Chapter 6 Passband Data Transmission

Similar documents
MSK has three important properties. However, the PSD of the MSK only drops by 10log 10 9 = 9.54 db below its midband value at ft b = 0.

Digital modulation techniques

Digital Modulation Schemes

DIGITAL COMMUNICATIONS SYSTEMS. MSc in Electronic Technologies and Communications

Thus there are three basic modulation techniques: 1) AMPLITUDE SHIFT KEYING 2) FREQUENCY SHIFT KEYING 3) PHASE SHIFT KEYING

Chapter 6 Passband Data Transmission

Modulation and Coding Tradeoffs

EC 6501 DIGITAL COMMUNICATION UNIT - IV PART A

Content. Chapter 6 PASSBAND DATA TRANSMISSION. Dr. Samir Alghadhban 11/22/13

Amplitude Frequency Phase

Chapter 4. Part 2(a) Digital Modulation Techniques

Principles of Communications

Department of Electronics and Communication Engineering 1

Wireless Communication Fading Modulation

3. 3. Noncoherent Binary Modulation Techniques

QUESTION BANK SUBJECT: DIGITAL COMMUNICATION (15EC61)

Objectives. Presentation Outline. Digital Modulation Revision

Principles of Communications

Digital Communication

Downloaded from 1

COSC 3213: Computer Networks I: Chapter 3 Handout #4. Instructor: Dr. Marvin Mandelbaum Department of Computer Science York University Section A

CSE4214 Digital Communications. Bandpass Modulation and Demodulation/Detection. Bandpass Modulation. Page 1

Detection and Estimation of Signals in Noise. Dr. Robert Schober Department of Electrical and Computer Engineering University of British Columbia

Fundamentals of Digital Communication

Mobile & Wireless Networking. Lecture 2: Wireless Transmission (2/2)

Chapter 6 Modulation Techniques for Mobile Radio

Outline / Wireless Networks and Applications Lecture 5: Physical Layer Signal Propagation and Modulation

ENSC327 Communication Systems 27: Digital Bandpass Modulation. (Ch. 7) Jie Liang School of Engineering Science Simon Fraser University

COMMUNICATION SYSTEMS

Digital Modulators & Line Codes

QUESTION BANK EC 1351 DIGITAL COMMUNICATION YEAR / SEM : III / VI UNIT I- PULSE MODULATION PART-A (2 Marks) 1. What is the purpose of sample and hold

Lecture 3: Wireless Physical Layer: Modulation Techniques. Mythili Vutukuru CS 653 Spring 2014 Jan 13, Monday

EXPERIMENT WISE VIVA QUESTIONS

UNIT TEST I Digital Communication

Digital Modulation Lecture 01. Review of Analogue Modulation Introduction to Digital Modulation Techniques Richard Harris

Objectives. Presentation Outline. Digital Modulation Lecture 01

Digital Communication System

AN INTRODUCTION OF ANALOG AND DIGITAL MODULATION TECHNIQUES IN COMMUNICATION SYSTEM

UNIT I Source Coding Systems

University of Manchester. CS3282: Digital Communications 06. Section 9: Multi-level digital modulation & demodulation


Revision of Wireless Channel

OptiSystem applications: Digital modulation analysis (FSK)

CHETTINAD COLLEGE OF ENGINEERING & TECHNOLOGY NH-67, TRICHY MAIN ROAD, PULIYUR, C.F , KARUR DT.

About Homework. The rest parts of the course: focus on popular standards like GSM, WCDMA, etc.

Spread Spectrum (SS) is a means of transmission in which the signal occupies a

Digital Communication System

MODULATION AND MULTIPLE ACCESS TECHNIQUES

Chapter 7 Multiple Division Techniques for Traffic Channels

Modern Quadrature Amplitude Modulation Principles and Applications for Fixed and Wireless Channels

Mobile Communications

Chapter 14 MODULATION INTRODUCTION

Problem Sheet 1 Probability, random processes, and noise

Channel Estimation in Multipath fading Environment using Combined Equalizer and Diversity Techniques

Lecture 9: Spread Spectrum Modulation Techniques

21. Orthonormal Representation of Signals

Physical Layer: Modulation, FEC. Wireless Networks: Guevara Noubir. S2001, COM3525 Wireless Networks Lecture 3, 1

Basic Concepts in Data Transmission

Signal Encoding Techniques


ECE 4600 Communication Systems


UNIVERSITY OF CALIFORNIA College of Engineering Department of Electrical Engineering and Computer Sciences EECS 121 FINAL EXAM

EE 460L University of Nevada, Las Vegas ECE Department

Mobile Radio Systems OPAM: Understanding OFDM and Spread Spectrum

CHETTINAD COLLEGE OF ENGINEERING & TECHNOLOGY NH-67, TRICHY MAIN ROAD, PULIYUR, C.F , KARUR DT.

SEN366 Computer Networks

Point-to-Point Communications

Swedish College of Engineering and Technology Rahim Yar Khan

CHAPTER 2. Instructor: Mr. Abhijit Parmar Course: Mobile Computing and Wireless Communication ( )

ECE5713 : Advanced Digital Communications

EC6501 Digital Communication

EE5713 : Advanced Digital Communications

Fund. of Digital Communications Ch. 3: Digital Modulation

Lecture 3 Digital Modulation, Detection and Performance Analysis

Year : TYEJ Sub: Digital Communication (17535) Assignment No. 1. Introduction of Digital Communication. Question Exam Marks

Chapter 3 Communication Concepts

Revision of Lecture 3

CHAPTER 2 DIGITAL MODULATION

Syllabus. osmania university UNIT - I UNIT - II UNIT - III CHAPTER - 1 : INTRODUCTION TO DIGITAL COMMUNICATION CHAPTER - 3 : INFORMATION THEORY

Receiver Designs for the Radio Channel

EE3723 : Digital Communications

Universitas Sumatera Utara

GOPALAN COLLEGE OF ENGINEERING AND MANAGEMENT Electronics and communication Department

UNIVERSITY OF BAHRAIN COLLEGE OF ENGINEERING DEPARTMENT OF ELECTRICAL AND ELECTRONIC ENGINEERING

SPREAD SPECTRUM (SS) SIGNALS FOR DIGITAL COMMUNICATIONS

Columbia University. Principles of Communication Systems ELEN E3701. Spring Semester May Final Examination

28. What is meant by repetition rate of the AM envelope? (ADC,AU-2010) 29. Describe the upper and lower sidebands. (ADC, AU-2010) 30.

Channel & Modulation: Basics

Problems from the 3 rd edition

TELE4652 Mobile and Satellite Communications

Exercises for chapter 2

Other Modulation Techniques - CAP, QAM, DMT

Time division multiplexing The block diagram for TDM is illustrated as shown in the figure

CALIFORNIA STATE UNIVERSITY, NORTHRIDGE FADING CHANNEL CHARACTERIZATION AND MODELING

ISHIK UNIVERSITY Faculty of Science Department of Information Technology Fall Course Name: Wireless Networks

Communication Systems

EE3723 : Digital Communications

Chapter 7. Multiple Division Techniques

Theory of Telecommunications Networks

Transcription:

Chapter 6 Passband Data ransmission Different methods of digital modulation Outline PSK(Phase-shift keying), QAM(Quad. amp. mod), FSK(Phase-shift keying) Coherent detection of modulated signals in AWGN Carrier phase; Bit timing Noncoherent detection of modulated signals in AWGN: phase information is disregarded Modems for transmission and reception of digital data over PSN (public switched telephone network) Sophisticated modulation techniques over a wideband channel with medium to severe ISI Carrierless amplitude/phase modulation Discrete multitone echniques for synchronizing receiver to transmitter 6 -

Introduction Baseband Pulse transmission: a discrete pulse-amplitude modulation (PAM) is transmitted over a low-pass channel Digital passband transmission Data stream is modulated into a carrier with fixed frequency limits imposed by a band-pass channel of interest. Major issue is the optimum design of the receiver in the presence of noise. Communication channel: microwave radio, satellite Modulation processswitching (keying) amplitude, phase, frequency hree basic signaling schemes: ASK(Amp.-shift keying), PSK(Phase-shift keying), FSK(Phase-shift keying) > Unlike continuous-wave modulation, it is not difficult to distinguish between PSK and FSK signals. > Unlike ASK signals, PSK and FSK signals have a constant envelopeimperious to amplitude nonlinearities (microwave radio, satellite channels)preferred to ASK for passband data transmission over nonlinear channels. 6 - Modulation schemes ASK(Amp.-shift keying) PSK(Phase-shift keying) FSK(Phase-shift keying) 6-3

Hierarchy of digital modulation M-ary schemes: conserve bandwidth at expense of increased power M-ary modulation techniques: M-ary PSK, FSK, ASK Hybrid form: M-ary amplitude-phase keying (APK) special case: M-ary quadrature-amplitude modulation (QAM)special case: M-ary ASK Linear modulation: M-ary PSK, M-ary QAM > M-ary PSK: constant envelopenonlinear band-pass channel > M-ary PSK: changes in carrier amplitudenonlinear band-pass channel Classification of digital modulation techniques: the receiver is equipped with a phase-recovery circuit or not Coherent modulation techniques > M-ary PSK, QAM, FSK Noncoherent modulation tech. (no carrier phase information is needed) > ASK, FSK: impractical to maintain carrier phase synchronization > Differential phase-shift keying (DPSK): noncoherent form of PSK 6-4 Power Spectra o study power spectra of resulting modulated signals is particularly important in two contexts - Occupancy of channel bandwidth - Cochannel interference in multiplexed systems Bandpass signal (modulated signal) complex envelope s t s I t cosf c t s Q t sinf c t Res t expjf c t s t s I t + js Q t S B f: power spectral density of complex envelope s t Baseband power spectral density Power spectral density of s t S s f ----- 4 S B f + S B f + Bandwidth efficiency: ration of data rate in bits per second to effectively utilized channel bandwidth R b ---- bits/s/hz - multilevel encoding; spectral shaping B f c amp. f c power / /4 6-5

Passband ransmission Model Bandpass communication channel - Channel is linear: bandwidth is wide enoughs i (t) no distortion - Channel noise w(t): white Gaussian with zero mean and power spectral density / reverses operations performed in transmitter Minimize effect of channel noise Linear AWGN 6-6 Binary Phase-Shift keying (rror rate) Binary Phase-shift keying (BPSK) s t b ------- cosf c t s t b ------- cosf c t t ---- cosf c t f c n c t f Xj x j m i ------------ ---- x j s ij N i, exp o j f X x ------------ b exp ---- x j s N z --------- x + b o p f X x x d ------------ exp ---- x + b dx N o ------ exp z d z ----erfc b b p p P e P e p e p p + p p -----erfc ---- b n c : fixed integer s s s t t s t t ------- b A b s A ---- b P e s b b signal-space diagram for coherent BPSK;n c transmitted signal energy per bit 6-7

Binary Phase-Shift keying (Power spectra) in-phase component +g(t), -g(t) g t b, t b, otherwise Power spectral density ---- G f s G f b sinc f G f b sin c f S B f b sin c f b sin f ----------------------------------------- f falls off ~ inverse square of frequency BW(PSK) < BW (MSK or FSK) Sidelobe(PSK) > Sidelobe(FSK) Binary PSK transmitter Coherent BPSK receiver 6-8 Quadrature Phase-Shift keying (QPSK) Goal in the design of digital communication: Low probability of error efficient utilization of channel bandwidth. BPSK QPSK: N, M4 s i t ----- cos f c t + i -----, t 4, elsewhere ----- cos i ----- cosf c t 4 ----- sin i ----- sinf c t 4 orthonormal basis functions t --- cos f c t t --- sin f c t t i 3 4 signal vector cos i ----- 4 s i sin i ----- 4 ----- A A ---- : symbol duration; : nergy/symbol f c n c / for some fixed integer n c Gray-encoded set of debits:,,, Gray-code 6-9

QPSK received signal Observation vector X in-phase x Quadrature x x t t + w x t t + w rror Probability of QPSK x t s i t + w t bit error in in-phase and Quadrature channels are statistically independent P c P ' average prob. of a correct decision Symbol error probability if» P e P c erfc ------- ----erfc ------- 4 erfc --------- BR ----erfc BR (BPSK) BR (QPSK) with the same b /, but BW(QPSK) / BW(BPSK) white Gaussian noise with power spectral density / wo BPSK with b / and power spectral density / -------- b P b -----erfc ---- P ' ----erfc ---------- BPSK QPSK s s bit error rate for each channel 6 - Binary sequence is divided into two other sequences - wo waveforms may individually be viewed as a BPSK signal Generation and Detection of QPSK QPSK ransmitter QPSK Receiver 6 -

Symbol shaping function g t in-phase Quadrature power spectra, t, otherwise G f sincf S BI S BQ Power Spectra of QPSK ----- G f sin cf S B f S BI + S BQ sin cf 4 b sin cf in-phase and Quadrature components have a common power spectra in-phase and Quadrature components are statistically independent BW(PSK) < BW (MSK or FSK) Sidelobe(PSK) > sidelobe (MSK or FSK) no spike in spectra 6 - M-ary Phase-Shift keying (M-ary PSK) ----- cos f s i t c t + i --, t M, elsewhere i i i M ----- A A ---- M : symbol duration; : nergy/symbol f c n c / for some fixed integer n c M message points are equally spaced on a circle of radius for the case of M 8 d d 8 sin -- M Average symbol error P e f X x j x d erfc sin -- M P e d erfc ---- sin -- M (M-ary PSK) M4 8-PSK he approximation becomes extremely tight, for fixed M, as / is increased. - For M 4, the same form for QPSK 6-3

rror Probability of M-ary PSK r t Acosf c t + k + n x t cosf c t n y t sinf c t Received M-ary PSK P e 8 Pn y Asin --+ P ny Asin 8 -- P Common Area 8 P n y A sin-- + P ny A sin-- 8 8 --erfc Asin 8 + ---------------------- n y --erfc ---------------------- Asin 8 n y erfc -------- A sin -- erfc sin 8 -- 8 erfc sin-- P M e M M ---P e 8 erfcsin -- M n 8-ary PSK signal-to-noise radio ----- S N S A N n B quivalent noise bandwidth as B for integrate-and-dump circuit, matched filter and correlator S S S ----- ------- ----- N B ---- s Q 4 n y n y n A x A sin 8 8 I nx ny n B n t n x t + jn y t Common Area B 6-4 Power Spectra of M-ary PSK S B f sin cf b log Msin c flog M normalized power spectral density S B S B f b log M : symbol duration Channel bandwidth required to pass M-ary PSK signal (main lobe, null-to-null) / Bandwidth efficiency (spectral efficiency): the ratio of data rate to channel bandwidth fficient modulation maximize bandwidth efficiency - achieve this bandwidth at a minimum of average signal power or average SNR Bandwidth efficiency R b ---- B B ----- --------------------- log M Data rate R b (bits/s/hz) Channel bandwidth R ------------- b log M f b vs normalized frequency f R b ---- B log M --------------- M is increased, bandwidth efficiency is improved at the expense of error performanceincrease b / f 6-5

M-ary Quadrature Amplitude Modulation (QAM) M-ary QAM is a two-dimensional generalization of M-ary PAM a k b k 3 5 s i t ------- a k cosf c t ------- b k sinf c t t k orthonormal basis functions : nergy of signal with the lowest amplitude ( t --- or ) channel cos f c t t d min ------- A wo distinct QAM constellations - Square constellations: number of bits per symbol is even - Cross constellations: number of bits per symbol is odd minimum distance between any two message points in constellation Amplitude for I channel cosf c t or Q Channel sinf c t d min t --- sin f c t M-QAM for M 6 with Gray-encoded t corresponding 4-PAM 6-6 rror probability of Square QAM With an even number of bits per symbol, L M (positive integer) - M-ary QAM square constellation Cartesian product of a one-dimensional L-ary PAM probability of symbol error probability of correct detection for M-ary QAM P c P e ' P e ' : probability of symbol error for corresponding L-ary PAM P e ' -------- M erfc ---- P e P c P e ' P e ' probability of symbol error for M-ary QAM -------- erfc ---- M average value of transmitted energy symmetric L av ------- i L f i, f M --------------------- 3 Average Symbol nergy P e -------- 3 erfc ------------------------ av M M M-QAM for M 6 with Gray-encoded corresponding 4-PAM 6-7

rror probability of Cross QAM Cross constellations: number of bits per symbol is odd d Construct a signal constellation with n bits per symbol - A square constellation with n - bits per symbol - xtend each side of square constellation by adding n-3 symbols - Ignore corners in the extension -3d -d -d d 3d x: 4x n-3 n- n- + n- n, n bits per symbol It is not possible, it is not possible to express a QAM cross constellation as product of a PAM constellation - It is not possible to perfectly Gray code a QAM cross constellation - Complicates determination of symbol error probability. P e --------- M erfc ---- o N 6-8 rror Probability of M-ary QAM M-ary PAM r t a k cosf c t+ n x t cosf c t n y t sinf c t -3d -d d M P e ----------------P n y d M -- M P n M + y d -----------P n y d M M -----------erfc ---------- d M -----------erfc S ----- ------------------- 3 M -----------erfc M M M M --- S ------------------- 3 N M a M k ---- M m d ------------------------- M d P AV S s t m 3 -- M d ------------------------- 3 S S d P e ---- erfc --- --- ----- 3 S -PAM 4-PAM P N N e --erfc ------ 4 5N M-ary QAM r t a k cosf c t b k sinf c t+ n x t cosf c t n y t sinf c t P e ------- d M erfc ---------- P AV S s t d -------- S M erfc ------- 3 N ---------------- ---------------------- M d 3 M 4-QAM P e erfc ------ S N --- S N ----- d -3d -d -d d --- S N 3d -------- 5d 3d 6-9

Coherent Binary Frequency-Shift keying (BFSK) Symbols and are distinguished from each other by transmitting one of two sinusoidal waves that differ in frequency by a fixed amount Sunde s FSK: continuous-phase signal phase continuity is always maintained including inter-bit switching timescontinuous-phase frequency-shift keying (CPSK) i ------- b cosf s i t i t, t b, elsewhere b : transmitted signal energy per bit n f + c i i ------------- n c : integer i i t ------ cosf i t s b, i j ij, ij M, N i j ------- b A b A ---- 6 - rror Probability of BFSK s b s b d Y X + X b Y X + X b Var Y Var X + Var X f y y --------------- exp ------- y + b p Py symbol f y y y d b x x symbol : symbol : y + --------------- b exp ------------------------- dy N o ------ z exp d z ----erfc --------- b b P e -----erfc ------- b P e BFSK P e BPSK d b b N d x t t x t t x x y x x symbol : x x y symbol : x x y N --- o N + --- --- o 4 4 b x x BPSK b 6 -

Generation and Detection of Coherent BFSK On-off level encoder: - volts; volts b Inverter: - f on, f off; f off, f on f i and f are chosen to equal different integer multiples of bit rate / - f i and f are synchronized - A single keyed (OSC) oscillator - modulated wave is shifted he detector consists two correlators - correlator outputs are subtracted - y > symbol y < symbol BFSK transmitter BFSK Receiver 6 - s t b ------- cos f c t----- t indep. of input binary wave Power Spectra of BFSK ------- b ----- t cos cosf c t ------- b ----- t sin sin f c t for all f b ------- f ------- + f + ------- t g f 8 bcos f ------------------------------------------ 4 b f t f f S B f b ------- f ------- + f + ------- 8 b cos f f f ------------------------------------------ 4 b f f 4 FSK (continuous phase) falls off ~ f -4 does not produce as much outside signal of interest FSK (Discontinuous phase) falls off ~ f - - f and f operate independently FSK has a smoother pulse shape and lower sidelobes than PSK Falls off ~ f -4 Smoother Lower sidelobes 6-3

Coherent Minimum-Shift-Keying (MSK) Coherent detection of BFSK - phase information is not fully exploited - other than to provide for synchronization of receiver and transmitter - proper use of phaseimprove noise performance of receiver - his improvement is achieved at expense of increased receiver complexity Sunde s FSK - Deviation ratio is exactly unity (f -f / ) - Phase change over one bit interval is radians - here is no memorychange occurred in previous bit interval provides no help in current bit interval waveform of MSK signal 6-4 Coherent MSK vs CFSK Continuous-phase frequency-shift keying (CFSK) ------- b cosf t + symbol s t t ------- b cosf t + symbol Another useful way of representing CFSK s t b ------- cosf c t + t continuous function of time including switching time t -----t h f c f c h + ------- f h ------- f b + symbol ;symbol f c -- f + f h f f Deviation ratio h symbol h symbol t phase Continuity Sunde s FSK h h / Phase rellis even odd 6-5

Signal-Space Diagram of MSK h /, frequency deviation, difference between two signaling frequencies f and f, equals half bit rate. f f. - Minimum frequency spacing allows two FSK signals representing symbols and ; coherently orthogonal - CPFSK with a deviation ratio of one half minimum shift keying (MSK) s t b ------- cost cos f c t ------- b sint sin f c tt, -------t, t b or h / s I t b ------- cost ------- b cos cos -------t ------- b cos -------t t half-cycle cosine pulse b s Q t b ------- sint ------- b sin b sin -------t ------- b -------t cos t half-cycle sine pulse b, b symbol, b symbol, b symbol, b symbol 6-6 s I t b ------- cos -------t t s Q Signal-Space Diagram of MSK t b ------- cos -------t t s t b ------- cost cos f c t ------- b sint sin f c t t ---- cos t ---- sin -------t -------t cosf c t sinf c t t s t s t + s t t integral for time interval s s t t b cos t s s t t b cos t 6-7

rror Probability of MSK, b symbol, b symbol, b symbol, b symbol Both integrals for a interval Lower and upper bound for s and s - s shifted by to s t is common to both integrals - and b are defined Average Probability of errors x x t t s + w, N t x x t t s + w, t b P e -----erfc ----- he same as PSK, QPSK, Detection ~ observation over 6-8 Generation and Detection of MSK s t s t + s t, t t ---- cos -------t cosf c t b t ---- -------t sin sinf c t b wo phase-coherent sinusoidal waves at f and f and for h / wo narrow-band filters orthonormal basis functions t and t. a and a : bit rate / h f c + -------- f h f c -------- f t integration interval t 6-9

Power Spectra of MSK s t b t ------- cos------- cos f c t b t ------- sin------- sin f c t g I t g Q t g I t, gi f ---------------- 3 b cos f g Q t ------------------------------ 6 b f gq f g f S B f g f --------------- 3 b ---------------- b cos f ------------------------------ 6 b f MSK produces less interference than PSK - MSK falls off ~ f -4 PSK ~ f - - he desired characteristics of MSK especially when operates with a bandwidth limitation GMSK: satisfy the stringent requirements of certain applications such as wireless communication normalized to 4 b t t Falls off ~ f -4 Smoother Lower sidelobes 6-3 Gaussian-Filtered MSK (GMSK) Desirable properties of MSK - Constant envelope - Relatively narrow bandwidth - Coherent detection performance equivalent to that of QPSK Power spectrum a compact form: premodulation low-pass filter (Pulse-shaping filter) - narrow bandwidth; sharp cutoff - impulse resp. with relatively low overshoot - volution of a phase trellis MSK Pulse-shaping filter Gaussian function Adjacent channel interference (ADJ) of wireless communication system using MSK is not low enough H f exp ----------- log -- f W frequency-shaping pulse g t log 9 9.54dB h t ----------- W exp ----------- W t log log g t ----------- W log exp -----------W log t shifted in time by.5 6-3

Power spectra of GMSK Frequency-shaping pulse g t ---- t h t ht d ----------- Wexp ----------- W log log t ime-bandwidth product (W ) - play role of a design parameter - W is reduced, time spread of frequency-shaping pulse is increased - W MSK - W more of transmit power is concentrated inside passband of GMSK Undesirable feature of GMSK - signal is no longer confined to a single bit intervalpulse spreadisi Power spectra of MSK and GMSK for varying time-bandwidth product (W ) 6-3 rror Probability of GMSK ISI increases with decreasing W a trade-off between spectral compactness and performance loss is a constant whose value depends on time-bandwidth product W - log (/) in db: a measure of performance degradation compared to MSK - W MSK W.3,performance degradation ~.46 db - a small price to pay for highly desirable spectral compactness of GMSK signal - An important applicationgsm rror probability of GMSK in the presence of AWGN W.46.3 P e ----erfc --------- b a small price to pay for highly desirable spectral compactness 6-33

GMSK for GSM Wireless Communications GSM: An important application of GMSK in a standardized wireless communication system; a time-division multiple-access system; W.3 - he best compromise between increased BW occupancy and resistance to CCI - 99% percent of power is confined to BW 5 khz, sidelobe~ outside this band CCI < 4dB Spectrum khz-wide subchannels - each subchannel at 7 kb/s - RF power spectrum (shaded subchannel) is down by an amount larger than 4dB at both adjacent subchannelseffect of CCI is practically negligible Power spectrum of GMSK signal for GSM wireless communications 6-34 Coherent M-ary FSK i M s i t ----- cos --n c + it, t P e --- M erfc ------- N good approximation for P e 3 Signal frequencies are separated by / M-ary FSK channel bandwidth B log M R b R B b M R ----------------- ---- b log M B log M ----------------- M M ------ M-ary PSK is spectrally efficient; M- ary PSK is spectrally inefficient - PSK: increase Mincrease - PSK: increase Mdecrease s i ts j tt d ij i t --------s i t n n ----- c c : integer minimum distance d min f c Spectral analysis of M-ary FSK signals is much more complicated particular case frequency deviation k.5; M signal frequencies are separated by / Power spectra of M-ary FSK signals for M,4,8 6-35

Optimum Quadratic Receiver Coherent detectionassumptions - Perfectly synchronized to transmitter - only channel impairment is noise Uncertainty due to randomness - distortion in transmission medium - common parameter is carrier phase - especially true for narrow signals Synchronization with phase may be too costly - Disregard phase information at expense of degradation in performance noncoherent wo equivalent forms of Quadratic Receiver - quivalent matched filter - Noncoherent matched filter noncoherent Quadrature receiver using correlators noncoherent Quadrature receiver using matched filters Noncoherent matched filter 6-36 Noncoherent Orthogonal Modulation Noncoherent Orthogonal Modulation - noncoherent BFSK - Differential PSK (DPSK) ransmitted signal s i t ----- cosf i t, Received signal f x t ----- cosf i t + + w t Noncoherent Binary FSK and DPSK x t g t + w t, g t + w t, s tsent s tsent i t t Binary receiver for noncoherent orthogonal modulation i t andˆi t are orthogonal to each other i t m t cosf i t ˆi t m t sinf i t m(t) is a bandlimited message P e -- exp -------- Quadrature receiver equivalent to either one of two matched filters 6-37

rror rate of Noncoherent BFSK and DPSK BFSK signal s i t -------- cosf i t, upper matched to lower matched to cosf t cosf t f i n i t b P e -- exp ------- DPSK signal s t symbol phase unchanged -------- cosf c t -------- cosf c t,, t t s t symbol phase unchanged -------- cosf c t -------- cosf c t +,, t t Noncoherent receiver for detection of BFSK DPSK is a special case of noncoherent orthogonal modulation with b b P e -- exp ---- 6-38 Generation and Detection of DPSK DPSK is noncoherent version of PSK - Incoming binary symbol b k is, leave symbol d k unchanged with respect to previous bit - Incoming binary symbol b k is, change symbol d k with respect to previous bit DPSK transmitter Signal-space diagram of received DPSK signal 6-39 DPSK receiver

Comparison of Digital Modulation schemes (Probability of error) BR decrease monotonically with increasing b /N - curves ~ shape in the form of a waterfall Coherent binary PSK, QPSK, MSK produce a smaller BR than any of other schemes b /N (coherent) b /N (noncoherent)3db less - BPSK vs BFSK; DPSK vs BFSK (incoh.) At high b /N, coherent db less than noncoherent 3dB - BPSK vs DPSK - BFSK (coh) vs BFSK (incoh.) db 6-4 Comparison (Bandwidth fficiency) Power-bandwidth for coherent PSK - QPSK: best trade-off bet. power and bandwidth - M > 8: excessive power; complex equipment M-ary QAM is better than M-ary PSK for M >4 - QAM can be realized if channel is linear M-ary FSK: increasing M reduced power requirement increased channel bandwidth M6 signal constellations of M-ary PSK 6-4 M-ary QAM