pulse is represented by rect (1) abbreviation. i.e.,Communication Systems - 1 42 Signals and Systems rrect(t) = 1 for 4 teh Rectangular pulse (Unit duration and unit amplitude) = 0 elsewhere os 05.4) Representation of rectangular pulse of width ~“I’ and amplitude ~A’ The pulse is shown in Fig, 2.5.3. This pulse can be represented using definition of rect (f) function given in equation 25.4. Fig. 2.5.3 Rectangular pulse of duration “T’ amplitude “A” Rectangular pulse : (4) =A ele pel T 2 2 =0 elsewhere ws (2.5.5) rect Duration'T' & amplitude 'A' ‘The above equation follows from equation 2.5.4, 4, Delta or Unit impulse function [8(¢)] Delta or Unit impulse function [6()] plays very important role in communication systems. As we have seen earlier that the causality and stability of the system is determined from the impulse response. Impulse response is obtained at the output when input is delta or unit impulse function.The delta function has zero amplitude every where except at t =0 and at t =0, the amplitude of delta function is infinitely large such that the area under its curve is equal to 1. The delta function is shown and defined in Fig. 2.5.4. x(t) = 8(t) : A unit impulse or delta function. “1' near the arrow of the pulse represents area under the pulse. 3@=0, #0 a» (2.5.6) Area under unit impulse soma w= (25.1) Sifting property of delta function The sifting property of delta function is,Communication Systems - | 43 Signals and Systems Sifting property : J x()8(t ty) dt = at) ww (2.5.8) 3(¢ -t,) is the time sifted delta function. The left hand side of the above equation sifts out r(f,) of x() at t =1p. Replication property of delta function This property is derived from the sifting property. We know that delta function is even function of time. i.e., BE -ty) = SHE -ty)] = Sty -) w= (25.9) Using above result we can write equation 2.5.8 as, { x) (t, -f) dt = tty) Let t =7 in the above equation (just for convenience of notations), ie, J x(t) 6(ty —1) dt = x(tp) Lett, =t in the above equation (just for convenience of notations), ie., f2@at-9 de = xO 5 w= (2.5.10) This equation represents the convolution of x(t) and a(t and it shows that, convolution, of any function with deli function leaves that function unchanged, This is known as Replication property of delta function. Equation 25.10 can also be represented as, Replication property : x(t) + 8(f) = x(¢) w= (2.5.11) 5. Exponential function Exponential function is used in phasor notations. Therefore exponential function is used extensively in mathematical analysis. The growing and decaying exponential functions are shown and defined in Fig. 2.5.5. Fig. 2.5.5 (a) : Decaying exponential function. wv (2.5.12)Communication Systems - 1 4 Signals and Systems Fig. 2.5.5 (b) : Rising exponential function, ww (2.5.13) 6. Sinc pulse Sine pulse or sinc function is very important mathematical model. It is used extensively in mathematical analysis of communication systems. Sampling theory, digital transmission uses this model. Sinc(x) is defined as, sin (nx) sinc(x) = se» (2.5.14) Here x is an independent variable. sinc function is an even function. * Ex. 2.5.1 : Prove that sinc (0) = 1 and plot sinc function. Sol. : We can write sin (nx) as, git gine io. gle sin (nx) = ————-. since ——"—=sin0 _... (2.5.18) 2 2j The exponential series is given as, 4 5 AL ; d= Fg -» (standard relation) . re ae ae 4 ae 2) . gages EE ee ee 2! 3! 4! 3! (Here ¢ = j rx) .. (2.5.16) -(jnx) (ixx? (jx2)° Ginn* (jx) and @ UT) 21 (jn + VEY URED ,ury ny, 21 3! 4! 5! 2 wm (Here t =—j m2)... (25.1 From the above two series expansions we can wgite, os Usa) B54) FF) _ URE ony +5 Gms) +R ae ewe Now divide the above equation by °2j’ on both sides we get, (jnx) _ (ins) inv? av? i A es ee 2j 3! Qj BI Oj 1 (jinx? 1 (jn i = — Sy is on HERA sin(nx) = mx 4a 7 aj + (2.5.18) * This example is optional. Students can skip this example.Communication Systems - | 45 Signals and Systems Left hand side of above equation is obtained from equation 2.5.15. If we divide the above equation by nx we obtain sinc(x). i.e., a ‘ 3 ¢. 5 sine) 82 1 Gey” LUE mx mx 3! jax 5! jax sin (xx) sinc (x) = aid Gan? +E (ina! pesca 12.2.2 1.4 4.4 Sp w rtp wx - 3! 5! 22 44 mx mx + 3! 5! Thus we obtained the series for sinc (x) function as, a+ =1- ws (25.19) sinc (0) = 1 from above equation. Other values can be obtained from definition of equation 2.5.14. sinc (1) = 38 Putting x = 1 in ® definition of equation ...2.5.14 =0 sinc (1) = 889.9 =n sine @) = 2228 29 ined), =, S29 29 2n Qn Thus at all multiple values of x sinc function will have zero value. sinc (0.25) = sine (- 0.25) = S277 . 9 sinc (0.5) = sna) = 0.6366 sine (- 0.5) = 0.6366 oe et Sih sinc (0.75) = 28%) - 93 sine (- 0.75) = SRC _ 93 34 sine (15) = 0.2122. sine (1.5) = 2159 _ _ 2102 =15n sinc(25) = 2259 «93973 sine(-2.5) = 82259) _ 9.4275 25x 25%Communication Systems - | 46 ‘ Signals and Systems The waveform obtained by plotting all the above values is shown in Fig. 2.5.6. Fig. 2.5.6 : The sinc function : Figure shows the smoath curve of sine (x) versus x. The curve passes through zero at multiple vatues of x. 2.6 Fourier Series Fourier series represents the periodic waveform in terms of infinite number of sine and cosine terms. These sine and cosine terms are called basis functions. Fourier representation of signal gives information about harmonic content of that signal. Fourier series is the representation of the signal in time domain series only. 2.6.1 Quadrature/Trigonometric Fourier Series Consider x(t) be a periodic signal. A fourier series can be written for this signal as follows. > 2nnt < sin | 220 +2 we 7 |: dons oe | w= (2.6.1) 0 0 x0 Here T, = r is the period of the signal such that, 0. x) = xt +7) In above equation if we write a, = 2nf, = = O a) = ay + > 4, COS (N@gt) + Za, sin (nat) wn (2.6.2) nal es Here a, ,a, and b,, are given as followsaa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.tion Systems - | 81 Signals and Systems + 5 cos (4 nt) +. 44160 In the above fourier series one component is cos (32). It is also given 449K" that one component is A cos (3 m1). Equating these two we get, Acos(3nt) = —° + 0s (31) 4490 A= z This is the required value of A. 4+9n Table 2.6.6 : Summary of fourier series Qustrature/Trigonomettic fourier Polar. fourier series Complex exponential fourier serles series @ 2 © og xl)=a9+ E ay coo[242!) w)=09+ ¥ D, we. Fe a4 fo net neve annt , Pe Here, Here, Here, I Tp WT . 1 1 -[2nntlly age fo x(n a Dy =aq-+ x(yat | Get foxme 0" 7, f 0- 40-7, ! anz, J ito J xq) at = ag t 89 = Co Dy = Co Cn= $(@n = tm) 22h * 4 an = BAe Cy oh = Ment tm) By eseinlcyl fy 20 (= tan [ C= Cn an)82 ‘Signals and Systems torn >0 forn=0 0.4240. np forn0). Therefore integration can be performed from -« to 0 instead of —0 to +00. Fourier transform of x(t) becomes XO) = f xe? at by equation 2.8.1 <2 bxi0 t@ Fig. 28.5 (a) Amplitude spectrum and (b) Phase spectrum of rising exponential pulseaa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.99 Signals and Systems Communication Systems o+ 2 ot PT ate fe PM are fot PM at 0- O+ ‘An integration of first and 3 term in above expression we have obtained in example 2.8.2 and example 2.8.3 respectively. And integration of middle term will be, o+ oe fue? dear fe? at ai fe* -e]= 10-0 =0 o- o- Or in other words integration at single point with upper and lower limits same is zero only. —1_40+1_ a-j2nf a+janf = = oe wn (2.8.27) a’ +(2nf) X(f) = Equation 27.25 can be written in short hand as e~“! ot yt) =e ul]. 'u(). [Here when <0 ; Fourier transform pair is represented as shown below, Symmetric double exponential pulse : e"*! «24 a” + (2xf) wns (2.8.28) 2a Here IX(f)| = =——z a? +Qnf) and ef) = 0 Fig. 2.8.7 shows the amplitude spectrum of this function. There is no variation of phase with frequency, it is zero at all frequencies. bl Fig. 2.8.7 Amplitude spectrum of symmetric double exponential pulseCommunication Systems - | 100 Signals and Systems Ex. 28.4 Obtain the Fourier transform of the antisymmetric exponential pulse as shown in Fig. 2.8.8 below. Fig. 2.8.8 Antisymmetric exponential pulse Sol. : This pulse can be very easily represented with the help of equation 2.8.26. We have used in last problem. Here overall value of the pulse is negative for negative time. nat ie, x@) =e"; t>0 = |p; t=0 =-e"; t<0 wn (28.29) ~. Fourier transform will be, Xf) = J x@ el?" dt from equation 28.1 e o- os @ . = fant PM ae FP an fore oa ay Soi o- osaa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.Communication Systems - | 114 ie. Ts (x, t=) dt & X,(/) X, (f) Ex. 2.8.10 Obtain the Fourier transform of the antisymmetric double rectangular pulse shown in Fig. 2.8.17 below. Fig. 2.8.17 Double pulse Sol. : The signal x (t) can be described as x) =A; -T 8(t) form a Fourier transform pair of de signal in time domain and delta function in frequency domain, then i cos (2nft) dt = 8(f) o» (2.8.67) Sol. : By definition of FT, « , x)= f x(tye 2 at Now if x() = 1,X(f)=8(f) from the given condition in example. f rei? dt = 8(f) 2, or f oPM dt =agy a» (2.8.68) j2mft . -jrnft cos (2nft) = © a and «- (2.8.69) pest | izes sin (nf) = —— 5 5 (2.8.70) i From above two equations we have, eft, -j2nft ,f2nfl cos (2 nf t)—j sin(2nft) = ot ; a ites w= 2.8.71) ~ jane Now substitute value of ¢ from above equation in equation 2.8.68 ie., j [cos (2 xf t)— jsin(2nfp] dt =8(F) -e Now since 5 (f) is real valued, complex function will have zero value in above expression,Communication Systems - | 123 Signals and Systems i cos (2m ft) = d(f) Hence the result is proved. Ex. 2.8.18 Obtain the fourier transform of a) = el 7h! Sol. : By definition of FT, xy = f xe l?"!* at - I eh 2thel i2Rft gy ee LAL =fe ON dt =d(f-f) from equation 2.8.66 Fig. 2.8.24 shows the amplitude spectrum of x(f). It has only one frequency component at f.. Fig. 2.8.24 Magnitude spectrum of e! 7"! Amplitude spectrum of e!?*' has only one frequency component f. of unity amplitude. Ex. 2.8.19 Find out the fourier transform of cosine wave shown in Fig. 2.8.25 below.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.Communication Systems - I 140 Signals and Systems Ex. 2.8.28 Find out the fourier transform of the periodic pulse train shown in Fig. 2.8.38 below. x(0 Fig. 2.8.38 Periodic pulse train of rectangular pulses of amplitude A Sol. : For the periodic pulse train by above figure we have obtained C, in example 2.68. co, #4 wpe (F) By equation 2.6.49 ...(2.8.107) Fourier transform of a periodic signal is given by equation 2.8.90 as, Xe z Cy =f) Putting value of C,, given by equation 2.8.107 in above equation, xy= ¥ - sine (=) 3 (f-"fy) as (2.8108) This equation gives the fourier transform of rectangular pulse train. Ex. 2.8.29 Find out the fourier transform of the cosine wave represented by Fig. 2.8.39 x) = Acos(2nf, u(t x) Fig. 2.8.39 Cosine functionCommuniéation Systems - | 141 _ Signals and Systems Sol. : By definition of FT, x= f x0 arial ag =f Acos2nf.y oP" at 0 =47 [ere « ores i2nft ay 2 0 jo, jo [soe ‘ east] 2 EAT jae felt eraden) 2 0 JA j eo 2a fot arsfe FERAL ay 2 oO Af, 1 a-fot _d ePrfo] 2) j2xf-f) J2aF+f,) lo Beep oe ign 2 [j2mF-f,.) 2m +f.) A [4 1 =A et ian |f-f f+f. wf «=» (2.8.109) Fig. 2.8.40 shows the amplitude spectrum of the cosine function for #> 0: + Fig. 2.8.40 Amplitude spectrum of cosine wave : A cos (2.x f,t)u(t)Communication. Systems - | 142 igpals and Systoms Ex, 2.8.30 Find out the fourier transform for A sin (2nf,t) u(t). The sine wave is shown in Fig. 2.8.41. x(t) = Asin 2nf,t) ut) Fig. 2.8.41 Sine wave Sol. : By definition of FT X(f) = j x(t) eT? at Here x(t) = Asin(2nf,t) u(t) xy) =f Asin(anf.t) el? dt 0 fee /® We know that sin® = —a then above equation will be, J x(f) -4 i [eine _ oFnhet) en lPtft ay lo Asin (2nf,t)u(t) w+ (28.110) Fig. 2.8:42°shows'the spectrum of sine wave. a Fig. 2.8.42 Spectrum of sine wave :-A sin(2nf,t) u(t)aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.Communication Systems - | 149 Signals and Systems 1 1 Hyy< Lt T,, RC ae titas te T4+j2nfRC & X(f) =F {re woh -f petlRC gr i2S ay % anf) (1+ j2nfROy Rc”! 2 2 1 te" u(t) “(axa ww (28.123) The students are given as an exercise to prove this property. Y(f) = XY) HY) ROP 1 (14 j2uf RO? (1+ j2xf RQ) yt) =F Yl ye ee | RO (+ j2nfRO? 2 -t/ RC = - tre 7 u(t) From equation 28.123 extension.Communication Systems - | 150 Signals and Systems Ex, 28.37 Impulse response h (1) is given by h(t) = eh u(t) +e"! u(t) The system is excited by x) = eu) Find out output y (t) Sol. : Output y(t) = x(t) * A(t) & Y() = xf) H(f) Here y@) = EMO) By definition of FT, H(f) = FIO) = j [een wee ia dt Q 2 . = fe Pt ate fet oP at he 0 2 1 (2-iaepe i ph ne itapit |? 2-j2af mo 1+janf 0 wh SG EH 2-j2nf 14 j2nf pe grey 2—j2nf 1+j2nf (1+j2nfyQ-j2xf) & X(f) = Ax@) ~~~ jdaft a fe ple s2as yt foeve dt =[Pe dt 1 24 j2nf Now, Y(f) = Hf) Xf) =——___4__, (14 j22f)(Q—j2mf\Q+j2nf) ao, b ie 1+j2nf 9 2-j2nf 2+j2afCommunication Systems - | 151 Signals and Systems Here, ee =1 2-j2efQ+i27H)| jagp oa 3 ie © 4 j2nf)Q-i2xf) janpen “1/4 3 = =-3/4 OF F2RNC-727A)| jagp a 1/4 B3/4 jay Q-janf 2+j2nf The inverse fourier transform of af 1/4 “4 1/4 12 pf apt ft/s _| oul Pg 2 ee (at ae) This function approaches to - 2 as t->+20 and approaches to ‘0’ as t > -. Therefore this term can be identified by negative time function. -. By taking IFT of equation 28.124 and applying equation 2.8.15, yt) ee'u@y - 36? weyet ce u(-t). Y(f) = wu (2.8.124) Ex. 2.8.38 Modulation Theorem The signals like x(t) cos(2nf, t+) represent carrier modulation. Here x(t) is the modulating signal and cos(2nf.t+) is the carrier. The process of modulation translates the spectrum of the signal up and dew by frequency f,. Thus, x(t) cos (anf t+4) o£ see x(f f. deze HKG) This relation is called soon theorem. [Nov.-95, 6-Marks] Proof : Let the fourier transform of x(t) be X(f). Then, land tie | iid ie : By Euler's theorem. j* 2a i ~j2n, =F xen fel + abe " ‘] 3 js -i . =f amet ae F [roe P¥e | 2 2 let us use the frequency shifting property which is given as, Por. fie, F [x (¢) cos (2nf,t+9)] = reaa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.Communication Systems - | 153 Signals and Systems Ex. 2.8.40 DC Signal as a limiting case of sinc pulse From example 2.8.8 we know that AsincQ2W8) «> rect (4 =» (28.132) 2W 2W This is a fourier transform pair. The sinc pulse represented by above equation is shown in Fig. 2.8.13 Let x(t) = A sinc(2Wt) x(t) will represent DC signal as W > 0ie., x(t) = yin, Asinc(2Wt) =A w= (28.133) Applying this limit to the fourier transform pair of equation 2.8.132 we get, lim Asinc(2wt) @ lim 4“. rect (4) woo wo 2W 2W A @ AB(f) Since rectangular pulse becomes delta function as W -> 0. 1 © 8(f) which is same as equation 2.8.65. Ex, 2.8.41 Find the fourier transform of the unit step function and hence find the fourier transform of the function. x(t) = u(t) sin (2nf,t) Here u(t) is a unit step function. Sol. : From equation 2.8.76, the fourier transform of unit step is given as, ut) o wn (28.134) j2nf The given signal x(t) can be written as, ¢ Fd _~ Pad x(t) = u(t) | —— ——_ 2j 1 J2nft a {u(t b fen W -upe Pt Taking fourier transform of the above equation. X(f) = 4 {F [ue eho -F [xe ere J The fourier transform of u(t) is a and by applying the frequency shifting -y property of equation 2.8.43 to above equation we get,Communication Systems - | 154 Signals and Systems -eijJii__-_1_ 0b mara _1,_f Inf? 5p? Observe that this result is same as that we obtained earlier in equation 2.8.110 with A = 1. Ex. 28.42 Find the fourier transform of B TG 1+(2nt) Sol. : From equation 2.8.28 we know that, = alt} 2a Be P22 z a? +(2nf) Here we will apply the duality property given by equation 2.8.39 by replacing f with t and t with - f ie, x(t) = ages a? +(2nt)? ol 6 a? +(2nt) eth es ie. Let a = 1 in above relation, elt yo 2 1+(2nt) 1 -¥l o 1 ; 2 1+(2nt) Multiply both sides by B, Bt B Seo, 2 1+(2nt) B or r|—4 1+(nt)?Communication Systems - | 155 Signals and Systems Ex. 2.8.43 Find the fourier transform of the trapezoidal pulse shown in Fig. 2.8.48 below. xt) Fig. 2.8.48 A trapezoidal pulse Sol. : This problem can be solved very easily by using differentiation property. Lets differentiate the trapezoidal pulse two times. Fig. 2.849 shows the signal after first differentiation and after second differentiation. x(t) Fig. 2.8.49 Trapezoidal pulse and its first and second differentiated signals, The waveform of Fig. 2.8.49 (c) is, d? x(t) ae We know that 8(t)< 1 and from time shifting property of equation 2.8.42, the fourier transform of above equation will be, = k3@+t,)—k, 6¢+,)—k, 5 -£,) +k, Sit)Communication Systems - | 156 Signals and Systems 2 F [2 al 2 ke ky oP ig oP ag, ge L gPaa p49 PTS g PAN gg PHA = 2, |= ___ - $_** __ 2 2 = 2k, [os (2xf ty) - cos (2aft,)] ws (2.8.15) From equation 2.8.47 we know that, a wo © (j2af)X (Ff) If we differentiate above equation ‘n’ times it will be fy) o (j2nf)" X(f) (2.8136) F / Applying this property with n = 2 to equation 2.8.135, Z 2] = (amy? Xf) dt ‘| 1 ys -2k, [cos (2nf t,) ~cos (2xf (nf t 2 k Rs af [cos (2nft,) -cos (2aft,)] S The value of k, can be easily obtained as then above equation will be, x(f) = w+ (28.137) This equation gives spectrum of trapezoidal pulse. Ex. 2.8.44 Determine the fourier transform of the square wave shown in Fig. 2.8.50 below. Fig. 2.8.50 Square waveCommunication Systems - | 157 Signals and Systems Sol. : Let us first calculate the fourier transform of one cycle of the waveform as shown below. Fig. 2.8.51 One cycle of square wave This cycle can be thought of as two rectangular pulses. Signal x(t) can be written as, x(t) = v¢) - v¢-2) a» (2.8.138) By taking fourier transform of both sides of above equation and using timeshifting property for second term on RHS, Xf) = VUA)-ViF) Pt? = Vif) ft - on as (2.8.139) Here v(t) is a rectangular pulse. For this pulse we have obtained fourier transform ie. for pulse of amplitude A and width (- A rea( 2 © AT sinc(fT) By equation 2.8.41, Here, A= 1 and T =2 then above pair will be, rect (5) & 2sine (2F) or V(f) = 2sinc(2f) Putting value of »(f) in equation 2.8.139, X(f) = 2sine(2f) fei) This is the fourier transform of one cyde of square wave of Fig, 2.8.51. For the periodic waveform of Fig. 28.50, X(f) is written as, X (af,). Since f=nf, and n = 0, +1, £2,.... $0 X(nfy) = 2sine(2n fy) [re Mm]Communication Systems - | 158 Signals and Systems The period of the square wave is Ty =4. fy Then above equation will be, cited 2 X(nfy) = 2sinc (2n- a 4 ( = O8ine “GF om] oe ( ze = [cos tam) ~ j sin (nn)]} = 2sine () {1 - cos (nn) } Since sin (n x) = 0 for integer values of n. In the above equation, 1-cos(nn) = 0 for n= 0,2, 4,.... even ‘n’ =2 for n=1,3,5, .. odd ‘n’ xorf)=asine{ 3) for n=1, 3, 5, ....0dd'n’ 2 ++ (2.8.140) =0 for n=1, 3, 5, «..odd'n’ From equation 2.8.98 we know that the fourier transform of a periodic signal is given as, TE Kop) g-nf) i tes (odd) sy 4sne (3 )sy- nh) (odd) xf) - & sine (5 )5g- nf) - (2.8.141) neve (odd) This is the required expression for fourier transform.Communication Systems - | 159 Signals and Systems ‘Table 2.8.2 : Summary of fourier transform equations Nonperiodle Signals X(t) =" f xe Pt x(t) = { xu) ot a Periodic Signals LO dlt-nh) new 1+T J sme i - j2nnfyt St oertyt ‘Specirum is continuous for onperiodic signals. ‘Spectrum is discontinuous for periodic signals. Table 2.8.3 Properties of fourier transform ‘Sr. No. Property ‘Mathematical Description 1. | tinearty 2%; (N)+ Bxp it) © aX4(F)+ OXa if) a & b ate constants ey xo + x{2) a is constant ii “a 3. [ Duality 1) @ X(f) then X(Q 6 x(-f) Tart h 4. | Time shifting x(t) © xu) etethO u i aR 5. | Frequency shiting fF E? 5a) Xf) 6. | Area under x () 2 [x0 a= x (te Fratt=0 7. | Area under X () 2 J xX a = x(ierat=o & | Differentiation in time domain é. suisse LD) 9. | Integration in time domain t 1 Luo ae Bax) 10. | Conjugate functions x() @ XU) then x(t) 4 X" (=f) 11. | Mutipization in time domain = y(t xp(t) of X4()X_f-ada 12: | Convolution in time domain % Pein tg tn oe @ mr) xpUt)Communication Systems - | 160 Signals and Systems Table 2.8.4 : Summary of fourier transform pairs Name of the signal Mathematical representation Fourier transform Rectangular puse of width T | (! } T sine (77) © 2 | Sing pulse with zero crossing | sinciam) tea ry at aw ax) Ww 3 | Decaying exponential signal for | 5") a > o i t>0 avjent | Double exponential signal r*Maso 2a 3 a? + (ant)? Sp” | Tisaer 1-2 sr T sine? (fT) loiter 6__| Unit impulse 0 7 7._| DG Signal 1 3h) | Delayed unit Impulse 0%) eta ‘| Phasor of frequency fg genet 80%), 10._| Cosine wave signal cos (2 Rh f) fa a lee) 3 BU -m+ 8+) 11. | Sine wave signal sin@xt, i, mi ay four - t+ 806 +] 12 | Signum function 390 () — Tat 13. | Unit step vO 1 2 tae 14,_| Impulse train or samy 2 z function = mT) ZX 8(f~alg) Exercise 1. Define fourier transform for nonperiodic signals and explain its significance. 2. Explain the conditions which are required to be satisfied for the signal to be fourier transformable. 3. State and prove following properties of fourier transform. i) Multiplication in time domain. ii) Convolution in time domain. iii) Differentiation in time domain. iv) Integration in time domain. 4. State and prove following properties of fourier transform. i) Time scaling ii) Time shifting iii) Frequency shifting 5. Define the fourier transform for periodic signals and explain its significance.Communication Systems - | 161 Signals and Systems 6. What is Poisson’s sum formula, how it is defined ? 7. What is sampling function ? Comment on its fourier transform. Unsolved Examples 1. Find the fourier transform of the half cosine pulse shown in Fig. 2.8.52. [ Hint : The pulse of above figure can be considered to be produced of A cos(nt /T) and rect (t/T). And multiplication in time domain is converted into convolution in frequency domain]. x) Al Tr Tr Fig. 2.8.52 Half cosine pulse [Ans. : X(t + sine (+4) 2. Find the fourier transform of the damped sinusoid given by x() =e! sin (2nf,t) u® anf, ans: K(f) = —— "= _] (2nf, ++ 5200) 3. Find the fourier transform of x@) = — 1+(2n1) [Hint : Apply duality property) fAns.: X(f) = 1007!fly 4. Prove following relations ‘ 2a FE) Gans" XY) [Hint : This can be proved using differentiation in time domain property]. i) M2@) eo (4) x yy 2x | Hint : Differentiate both sides of FT equation ‘n’ times with respect to f].Communication Systems - | 162 ‘Signals and Systems 5. If x(t) = (¢-3)", then find out the value of i) J x@ 8 erat ii) x) * 8(t+4) [Hint ; Apply sifting and replication properties of delta function. : [Ans. : (i) 49, Gi) (t+1)7] 6. Using the convolution property find out the spectrum of x(0 = sin(2nf, t):cos(2mf, 0) [Hint : First find FT of sin(2mf,!) and cos(2nf,#). Then take their convolution using convolution property] tans.: 4 {0 +h, th) +8(F th ~h)-8(f-fy th)-8-f “fh 2 7. Find the spectrum of x(t) = e ™"'/7? ot (Ans: X(t) =Te™™ ] 8. If x(t) has the fourier transform given as, = i2ef Xf) = 1+j2nf Then find out fourier transform of x (2t+2) [Hint : Use time scaling and shifting properties] jnf int (Ans. : F [x(2t+2)] = —~—_- ] MSEC Span 9. Using modulation theorem find the spectrum of x(t) = Au(t) cos (2nf, t) [Hint : Refer example 2.8.41] tans: AL. j2ne 10. Prove that X(E-ty) +x +t) > 2X (fF) cos2nftyCommunication Systems - | 163 Signals and Systems University Questions November - 95 Q.1 Define unit impulse function and find its fourier transform. [6 Marks] ‘Ans.: Unit impulse function is defined on page 42 and its fourier transform is given in Ex. 2.8.15 (page 120). Q.2° List out various properties of fourier transform and prove modulation theorem stating its significance in analog communication system. [6 Marks] Ans.: Properties of fourier transform are listed in table 2.83 on page 159. Modulation theorem is proved in Ex. 2.8.38 (page 151.) The significance of modulation theorem is also stated in Ex. 2.8.38. May - 96 Q3 List the merits and limitations of fourier transform. [4 Marks] Ans.: Advantages of fourier transform 1. Fourier transform is linear. 2. Fourier transform of periodic as well as non periodic signals can be calculated. 3. It gives friequency spectrum of the signal. 4, It is also used to study esd and psd of signals. Limitations of fourier transform 1. Fourier transform can be calculated if the signal is absolutely integrable. 2. Fourier transform can not be calculated for discontinuous signals. QA Find the fourier transform of the RE pulse of Fig. 2.8.53 and discuss the result. [6 Marks] Fig. 2.8.53 RF pulse Ans.: Fourier transform of RF pulse is calculated in Ex. 2.89 on page 108. Its fourier transform consists of two sinc functions located at + f.. Q.5. State and prove two convolution theorems. : [6 Marks} ‘Ans.: Convolution theorem in frequency domain is given as, 0) O XDCommunication Systems - | 164 Signals and Systems —— 7 This property is also called multiplication theorem. It is given by property 11 on page 112. ‘The convolution in time domain is given as, 1020) & X(NX This is given by property 12 on page 113. December - 96 Q.6 Show that compression in time domain is equivalent to expansion in the frequency domain and vice versa. [6 Marks) Ans.: This can be proved with time scaling property of fourier transform. It is given as, xat) ae ) If a < 1, then there will be compression in time domain and expansion in frequency domain. . Q.7 State and prove the frequency translation theorem. Explain its utility in transmission and reception of AM signals. [6 Marks] Ans.: The frequency translation theorem is nothing but modulation theorem. It is discussed in Ex. 2.8.38 on page 151. Q.8 An audio oscillator is switched on at t = 0. Find the spectrum of the generated (i) cosine waveform and (ii) sine waveform | > [6 Marks) Ans, : Since the signal is generated at t = 0, the generated sine and cosine signals will be, x(t) = cos(2nft)u(t) 2x,(0) = Asin(2nf.t) u(t) The spectrums (Fourier transforms) are obtained in Ex. 2.8.29 and Ex. 2.8.30. May - 97 Q.8 List the merits and limitations of fourier transform. [4 Marks] Ans.: Refer to the answer of Q.3. Q.9 Write and explain the derivation of the fourier transform pair from fourier series for a periodic signal. [4 Marks] Ans.: Fourier transform of a periodic signal is obtained from exponential fourier series. This result in derived in Sec 2.8.4 on page 131 and 132. December - 97 r Q.10 State and derive convolution theorem in time domain and explain its significance and use in communication systems. [2 Marks] Ans.: Convolution in time domain is given by property 12 on page 113. Let h(t) be the impulse response of LT1 system and x(t) be the input. The output of such system is given as,Communication Systems - | 165 Signals and Systems yt) = x(rh(t) Thus output is the convolution of input and inpulse response. With the help of convolution theorem we can write above equation as, Vf) = X()-HY) Here H(A) is called transfer function of the system. Q.11 Find the fourier transform of a sequence of equidistant inpulses of unit strength separated by T seconds. [6 Marks} Ans.: This is basically impulse train of period T. The fourier transform of such signal is calculated in Ex. 28.27. Q.12 Fourier transform is a limiting case of fourier series by letting the period of periodic function inifinite, Justify the statement with an example of periodic gate function. [8 Marks] Ans.: We have obtained the fourier series for periodic gate function in Ex. 2.6.8. The gate function and its spectrum is given below. The spacing between the frequency components is fy. x(t) - -1 To 2 2 Fig. 2.8.54 Periodic gate function and its spectrumCommunication Systems - | 166 Signals and Systems AS Ty, ~~, the periodic gate function is converted to a single rectangular pulse. This pulse is shown in Fig. 2.8.55. The spectrum of this pulse is calculated in Ex. 2.8.7, which is given as, X(f) = Arsine( fr) This spectrum is also shown below : xt) Fig. 2.8.55 Rectangular pulse and its spectrum 1 0 continuous as shown in above figure, That is the spacing between the frequency components is zero. Thus fourier transform is the limiting case of fourier series. May - 98 Q.13 State and derive convolution theorem in time domain and explain its significance and use in communication systems. [2 Marks} Ans.: Refer to the answer of Q.10. Q.14 Show that the relationship between the time domain and frequency domain descriptions of a signal isan INVERSE one. That is a pulse, narrow in time has a significant frequency description over a wide range of frequency. {10 Marks} As Ty >, f,>0 since Ty =. Hence the frequency components in the spectrum are Ans.: — Fig. 2.8.55 shows the rectangular pulse of width + and its spectrum. As the width + is reduced, the width of the main lobe increases. This is shown in Fig. 2.8.56.Communication Systems 167 x(t) Fig. 2.6.56 Reduction of width in time increases significant frequency range As t->0, the rectangular pulse becomes impulse function and its spectrum will be as shown below. The spectrum of impulse function is de signal. It contains all frequencies. Thus as time description of the pulse is narrow, then its frequency content is wide. 5) + i x) Fig. 2.8.57 Impulse function contains all frequency components Q.15 Show that the spectral pattern of the wave form x(t) = cos(2nf,t)cos(2nf) has 4 spectral lines corresponding to two real frequencies(f, + f-,) and (f.— fy) {8 Marks] Ans.: The given signal is, a(t) = cos(2nf,f)cos(2nft)Communication Systems - I 168 Signals and Systems We know that cos(x) cos(y) = Jeos(x-y) +cos(x+). Hence above equation becomes, a) = Zeosl2f+ fl +L cos ffl Fourier transform of cosine wave of frequency fy is given by equation 2.8.72 as, cos 2nf) © 8 —fo)+ AF + Hence fourier transform of given signal becomes, RD = Bef +00 fet # HOPI, Sy) # P+ f Sal = ERC fh EMU Ue fl} + FBC ~Ce— fll +E 3U + Ce fal This spectrum shows four frequency components located at (f+ f,), - Ue +f fe) and ~ (f,~f,) The spectrum is shown in Fig. 28.58. . xf) = (fotfm) — (ferfm) Fig. 28.58 Spectrum of x(t) = cos (2rf,,t) cos (2xf,) December - 98 Q.16 Explain that the signal cannot be band limited and time limited simultaneously. (6 Marks.) Ans. : The rectangular pulse of width ‘T’ and sinc pulse is the fourier transform pair ie., A reo 2) AT ssinc({T) from equation 2.8.41 Here observe that rectangular pulse is time limited. But its spectrum is not band limited, since sinc function has values from -© to + 2. Now consider the case of rectangular pulse of width 2W in frequency domain. Its inversed fourier transform is given as,Communication Systems - | 169 Signals and Systems A f . A Wi ssrect = — .8 sinc(2W1) Are(£] from equation 2.8.41(a) Here observe that the spectrum is bandlimited, i.e. rectangular pulse. But its time domain signal is sinc function. It has values from t = —« to +. Hence it is not timelimited. This shows that the signal connot be bandlimited and timelimited simultaneously. May - 99 Q.17 List out various properties of fourier transform and prove modulation theorem stating its significance in analog communication system. {6 Marks} Ans.: Properties of fourier transform are listed in table 28.3 on page 159. Modulation theorem is proved in Ex. 2.8.38 (page 151.) The significance of modulation theorem is also stated in Ex. 28.38. Q.18 State the conditions for fourier transform to exist. [2 marks} Ans.: The existence of fourier transform is discussed in Sec 28.2 December - 99 Q.19 Show that the relationship between the time domain and frequency domain descriptions ofa signal is an INVERSE one. That is a pulse, narrow in time has a significant frequency description over a wide range of frequency. [10 Marks} Ans.: — Refer to answer of Q.14. 0.20 Show that a rectangular pulse and a sinc function constitute a fourier transform pair. {4 Marks} Ans.: This is proved in Ex. 287 Q.21. Write short note on symmetry property of fourier transform [6Marks] ‘Ans: Symmetry property is given by property 3 on page 103. May - 2000 Q.22 State the following properties of fourier transform with necessary mathematical expressions. i) Scaling i) Time iii) Frequency translation iv) Time convolution. (8 Marks} Ans.: — These properties are discussed in Sec 2.8.3. December - 2000 (Q.23 Find fourier transform of following functions. ay pvaltl Fr _[ 1 fort>o pel for a>O ii) sgnt=) “pete where sgn tis a signum function, Also sketch the functions and amplitude spectrum. {8 Marks} . Ans: Fourier transform of e~“'‘! is obtained in Ex. 2.83. Fourier transform of signum fanction is obtained in Ex. 28.5Communication Systems - | 170 Signals and Systems May - 2001 Q.24 State the following properties of fourier transform with necessary mathematical expressions. i) Scaling ii) Time iti) Frequency translation iv) Time convolution. {8 Marks] ‘Ans,: These properties are discussed in Sec 2.8.3. Q.25 Find the fourier transform of gate pulse given by equation g(t) =rect (4) 7 © 1 for|t}<— where rect, ( 4) = 2 TY Jo foritl>= 2 Also sketch its fourier spectrum. [8 Marks] Ans.: — Fourier transform of gate pulse is obtained in Ex. 2.8.7. December - 2001 Q.26 Find the fourier transform of following functions and sketch its amplitude spectrum. De Midt) fora>0 i) Modulated signal g(t)coswgf in which g(t) is a gate pulse, rect{ es, [2 forlti< = where rect) = 2 0 forltt>= frites [8 Marks} Ans.: i) Fourier transform of ¢“u(t) is calculated in Ex. 28.1. fi) When cosmpf is multiplied by a gate function, the signal becomes, T sts in COS Oye tye [OO oO elsewhere Fourier transform of such signal is obtained in Ex. 2.8.39. Q.27 Show that a rectangular pulse and sinc function constitute a fourier transform pair. [4 marks} Ans.: This is proved in Ex. 2.87. May - 2002 Q.28 State and prove following properties of fourier transform. i)scaling i) Frequency shifting Also explain significance of these properties in communication systems. [8 Marks] Ans.: These properties are explained in sec 2.8.3aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.Communication Systems - | 173 Signals and Systems (Ay rect? (4) af For sinc pulse. eS — ge Since rect (4) is rectangular pulse of unit amplitude over (-W, W). Exercise 1. State and prove Parseval’s power theorem. 2. State and prove Rayleigh’s energy theorem. Unsolved Example 1. Calculate the energy of the casual exponential pulse shown in Fig. 2.8.2. [ans.2E = University Questions May - 96 Q.1 State and prove Rayleigh’s energy theorem. : [5 Marks} Ans.: Refer to Sec 2.9. 2.10 Spectral Density Functions The spectral density functions of the periodic or nonperiodic signal x(t) represents the distribution of power or energy in the frequency domain. In other words for a energy or power signal, the total area under the spectral density curve plotted as a function of frequency is equal to total energy or average power of the signal. For the energy and power signals two types of spectral density functions are normally defined, ie. energy spectral density (esd) and power spectral density (psd). Let us now discuss these density functions. 2.10.1 Energy Spectral Density This spectral density function gives the distribution of energy of the signal in the frequency domain. We know that, s ® = flxora= fixer ¢ By equation 2.9.1; Rayleigh’s energy theoremCommunication Systems - | 174 Signals and Systems This equation gives the total energy of the signal from its frequency components energy in frequency domain. In this equation |X(f)| is the amplitude spectrum of the signal. Let us denote the squared amplitude spectrum |X (f)|° of the signal by y (f) ie., w-- (2.10.2) Putting above value in equation 2.10.1, an energy of the signal x (t) is expressed in terms of y (f) as, E= fw (af .. (2.10.3) This equation shows that total energy of the signal is given by total area under the curve y (f). That is y(f) represents the energy density of the signal x (t) in joules per hertz. The function y (f) is therefore called energy spectral density of the signal x(t). The energy w (f) carried by every frequency component ‘f’ is added (integrated) over the complete range of the signal to give its total energy. We know that amplitude spectrum of real valued signal is an even function of f. Therefore energy spectral density (esd) of such signal is symmetrical about the vertical axis at frequency f = 0. 2.10.2 Power Spectral Density This function gives the distribution of power of a signal at various frequencies in frequency domain. From Parseval’s power theorem we know that, BS > C,P ws» (2.10.4) For the periodic power signals C, is given by equation 2.8.95 as, c= me X(nf,) Putting above value of C,, in equation 2.10.4, the total power will be, e Zz xf)? ews (2.10.5) The power spectral density of the periodic signal x (t) can be defined as Power Spectral Densi = (2.10.6) This definition shows that power spectral density of the periodic signal is discrete function of frequency and it is defined at harmonics of fundamental frequency f,. Therefore 6(f—nf,) function appears in above equation. The total area under the power spectral density curve gives average power of the signal x (t). HenceCommunication Systems - | 475 Signals and Systems [sne- fA SX pope syne -« n=- z Using the sifting property of delta function given by equation 2.5.8 we have, fsma-k ¥ ixoge Sy Ty nite =P By equation 2.105. He pe j sifyadf ws (210.7) Thus average power is the area under the power spectral density curve. 2.10.3 Properties of Spectral Density Function The following properties can be summarised from the preceding discussion of power spectral density and energy spectral density. Sometimes these properties are also called combinely as properties of spectral density functions. 2.10.3.1 Properties of Energy Spectral Density Function 1) The total area under the energy density function gives total energy of that signal. ie. E= [yds By equation 2.10.3 ss nN If x(t) is input to a linear time invariant system with transfer function H (t), then input and output energy spectral density functions are related by, vo (f) =H yf) =» (2.108) Here wy, (f) is output energy spectral density function and y, (f) is input energy spectral density function. |H4f)|° is energy gain at frequency f Proof : Consider the ouitput be represented by y (t). Then in frequency domain we can write, Y(f) = A(f) X(f) Squaring the magnitudes of both sides, IMP =1H@P IX~P We know that |X(F)[> = y, (Ff) and [Y(F)? = yg (F) vo) =1HOP ¥.) 3. The autocorrelation function R(x) and energy spectral density function y (f) form a fourier transform pair,Communication Systems - | 176 Signals and Systems 2109) This property will be proved in next section. 2.10.3.2 Properties of Power Spectral Density Function 1) The area under the power spectral density function gives average power of the signal. ie, P= { s(f) af By equation 2.10.7 —s 8 2) If |H(f)[? is the power gain of the linear time invariant system, then output power spectral density S,(f) and input power spectral density 5,(f)_are related by, 5,0) =1H@P $f) ww. (2.10.10) Here S, (f) is output power spectral density and 5, (f) is input power spectral density. . Proof : Let x (t) be the input and y (t) be the output. For’the periodic signals we can represent the fourier transforms of x (t) and y (t) as, x= EL xa) 3y-m) 7 & Y(f) = yi Y (fy) 5 (f - nf) By equation 2.8.96 woo We know that, YU) = HOXP) Squaring both sides, men = RP x~e 7 Map)? 8 fmf.) = IHEP - z az em BF nfo) 0 o By equation 2.10.6, we can write above equation as, 2 5, (f) = IAW) 5, ) 3) The autocorrelation function R(x) and power spectral density function S (6) form a fourier transform pair. ie. R() @S(f) ves (ZOD) This property will be proved in the next section. Ex. 2.10.1. Find out the energy spectral density of the sinc pulse. Find out the total energy content also. ,Communication Systems - | 177 o Signals and Systems In example 2.8.8 we have derived that sinc pulse and rectangular pulse form a Sol. : fourier transform pair. ie., 7) Assinc (2Wt) *. rect (5 2w 2w X(f) == A ra (5) The energy spectral ders is given by equation 2.10.1 as, v@) =Ik~P a A f) of wef we ta! Total energy is given as area under the w (t) curve by equation 2.10.3, ie. = fun Ww 2 =f A. rect (Se _w 4W 2W Here limits are changed to (-W, W) since rectangular pulse is present over (AW, W). The value of ret F) is 1 over (-W, W). Therefore above equation will be, wo 42 E-f Ang -w 4W 2 2 -4!l wes aw? 2w Ex, 2.10.2. Find out the area under sinc? (t). Sol. : Consider the given sinc pulse as x (1) = sinc (t). Thus we have to evaluate, I sinc’ (dt = f (at From example 28.8 we know that, Asine (2W1)-A. rect (4 f ) (2.10.12) 2W Ww With A = 1 and 2W = 1, above pair becomes,Communication Systems - | 178 sine(t) © rect (4) f) 1] Thus the given sinc pulse is transformed into rectangular pulse of width f 1 5 1) Since W *5 and amplitude of 1. ie. F[sine(t)] = rect ( f or K(f) = ret | -, Energy spectral density wi) =1X@r = rec” (£) The total energy is given by equation 2.9.2. as, j was = J XQ af E A, ff rect? (f) af -Y 4 11 ; 2 why hid Since rect vm» (2.10.13) he " ¢ ) By Rayleigh’s energy theorem, ° De e 2 B= [ixmra= flee) ae i t 2 ie. J ixoPar= I IX(Qe as from eq. 2.10.13 J sinc? (t) dt This is the area under sine pulse Ex. 2.10.3 Fig. 2.10.1 shows the psd of the signal x (1). Find out its average power.Communication Systems - | Signals and Systems. 2 A a 4 2 f Fig. 2.10.1 psd of x (t) Sol. : From equation 2.10.7 we know that the total average power Is equal to the area under the psd curve. ie. ” u [snus frase i rips fids 22 34 1 = 6 (Ans.) Ex. 2.10.4 For the signal x(t)=e"" u(t), find out the total energy contained in the frequency band |f| an 2 2 gts nk Therefore equation 2.10.14 will be, 2 2 xe > (eZ) (f-"fy) wis (2.10.15) (odd) From equation 2.10.6, psd is given as, ee SI =z Leip? 8 -%) ) nsne -KOP since xy= 2 F xlof) 8G =m) For periodic signals. 4 = yY a ¢-m nee on (odd) Putting X (f) from equation 2.10.15 Here observe that 5(f~nf,) represents unit impulse at nf), therefore even after squaring, its value remains same. Since ‘n’ is squared in above equation we can write, s= YS 2. bos gm n= 1,35. nn 2 a(f- et $ SUE) «.- (2.10.16) nu n=1,35... n This is the required: relation for psd. By equation 2.10.7, average power is given as,Communication Systems - | 181 Signals and Systems Pe Actually average power can also be directly obtained by t+Tp Bek j Ix@p at 1, t This is left as an exercise to verify the answer. Ex. 2.10.6 Show that the two pulses shown in Fig. 2.10.2 (a) and (b) below have same energy spectral density which is given as, 4A2T? cos? (nf T) vif) = 7 «leg? -1] x(t) A Te 7 7 o T t 10-2 (@) Fig. 2.10.2 (b) Sol. : Consider the pulse of Fig, 2.10.2. (a). This is a part of cosine wave. It can be represented as product of cosine wave and rectangular pulse. i.e., x(t) =A rect (2) cos (2nf,t) (2.10.17) By modulation theorem we have obtained the fourier transforn{ of this function in example 2.8.39. It is given by equation 2.8.131. ic., AT AT x) = AE sine gt +82 sine + fT f “+ i.e. the frequency of the cosine pulse of Fig. 2.10.2. (a). Therefore above equation will be,Communication Systems - | 182 Signals and Systems AT Pome x= J )t eS sine (pr | By fee it ein(T- 3) sin| sf T 5) ® (fT -1) QfT +1) _ AT cos (nfT) {- 1,1 } z QfT-1 QfT+1) = 2AT cos (af T) vo» (2.10.18) ah -(2fT) Power spectral density of this pulse is given by equation 2.10.2. as, wif) =X? _ 4A? 1? cos (af 1) ee? ° fi -eT)'| _ 4A°T? cos? (nf T) 3 27 z lern -1] 2 Since [1-7 | = fern? -a] Now consider the pulse y(t). This pulse is delayed with respect to x{t) by z ie, y(t) jee wy =xyye 2 By equation 2.8.42 yf) = xy eTCommunication Systems - | 183 Signals and Systems OL = x) Since e-/*/™ does not affect magnitude. wif) = XP =P Thus the two pulses have same energy spectral density. This shows that energy spectral density is independent of time shift. Ex. 2.10.7 Find out power spectral density of the modulated signal x(t) cos (2 nf). Sol. : Let the signal be written as, y(t) = x(t) cos(2nf, t) From equation 2.8.125 we know that Here x(t) cos(2xf, #) of X(f-f.) + ; X(F+A) vif) =i xy-sy+ixye YN) = 5 8AM) + 5 RUHL Thus the spectrum of modulating signal is placed at +f. (carrier frequency) in the modulated signal. If X(f) is bandlimited to +W, then the spectrums of Y(f).will not overlap if f, < W. This can be very easily verified. psd of Y (f) is given as, wn =ir~r 1 1 Exg-1 +4 gio +5 oP Let us denote psd of y (t) as w A (f) and psd of x(t) as y , (f), then above equation becomes, vy = vif +4 avs Fth) th Thus the psd of y (0) is t of psd of x(f)-placed at +f,Communication Systems - | 184 Signals and Systems ‘Table 2.10.1 Comparison of esd and psd Energy spectral density y (f ) Power spectral density S (f ) it gives disiribution of energy of signal in frequency | It gives distribution of power of a signal in frequency domain domain 2 = vit) = ee Sit) = Extn 8 ¢ ato) yy nee Total energy (E) = f o(f) ar Total power (P)= f Sif) df “Aulocorrelation function for energy signal and esd | Autocorrelation function of power signel and ped form a fourier transform pair. form a fourier transform pair. Exercise 1. Explain energy spectral density and state all of its properties. 2. Explain power spectral density and state all of its properties. Unsolved Exampies 1. Find the psd of x(t)=5+12cos (2 mf, t) and f, =10Hz. [Ans : S(f) = 255 (f) +365 (f-f,) +365(f +f, )] 2. Find the psd of x(t) =A sii: (2x f, #) and hence find total power. 2 A tans : $(f)=4—5 (1,443 (¢+f,) and P 3. For the triangular waveform shown in Fig, 2.10.3, find out plot psd. x(t) Fig. 2.10.9 Triangular waveform [ Hint : For this waveform, C, = 4. for n=1,3,5...And S(f) ={C,I"] (xn) 4. Calculate psd for the halfwave rectified sine wave of amplitude A. { Hint : Calculate C, and $(f)=IC, |}Communication Systems - | 185 Signals and Systems University Questions May - 96 Q.1 Define power spectral density and explain its properties. {5 Marks] Ans.: Refer to Sec 2.10.2 and 2.10.3.2. November - 97 Q.2 —_Definepower spectral density function for a periodic function. What is its relation with fourier series coefficients ? How do energy signals differ from power signals ? [8 Marks} Ans.: Refer to Sec 2.10.2. The frequency components are basically fourier series coefficients. psd shows the power associated with each frequency component. Comparison of energy and power signals is given in Sec 2.2 in table 2.2.1. May - 98 Q.3 Explain following terms with their significance in electrical commumication systems. [2MarKs] ‘Ans: Refer to Sec. 2.10.2. The power density spectrum gives distribution of power with respect to frequencies. It is used to determine significant spectral components of the signal. November - 99 Q.4 Write short note on ESD and PSD function. (6 Marks] Ans.: Refer to Sec. 2.10.1 and 2.10.2. December - 2001 Q.5 — Write a short note on power spectral density [3 Marks] Ans.: Refer to Sec. 2.10.2. December - 2002 Q.6 Define the following terms with suitable expressions : Energy spectral density and power spectral density. [4 Marks} Ans. : Refer to Q. 4 above. 2.41 Autocorrelation Function Autocorrelation function gives the measure of similarity, match or coherence between the signal and its replica delayed by a variable period. There is little . — difference in defining autocorrelation function of energy and power signals. The autocorrelation function becomes the function of a variable delay period. 2.11.1 Autocorrelation of Energy Signals Let us consider an energy signal x(1). Then autocorrelation function of this signal is obtained by integrating the product of signal x(f) and delayed version of its complex conjugate.Communication Systems - | 186 Signals and Systems ie, Autocorrelation of energy signals : R(x) = [ x() x’ (@- dt ~~ (2111) Here tis the time delay. In this equation the complex valued signal is delayed in positive direction, It is shifted by same period *t’ in negative direction, the signal x(t) will also be shifted by the same period + in negative direction. The equation 2.11.1 becomes, J x(t+)x () dt Autocorrelation of energy signal: ow (2.11.2) Properties of autocorrelation function for energy signals : Property 1: The autocorrelation function shows conjugate symmetry, i.e., RQ) =R'(-9) ww» (2.11.3) Thus it states that real part of R(t) is an even function of t, whereas imaginary part is an odd function of t, Proof : Consider equation 2.11.1, R(t) = J x(t) x" (f-1) dt R@= fv Oxt-oat By taking complex conjugate R(-) = J Po x@en de = R(t) from equation 2.11.2 R(t) = R*(-0) This is same as equation 2.11.3 Property 2: The value of autocorrelation function of 7=0 (i.e. at origin) is equal to the energy of the signal. wn (2A) Proof : Consider equation 2.11.1Communication Systems - | 187 Signals and Systems j x(f) x’ (f=1) dt R(t) t=Ogives, R(0) = J x() x () dt =f xO? at =E Property 3 : If + is increased in either direction, the autocorrelation reduces. As t reduces, autocorrelation increases and it is maximum at t= 0 ie. at origin. ie., R@| < RO for Proof : Consider the function x(f) andx(t+1). [x() + x(t+7)) is always greater than or equal to zero since it is squared. ie. x) + x7 (t40) + 2x xen 20 lt v= (2.11.5) ie. x) ex +n) > £2 x) x40 Integrating both the sides, of [xP ate J jx@+pp ate 2 J xOxeen de E+E > 2R(0), if x() is real valued function E> R(t) or R@ > R(t) since R (0) = E by equation 2.114 Property 4 : The autocorrelation function and energy spectral density function of energy signal forms a fourier transform pair. i.e, R@Mevi(f) ww (2.11.6) Proof : Consider equation 2.11.1, R(t) = J x() x’ (f-0) dt In above equation replace x” (f~ 1) by its inverse fourier transform i.e., 2 ~ y R@= fF xO) fxme?ve? “| dtCommunication Systems - | 188 Signals and Systems " oo s “| ix er itast-0 asa f late ad foe. co2ateaay -T/2 T+0eT T/2 [me + lim ze T+ jim 2 | om (aorta “Ta [cos@, + +0]asT >Communication Systems - | 202 Signals and Systems The average power can also be obtained from the property of autocorrelation function ie. from equation 2.11.9 ie. P = Ry 2 A = = c0s(0) ro a 2 Which is same as we have derived earlier. The power spectral density and autocorrelation function are related by fourier transform pair. ie. from equation 21111. S(f) = FIRG) 2 [Some ‘ 7 o%, 2 =f 4 cso, tei? at 2 -o a’? ~j2eft Sif)=< J cso,te at 2 oO Ai = BEL t8 FHA This is obtained from equation 2.8.72 and o, =2 nf, ero a2 sy) =A [a(r-4)+8(F+2)] Exercise 1, Define autocorrelation function of energy signals and state and prove all its properties. 2. Define autocorrelation function of power signals and state and prove all its Properties. Unsoived Example 1. Find out autocorrelation function of x(f)=5+12 cos (2nf, t) where f, =10-Hz and hence find out its psd. {Ans, : R (1) = 25+72cos f@,1) S(f) = F(R (01 = 255 (fF) +36 8 (f ~ f) +365 (F +f.)aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.Communication Systems - | 204 Signals and Systems at T nT ) = cos(F) ets 0 otherwise Find : i) Autocorrelation function ii) Energy spectral density (10 Marks] Ans. : Refer to example 2.11.2. November - 99 Q.8 — Write short note on Autocorrelation function. [12 Marks] Ans. : Refer to Sec. 2.11. Q.9 Define auto correlation function for deterministic signals. Prove that power density spectrum and auto correlation function from a fourier transform pair. Explain significance of this relation. [6 Marks] Ans.: Refer to Q.2. December - 99 Q.10 Define autocorrelation function for time domain function x(t). What is significance of this function in communication systems ? Prove that power spectral density of x(t) and its autocorrelation form a ‘fourier transform pair. . [8 Marks] Ans.: Refer to Q.2 and Q. 4. May - 2000 Q.11 Find the time autocorrelation function of the signal g(t) =e" “'u(t), and from it determine the energy spectral density of g(t). Draw the necessary sketches for the functions. [8 Marks} Ans.: Refer to example 2.11.1 for solution. December - 2000 Q.12 Define power spectral density and autocorrelation functions. Show that power spectral density and auto correlation function form a.fourier transform pair. (8 Marks] Ans.: Power spectral density is defined in Sec 2.10.2. Autocorrelation function is defined. In Sec 2.11.2 for periodic signals. Relationship between autocorrelation function and psd is given by property 4 in Sec 2.11.2. December 2001 Q.13 Write ashort note on auto correlation function {4 markes} Ans. : Refer to Sec. 2.11. May-2002 Q 14 Find the time autocorrelation function of the signal g(1) =e” “'u(t), and from it determine the energy spectral density of g(t). Draw the necessary sketches for the functions. [8 Marks] Ans.: Refer to Q.11. 2.12 Cross Correlation Functions The autocorrelation function is used to measure the match between the signal and its time delayed version. The cross correlation is used to measure the match or similarity between two different waveforms. For cross correlation also second waveform is time delayed. The cross cortelation is defined separately for energyCommunication Systems - 205 Signals and Systems signals and periodic signals. The following two subsections are used to describe the definitions of crosscorrelation and its properties. 2.12.1 Cross Correlation of Energy Signals The two signals x, () and x, () is a pair of complex valued signals of finite energy. The cross correlation of these two signals is given as, Cross Correlation of Energy Signals : Ry) = J Ox @-0at =2 w 2121) From this relation we observe that the autocorrelation of energy signals has same type of relation. But the difference is that the two signals are different in cross correlation whereas they are same in autocorrelation. If x, (0 and x, () have some similarity, then the cross correlation function Ry, (t) will have some finite value over the Tange of t. If j x, 0x, Odt = 0 ie, wu (212.2) R (0) = 0, 12 then the signals x, (t) and x, (f) are called orthogonal signals. That is the two signals have no cross correlation between them. The second cross correlation function for x, (!) and x, (f) is given as, Ry) = i yx; ¢-a)dt w (2123) Properties of Cross correlation function for energy signals : Property 1: The cross correlation functions exhibits conjugate symmetry. ie. R,@) = Ry) vo» (212.4) That is unlike convolution, cross correlation is not in general commutative ie., Ry) # Ry Property 2: If R,, (0) = Die, J =, 0 4 Wat = 0, then the signals are said to be orthogonal over the entire time interval. Property 3 : The cross correlation of the two energy signals corresponds to the multiplication of the fourier transform of one signal by the complex conjugate of fourier transform of the second signal.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.Communication Systems - | 208 Signals and Systems December - 2002 Q3 Define following terms with suitable expressions. i) Autocorrelation and crosscorrelation functions. {4 Marks] 2.13 Frequency Response of LTI Systems The LTI systems form an important class in communication. The amplitude and phase response, realizability, bandwidth, distortion during transmission of signal are all very important concepts related to design and implementation of systems. 2.13.1 Frequency Response The frequency response of the system gives the variation of magnitude and phase of the system output with respect to frequency on application of input. We know that the output y (t) of the system is given as, y(t) = i x(t)h(t-1)dt w» (2.13.1) This equation gives time response of the LTI system. The RHS of the above equation represents convolution of input signal x (t) and impulse response h (t). By applying fourier transform to above equation, Fly) = ‘| f scone-oe] We know that convolution of two functions is transformed into multiplication of their fourier transforms. By applying this to above equation, F J sores] = X(f)H(f) Het: H(f) is called transfer function of the system, : Y(f) = H(f) X(f) .. (2.13.2) Thus ar a Linear Time Invariant system fourier transform of the aatpiat is equal to product of the transfer function of the system and fourier transform of the input. The above equation gives frequency response of the system. 2.13.2 Distortionless Transmission Through System A distortion less transmission means output of the system is an exact replica of the input signal. The difference between input and output of such system is that, 1. Amplitude of the output signal may increase or decrease by some factor with respect to input and 2 The output signal may be delayed in time with respect to input signal because of system delay. Therefore output signal y (#) can be written interms of input x (t) as,Communication Systems - | 209 Signals and Systems. y) = Kx@-t)) w» (2.13.3) Here, K = constant represents change in amplitude. &t, = time delay in transmission of signal through a system. By taking fourier transform of both sides of above equation YQ) = Fly] = F {Kx@-t)} . From the time shifting property of FT, eo itAtto YQ) = KX(f) w» (2.134) Transfer function H (0) is given from equation 2.13.2 Ye Hf) = Xf) Putting for RHS from equation 2.13.4 in above equation, Hy) = ke .. (2135) This equation gives the transfer function for a distortionless system. It is clear from above equation that, the magnitude of the transfer function is.’K’, which is independent of frequency. That is the transfer function has constant amplitude;at all : frequencies. The phase shift of above equation is, : of) = ca vs pp RI9895 -2nty) f ‘That is the phase shift is sont proportional to frequency. Here the phase shift is linear at all frequencies. This can be expressed with the example A Let there be a signal in time domain as x( = cos (Qn ft) Now let the output signal be same in amplitude but shifted in time by t, seconds. y= cos [2nf(t-to)l os (2.43.7) This equation can also be written as, y() = cos (Qn ft-2 nf ty) = cos (2nft—0(F)] Thus phase shift of y (t) is, O(f) = -2nft, which is proportional to frequency *f'aa You have either reached a page that is unavailable for viewing or reached your viewing limit for this book.Communication Systems - | Signals and Systems Fig. 2.13.2 (a) Magnitude response of ideal low-pass filter (0) Phase response of ideal low-pass fliter pert eit ay h® = “3B - j el2Mt—0) ap ae - 1 [alee j2n-t) a - 1 [erro _ osteo j2n(t-ty) a eltett tod _ gc i2Bt—) a) 27 = —1__ sin (25B(t-1,)] n(t-fy) 0 = op ( Sle nBt-to)l 2nB(t-to) = 2Bsinc [2B(t-t,)] w» (2:13:10) By definition of sinc functiort. The above equation gives impulse response of an ideal lowpass filter.Communication Systems - | 212 Signals and Systems Fig. 2.13.3 shows the impulse response h (t) of above equation of ideal low-pass filter. Fig. 2.13.3 Impulse response of ideal low-pass filter The above figure shows that impulse response exists for negative values of ‘’. But actually unit impulse is applied at t = 0 always. Thus the response appears before the unit impulse is applied. Practically it is impossible to implement such a system. In section 2.4.3 also we have stated in equation 2.4.7 that h (t) = 0 for t < 0 for causal system. Therefore it is clear that although ideal low pass filter is very desirable it cannot be physically realizable. Practically unit impulse to the filter is applied at t = 0 [ie. 3()=1 applied at t = 0], then impulse response of the filter should start at the most at > 0, and not at t < 0, ie. negative values of ’t’. Ex. 2.13.1 Determine the energy special density of a square pulse shown in Fig. 2.134 and calculate its energy. If the signal is passed through an ideal low-pass filter of bandwidth f, Hz determine the energy E,, of the output signal y (1). x(t) 1 Fig. 2.13.4 Sol. : The square pulse is represented by equation. t (t) = A rect | — x() = Are (3) Here Azl vet — ; x(t) = rect (#) The fourier transform of x (t) will be from equation 2.8.41,Communication Systems - | 213 Signals and Systems X(f) = T sinc(FT) Energy spectral density y (f) from equation 2.10.1 will be, vw) = 1X = |T sinc(fT)|? = T? sinc? (FT) wee (243.11) Energy of the signal is given as, E= f xp at Ti2 = [ ide -Pfz: =T The output and input energy spectral density functions are related by wolf) = |H(f)I? y, (Ff) This is written from property (ii) of energy spectral density function. Here yy (f) is output energy spectral density function. vy; (f) is input energy spectral density function. & H(f) is transfer function of ideal low-pass filter. v,(f) = T? sinc’ (fT) From equation 2.13.11 Hy) =e Po From equation 2.13.9 vo (f) = 6/1/01? sinc? (FT) Energy at the output will be, B= J vo W4af From equation (2.10.3) : The low-pass filter has pass-band from -f. to +f, jAnfly : B=? f e sinc” (fT) df “he _ 1? sine” (fT) [earn — itt) ~ j4nty += T? (2f,) sinc? (FT) sine (4 f, tg) Ex, 2.132 A rectangular pulse of unit amplitude and duration T is applied to an ideal lowpass filter of bandwidth B. Deiermine the response y (t) of the filter. Sol. : The output is given as,