Digital Transmission

Similar documents
Calculation of the received voltage due to the radiation from multiple co-frequency sources

Phasor Representation of Sinusoidal Signals

Rejection of PSK Interference in DS-SS/PSK System Using Adaptive Transversal Filter with Conditional Response Recalculation

Revision of Lecture Twenty-One

Passive Filters. References: Barbow (pp ), Hayes & Horowitz (pp 32-60), Rizzoni (Chap. 6)

Adaptive Modulation for Multiple Antenna Channels

Section 5. Signal Conditioning and Data Analysis

To: Professor Avitabile Date: February 4, 2003 From: Mechanical Student Subject: Experiment #1 Numerical Methods Using Excel

A study of turbo codes for multilevel modulations in Gaussian and mobile channels

Space Time Equalization-space time codes System Model for STCM

Review: Our Approach 2. CSC310 Information Theory

1 GSW Multipath Channel Models

The Performance Improvement of BASK System for Giga-Bit MODEM Using the Fuzzy System

Parameter Free Iterative Decoding Metrics for Non-Coherent Orthogonal Modulation

antenna antenna (4.139)

Multicarrier Modulation

IRZ0140 Signals and Signal Processing

Walsh Function Based Synthesis Method of PWM Pattern for Full-Bridge Inverter

TECHNICAL NOTE TERMINATION FOR POINT- TO-POINT SYSTEMS TN TERMINATON FOR POINT-TO-POINT SYSTEMS. Zo = L C. ω - angular frequency = 2πf

Abstract. 1. Introduction. 2. OFDM Signal. where

High Speed ADC Sampling Transients

Test 2. ECON3161, Game Theory. Tuesday, November 6 th

POLYTECHNIC UNIVERSITY Electrical Engineering Department. EE SOPHOMORE LABORATORY Experiment 1 Laboratory Energy Sources

IEE Electronics Letters, vol 34, no 17, August 1998, pp ESTIMATING STARTING POINT OF CONDUCTION OF CMOS GATES

ECE 2133 Electronic Circuits. Dept. of Electrical and Computer Engineering International Islamic University Malaysia

A Comparison of Two Equivalent Real Formulations for Complex-Valued Linear Systems Part 2: Results

Index Terms Adaptive modulation, Adaptive FEC, Packet Error Rate, Performance.

Bit Error Probability of Cooperative Diversity for M-ary QAM OFDM-based system with Best Relay Selection

Analysis of Time Delays in Synchronous and. Asynchronous Control Loops. Bj rn Wittenmark, Ben Bastian, and Johan Nilsson

Figure 1. DC-DC Boost Converter

Low Switching Frequency Active Harmonic Elimination in Multilevel Converters with Unequal DC Voltages

NOVEL ITERATIVE TECHNIQUES FOR RADAR TARGET DISCRIMINATION

Novel Sampling Clock Offset Estimation for DVB-T OFDM

Chapter 13. Filters Introduction Ideal Filter

Power Minimization Under Constant Throughput Constraint in Wireless Networks with Beamforming

NATIONAL RADIO ASTRONOMY OBSERVATORY Green Bank, West Virginia SPECTRAL PROCESSOR MEMO NO. 25. MEMORANDUM February 13, 1985

Performance Study of OFDMA vs. OFDM/SDMA

RC Filters TEP Related Topics Principle Equipment

... -J to send radio signals through the air, how modulation , IJ."~ UNITED STATES~~ FREQUENCY ALLOCATIONS

AN IMPROVED BIT LOADING TECHNIQUE FOR ENHANCED ENERGY EFFICIENCY IN NEXT GENERATION VOICE/VIDEO APPLICATIONS

Multipath Propagation. Outline. What is OFDM? (OFDM) for Broadband Communications and. Orthogonal Frequency Division Multiplexing

UNIT 11 TWO-PERSON ZERO-SUM GAMES WITH SADDLE POINT

A Current Differential Line Protection Using a Synchronous Reference Frame Approach

Hierarchical Generalized Cantor Set Modulation

LOCAL DECODING OF WALSH CODES TO REDUCE CDMA DESPREADING COMPUTATION

THE GENERATION OF 400 MW RF PULSES AT X-BAND USING RESONANT DELAY LINES *

Learning Ensembles of Convolutional Neural Networks

THE USE OF CONVOLUTIONAL CODE FOR NARROWBAND INTERFERENCE SUPPRESSION IN OFDM-DVBT SYSTEM

ROYAL INSTITUTE OF TECHNOLOGY KUNGL TEKNISKA HÖGSKOLAN. Department of Signals, Sensors & Systems Signal Processing S STOCKHOLM

Comparative Analysis of Reuse 1 and 3 in Cellular Network Based On SIR Distribution and Rate

Design of an FPGA based TV-tuner test bench using MFIR structures

A thesis presented to. the faculty of. the Russ College of Engineering and Technology of Ohio University. In partial fulfillment

EE 508 Lecture 6. Degrees of Freedom The Approximation Problem

Prevention of Sequential Message Loss in CAN Systems

BER Performance Analysis of Multiuser Diversity with Antenna Selection in MRC MIMO Systems

@IJMTER-2015, All rights Reserved 383

Markov Chain Monte Carlo Detection for Underwater Acoustic Channels

SECURE COMMUNICATION BY USING MULTIPLE KEYS HAVING VARIABLE LENGTH IN A REAL TIME ENVIRONMENT FOR MULTIPLE STATIONS

Malicious User Detection in Spectrum Sensing for WRAN Using Different Outliers Detection Techniques

Power spectral density estimation for wireless fluctuation enhanced gas sensor nodes

Keywords LTE, Uplink, Power Control, Fractional Power Control.

PRACTICAL, COMPUTATION EFFICIENT HIGH-ORDER NEURAL NETWORK FOR ROTATION AND SHIFT INVARIANT PATTERN RECOGNITION. Evgeny Artyomov and Orly Yadid-Pecht

MTBF PREDICTION REPORT

64-QAM Communication System using Three-beam Spatial Power Combining Technology

29. Network Functions for Circuits Containing Op Amps

Direct Sequence Spread Spectrum (DSSS)

ECE315 / ECE515 Lecture 5 Date:

Optimizing Transmission Lengths for Limited Feedback with Non-Binary LDPC Examples

RESOURCE CONTROL FOR HYBRID CODE AND TIME DIVISION SCHEDULING

An Improved Profile-Based Location Caching with Fixed Local Anchor Based on Group Deregistration for Wireless Networks

Figure 1. DC-DC Boost Converter

Power Control for Wireless Data

Performance Analysis of Multi User MIMO System with Block-Diagonalization Precoding Scheme

Guidelines for CCPR and RMO Bilateral Key Comparisons CCPR Working Group on Key Comparison CCPR-G5 October 10 th, 2014

arxiv: v1 [cs.it] 30 Sep 2008

AN ADVANCED ALGORITHM FOR IMPROVING DVB-T COVERAGE IN SFN. V. Mignone, A. Morello, M. Visintin. RAI Research Centre, Italy ABSTRACT

4.3- Modeling the Diode Forward Characteristic

Control Chart. Control Chart - history. Process in control. Developed in 1920 s. By Dr. Walter A. Shewhart

A NOVEL PREAMBLE DESIGN FOR CHANNEL ESTIMATION IN MIMO- OFDM SYSTEMS RESULTING IN ENHANCED THROUGHPUT

A Fuzzy-based Routing Strategy for Multihop Cognitive Radio Networks

Novel Techniques of RF High Power Measurement

MASTER TIMING AND TOF MODULE-

Encoding of passive anticollision radio-frequency identification surface acoustic waves tags

Error Probability of RS Code Over Wireless Channel

Secure Transmission of Sensitive data using multiple channels

EEC173B/ECS152C, Winter Fundamentals of Wireless Communications

Energy Efficient Adaptive Modulation in Wireless Cognitive Radio Ad Hoc Networks

Unit 1. Current and Voltage U 1 VOLTAGE AND CURRENT. Circuit Basics KVL, KCL, Ohm's Law LED Outputs Buttons/Switch Inputs. Current / Voltage Analogy

High Speed, Low Power And Area Efficient Carry-Select Adder

Joint Power Control and Scheduling for Two-Cell Energy Efficient Broadcasting with Network Coding

Chapter 1. On-line Choice of On-line Algorithms. Yossi Azar Andrei Z. Broder Mark S. Manasse

Underwater Acoustic Networks: Channel Models and Network Coding based Lower Bound to Transmission Power for Multicast

Harmonic Balance of Nonlinear RF Circuits

Understanding the Spike Algorithm

A Simple Satellite Exclusion Algorithm for Advanced RAIM

A new family of linear dispersion code for fast sphere decoding. Creative Commons: Attribution 3.0 Hong Kong License

Uncertainty in measurements of power and energy on power networks

On Timing Offset and Frequency Offset Estimation in LTE Uplink *

On Interference Alignment for Multi-hop MIMO Networks

EE301 AC Source Transformation and Nodal Analysis

Transcription:

Dgtal Transmsson Most modern communcaton systems are dgtal, meanng that the transmtted normaton sgnal carres bts and symbols rather than an analog sgnal. The eect o C/N rato ncrease or decrease on dgtal transmssons appears n the orm o errors n the transmtted dgtal normaton. Derence between a ymbol and a Bt symbol s a pulse wth duraton T seconds (ths s a perod wth unts o seconds not temperature as was used beore) that may contan a bt or more o data. For: a symbol that takes one o 2 values, each symbol contans 1 bt T = T B & = B a symbol that takes one o 4 values, each symbol contans 2 bts T = 2T B & = B /2.. symbol that takes one o M = 2 N values, each symbol contans N bts. T = N T B & = B /N There s an nherent advantage to transmttng data usng symbols wth large values o M. The advantage s that more data s transmtted by usng the same BW. That s, the element that determnes the bandwdth n dgtal transmsson s number o transtons between symbols not the number o levels over whch we change. o, two sgnals wth 10 changes per second but one o them changes between 2 levels whle the other changes among 10 levels wll requre the same bandwdth. Types o gnalng (Lne codes) The realty s that normaton sources that provde dgtal data actually provde numbers that are not sutable or transmsson on any channel as they are. n analog to dgtal converter (DC) that converts an analog audo sgnal to a dgtal ormat provdes sample values every 50 to 100 mcroseconds. The sample values must be ormatted n a proper way to make them sutable or transmsson through the communcaton channel. When each dgtal value s represented usng a pulse or each bt (n bnary communcaton) or a pulse or multple bts smultaneously (or M-ary communcaton), the sgnal generated by stackng derent pulses s called a lne code. To put t n a smple way, the process o lne codng s usng the dgtal data obtaned rom an normaton source to

generate a voltage sgnal that represents the normaton. There are derent orms o lne codes that can be used to represent the normaton. The terms Return to Zero (RZ) and Non Return to Zero (NRZ) wll be used n descrbng these sgnal. ome lne codes are shown next along wth ther power spectral denstes, advantages and dsadvantages. We wll consder only bnary lne codes although many lnes codes can work wth bnary and M-ary transmssons: 1. Unpolar (On O) Non-Return to Zero (NRZ): In ths orm o lne codes, a bt o 1 s represented by some postve voltage (+5 volts or example) and a bt o 0 by 0 volts (justyng callng ths sgnal On O). The pulses correspondng to bnary 1 reman at the postve voltage or the whole duraton o the bt perod (t does not return to zero at any tme durng the bt perod justyng callng ths lne code NRZ). T 0 R 1 2 2T = R ω dvantages: 1. Very smple to generate (has only two levels whch can be easly generated usng smple dgtal electroncs) 2. pectrally ecent (requres the mnmum amount o bandwdth or a specc bt rate) Dsadvantages: 1. Does not provde any orm o bt or rame synchronzaton (or long sequences o ones or zeros, the transmtter and recever may get unsynchronzed) 2. Does not provde any orm o error detecton. 3. Generally has non-zero DC (even voltages o -5 V and +5 V are used, unless the number o zeros and number o ones are equal). Ths may be a problem or some communcaton systems that cannot transmt DC values. 2. Bpolar Non-Return to Zero (NRZ): In ths lne codes, a bt o 1 s represented n an alternatng orm by some postve voltage (+5 volts or example) once and the next tme a bt

o 1 appears t wll be represented by the same voltage but wth a negatve value ( 5 volts). bt o 0 s represented by zero volts. The pulses correspondng to bnary 1 reman at the postve and negatve voltages or the whole duraton o the bt perod (they do not return to zero). The advantage o ths lne code over the Unpolar (On O) NRZ s that t has zero DC value because bts o 1 alternate n usng the postve and negatve voltages. lne code wth zero DC s desred n some applcatons that requre that the transmtted sgnal to have no DC. T 0 R 1 2 2T = R ω dvantages: 1. pectrally ecent (requres the mnmum amount o bandwdth or a specc bt rate) 2. Provde synchronzaton between the transmtter and recever or long sequences o Logc 1. 3. llows some orm o error detecton (snce two consecutve logc 1 s are represented by a postve and negatve pulses, the recever detects two consecutve pulses that have the same polarty (both are postve or both are negatve), t can easly detect that there must have been an error n the transmsson). 4. Has zero DC value regardless o the number o Logc 1s and Logc 0s n the normaton to be transmtted. Dsadvantages: 1. Does not provde synchronzaton normaton or long sequences o zeros. 2. Requres more sophstcated electroncs to be generated because t uses sgnals wth 3 levels (or example, +5, 0, and -5 V) 3. Manchester (B-Phase): In ths lne code, a bt o 0 s represented by some postve voltage or the rst hal o the bt perod and some negatve voltage or the second hal o the bt perod. bt o 1 s smply the negatve o the zero bt so t s represented by the negatve voltage or the rst hal o the bt perod and the postve voltage or the second hal o the bt. Unlke prevously dscussed lne codes, whch carry the normaton n the levels o

the pulses, snce each o the two bnary values (0 and 1) n ths lne code are transmtted usng pulses that have hal o ther duraton beng a hgh voltage (or postve voltage) and the other hal beng a low voltage (or negatve voltage), the normaton s not carred n the levels but n the transton rom hgh to low voltage or vce versa n the mddle o the pulse representng each dgtal bt. transton rom hgh to low may represent a zero whle a transton rom low to hgh would then represent a one. Ths lne code s very good or nsurng synchronzaton between the transmtter and recever. For consecutve bts that are equal, a transton may occur at the border o bts. These transtons are smply gnored and do not carry normaton. R 1 2 2T = R 1.5R dvantages: 1. Provdes ull synchronzaton normaton or long sequences o zeros and long sequences o ones. lso, ths lne code can easly be used to also provde rame synchronzaton normaton by smply transmttng bts wth both o ther two parts beng hgh voltage or both are low voltage, where the transmtter can use ths method to sgnal to the recever that a byte, or example, has ended and a new byte t startng. 2. llows some orm o error detecton (snce a bt has two parts, so the two receved parts o a bt have the same value, an error may have happened (ths assumes that rame synchronzaton as dscussed n pont 1 above s not used) 3. Has zero DC value regardless o the number o Logc 1s and Logc 0s n the normaton to be transmtted. Dsadvantages: 1. Requres more bandwdth or transmsson (approxmately 1.6 tmes the bandwdth o the prevous lne codes) snce there are more transtons n the sgnal o ths lne code compared to the prevous lne codes or the same bt rate. 2. lthough ths lne code has only two levels, t s slghtly more complcated to generate compared to Unpolar NRZ lne code.

Dgtal Modulaton Methods everal types o dgtal modulatons exst. The three basc types o dgtal modulatons are named: 1. mpltude ht Keyng (K) 2. Phase ht Keyng (PK) 3. Frequency ht Keyng (FK) (Wll not be dscussed here) Next we descrbe several o these modulatons and shown ther tme doman pulses and ther constellaton. mpltude ht Keyng (K) Bnary mpltude ht Keyng (BK) In K, the derent pulses all have the same phase but derent ampltudes. In the bnary ampltude sht keyng (BK) modulaton technque we transmt one o two pulses or each bt: 1) or logc 0, we transmt nothng ( s 0 ( t) = 0 or 0 t T ) 2) or logc 1, we transmt a modulated pulse wth magntude s () t = cos2π t or 0 t T ) ( ( ) 1 c nce all the pulses have the same phase (you can thnk o the rst sgnal as s0 ( t) = 0 cos( 2π ct) or 0 t T ), the constellaton o ths modulaton technque (the representaton o the pulses o the modulaton algorthm) becomes M-ary mpltude ht Keyng (M-ary K) In ths dgtal modulaton technque, we transmt one o M pulses, where M s a power o two number such that

M = 2 n. The quantty n here s equal to the number o bts that are carred by each transmtted pulse. To determne whch pulse to transmt n each case, we wll have to dvde the sequence o normaton bts nto groups o n consecutve bts. The combnaton o bts wll be one o M = 2 n possble combnatons that wll determne the pulse to be transmtted or these n bts. o, n ths modulaton we wll transmt one o the ollowng pulses: 1) or bt sequence 00..00, transmt s () 0 or 0 00 00 t = t T 2) or bt sequence 00..01, transmt 00 01 ( π 3) or bt sequence 00..10, transmt 00 10 ( π 4) or bt sequence 00..11, transmt ( π ) s () t = cos2 t or 0 t T s () t = 2 cos2 t or 0 t T s () t = 3 cos2 t or 0 t T 00 11 : : M) or bt sequence 11..11, transmt ( ) ( π ) 11 11 c s ( t) = M 1 cos 2 t or 0 t T nce all pulses o ths modulaton technque have the same phase, the representaton o all pulses n the constellaton all on a straght lne. The constellaton o ths modulaton algorthm s shown below. c Clearly n ths case, the power requred or transmttng derent pulses s derent and the average power o transmsson can be obtaned easly by averagng all powers assumng that derent bt sequences have equal probabltes. Phase ht Keyng (PK) Bnary Phase ht Keyng (BPK) In ths dgtal modulaton technque, we transmt one o two pulses or each bt that are smlar n magntude but have a 180 phase sht between them (one o them s the negatve o the other): 1) or logc 1, we transmt the pulse 1 ( π 2) or logc 0, we transmt the pulse ( π ) The constellaton o the modulaton algorthm n ths case becomes. s () t = cos 2 t or 0 t T s () t = cos 2 t or 0 t T 0 c

It s worth mentonng that although ths algorthm s very smlar to the BK, ths s better because wth the same probablty o error as a BK, ths modulaton technque would requre less power (very ths). Quadrature Phase ht Keyng (QPK) In ths dgtal modulaton technque, we transmt one o our pulses that correspond to a group o 2 bts. ll pulses have equal magntudes but are separated n phase by 90 (each two are negatve o each other). Because o the phase sht o 90 between the derent pulses, two o the pulses wll be represented by dots on the x-axs and two wll be represented by dots on the y-axs. Ths means that the 4 pulses are 1) or logc 00, we transmt the pulse 00 ( π 2) or logc 01, we transmt the pulse 01 ( π 3) or logc 10, we transmt the pulse 10 ( π 4) or logc 11, we transmt the pulse ( π ) s () t = cos 2 t or 0 t T s () t = sn 2 t or 0 t T s () t = cos2 t or 0 t T s () t = sn 2 t or 0 t T 11 c The constellaton now s a 2-D constellaton that looks lke the ollowng

8-ary Phase ht Keyng (8-ary PK) The QPK modulaton dscussed above can be extended urther where more symbols are added so nstead o havng 4 symbols n the QPK, we now have 8 symbols. We transmt one o 8 pulses such that each pulse now carres 3 bts. ll pulses wll have equal magntudes but are separated n phase by 45. Thereore, the transmtted pulses are 0π 1) s000( t) = cos 2π ct or 0 t T 2 π 2) s001( t) = cos 2π ct or 0 t T 2π 3) s010( t) = cos 2π ct or 0 t T 3π 4) s011( t) = cos 2π ct or 0 t T 4π 5) s100( t) = cos 2π ct or 0 t T 5π 6) s101( t) = cos 2π ct or 0 t T 6π 7) s110( t) = cos 2π ct or 0 t T 7π 8) s111( t) = cos 2π ct or 0 t T o the constellaton o ths modulaton can easly be shown to be the ollowng (remember that ( α β) ( α) ( β) ( α) ( β) cos ± = cos cos sn sn ):

PK Modulatons wth hgher power o two values o M are possble but as the value o M ncreases, the modulaton algorthm becomes more necent as derent ponts n ts constellaton become very close to each other resultng n hgh probablty o error compared to the requred power o transmsson. We can easly mody the pulses that are requred to transmt the normaton such that the average power o transmsson s reduced and at the same tme, the probablty o error s ether lowered or at least mantaned at ts level. In ths case, other modulaton technques n whch the constellaton pnts are dstrbuted more evenly are used. These modulaton technques are called Quadrature mpltude Modulaton (QM), and rom ther name, they are a mx o both K and PK. Quadrature mpltude Modulaton (QM) Oten, we need to transmt more bts per symbol than what smple dgtal modulaton technques descrbed n above and n the prevous lecture can practcally do. We have seen that M-ary PK can be used or transmttng 3 bts per symbol or hgher bts per symbol but once the number o bts per symbol ncreases, the ecency (probablty o error) o the modulaton ncreases sgncantly because the ponts o the constellaton are conned to a crcle o a specc radus. Redstrbutng the constellaton ponts n a derent type o modulaton gves better perormance n terms o probablty o error at the expense o a more complcated system and a no longer constant ampltude or the derent pulses as t s the case wth all PK modulatons. We can generate M sgnals that have derent ampltudes and/or phases such that each carres n bts ( n= log2 M ) by havng the derent pulses n the orm

where {, } ( π ) ( π ) s ( t) = a cos 2 t + b sn 2 t or 0 t T c c s and = 1, 2,, M a b s a par o coecents that are gven by a matrx o pars dependng on the desred M-ary QM modulaton scheme (or sometmes called M-QM or short). For example, 16-QM wll have a matrx or {, } { a, b} a b gven by ( 3,3) ( 1,3) (1,3) (3,3) ( 3,1) ( 1,1) (1,1) (3,1) = ( 3, 1) ( 1, 1) (1, 1) (3, 1) ( 3, 3) ( 1, 3) (1, 3) (3, 3) o n 16-QM, each 4 bts are combned and based on the pattern o these 4 bts (gvng 16 possbltes), the correspondng par {, } pulse or these 4 bts s generated and transmtted. The constellaton o the above 16-QM s shown below: a b s selected rom the above matrx and the correspondng More complcated constellatons are obtaned or hgher levels o QM. Once the pulses have been transmtted, the receved pulses can be plotted on the same constellaton. The pulse assumed to have been transmtted s the one wth the pont n the constellaton that s closest to that o the receved pulse. QM s wdely used n dgtal communcaton systems ncludng the V.92 56k bps modem.