$ESIGN AND !

NALYSIS OF %STIMATORS FOR -ULTICARRIER -ODULATION AND 5LTRASONIC )MAGING

-AGNUS 3ANDELL

$ESIGN AND !NALYSIS OF %STIMATORS FOR -ULTICARRIER -ODULATION AND 5LTRASONIC )MAGING

-AGNUS 3ANDELL
,ULE 5NIVERSITY OF 4ECHNOLOGY $IVISION OF 3IGNAL 0ROCESSING ,ULE 3WEDEN

3EPTEMBER 

3UPERVISOR 0ROFESSOR 0ER /LA "RJESSON ,ULE 5NIVERSITY OF 4ECHNOLOGY

!BSTRACT
4HIS THESIS DEALS WITH TWO AREAS MULTICARRIER MODULATION AND ULTRASONIC IMAGING !LTHOUGH DIdERENT IN APPLICATION THESE ELDS HAVE A LOT IN COMMON &OR INSTANCE SYN CHRONIZATION AND CHANNEL ESTIMATION IN COMMUNICATION SYSTEMS CORRESPOND TO TIME DELAY ESTIMATION AND SYSTEM IDENTICATION IN ULTRASONIC IMAGING 4HIS THESIS CONCENTRATES ON THESE ESTIMATION PROBLEMS 4HREE TASKS IN ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING /&$-  A TYPE OF MUL TICARRIER SYSTEM ARE CONSIDERED SYNCHRONIZATION CHANNEL ESTIMATION AND CODING 4HE CYCLIC PREX A PREAMBLE WITHIN /&$- SYMBOLS USED TO AVOID INTERSYMBOL AND INTER CARRIER INTERFERENCE IS COMPOSED OF REDUNDANT DATA AND IS SHOWN TO CONTAIN SUbCIENT INFORMATION TO BE USED FOR SYNCHRONIZATION PURPOSES &ROM THIS REDUNDANCY IT IS POSSIBLE TO ESTIMATE BOTH TIME AND FREQUENCY OdSETS THUS REDUCING THE NEED FOR PILOTS KNOWN SYMBOLS  ! NOVEL SIMULTANEOUS MAXIMUM LIKELIHOOD ESTIMATOR OF THE TIME AND FREQUENCY OdSET IS PRESENTED #HANNEL ESTIMATION IN /&$- SYSTEMS IS USUALLY DONE WITH PILOT SYMBOL ASSISTED MODULATION IE PILOTS ARE MULTIPLEXED INTO THE TRANSMITTED SYMBOL STREAM %STIMATION OF THE CHANNEL WITH NITE IMPULSE RESPONSE &)2 LTERS HAS BEEN PROPOSED BUT WE SHOW THAT LOW RANK ESTIMATORS BASED ON OPTIMAL RANK REDUCTION CAN GIVE BETTER PERFORMANCE WITH THE SAME COMPLEXITY 4HESE NOVEL ESTIMATORS ARE PRESENTED AND EVALUATED BOTH IN TERMS OF MEAN SQUARED ERROR AND CODED BIT ERROR RATE 4HE MEAN SQUARED ERROR IS OBTAINED FROM ANALYTICAL EXPRESSIONS AND THE CODED BIT ERROR RATE FROM BOTH SIMULATIONS AND ANALYTICAL EXPRESSIONS 4HE ANALYTICAL METHOD WE EMPLOY FOR THESE EVALUATIONS IS ITSELF INVESTIGATED AS A PART OF THIS THESIS 4HIS METHOD OF ESTIMATING CODED BIT ERROR RATES PRESENTED IN THE LITERATURE A FEW YEARS AGO IS IN THIS THESIS EXTENDED TO /&$- SYSTEMS 3HOWING A HIGH POTENTIAL FOR ANALYZING CODED SYSTEMS IT CAN BE VERY USEFUL IN THE DESIGN OF AN /&$- SYSTEM SINCE IT DOES NOT REQUIRE LENGTHY SIMULATIONS 4HIS MAKES IT POSSIBLE TO VARY SEVERAL PARAMETERS AND OBTAIN BIT ERROR RATES MORE PROMPTLY 4HE PROBLEMS CONSIDERED IN ULTRASONIC IMAGING ARE SYSTEM IDENTICATION AND TIME DELAY ESTIMATION 4HE FORMER SEEKS TO OBTAIN THE SPATIAL IMPULSE RESPONSE OF AN ULTRASONIC TRANSDUCER IE THE ECHO FROM A IDEALIZED POINT REECTOR 4HIS CAN BE VERY USEFUL SINCE THE ECHO FROM AN ARBITRARY SURFACE CAN BE MODELLED AS A SUM OF THE ECHOES FROM THE INDIVIDUAL POINTS MAKING UP THE SURFACE 4HE SPATIAL IMPULSE RESPONSE CAN BE MEASURED DIRECTLY BUT THIS POSES SOME PRACTICAL PROBLEMS 4HE REECTOR MUST BE MADE SMALL ENOUGH TO RESEMBLE A POINT INCREASING THE REECTED ECHOS SUSCEPTIBILITY TO MEASUREMENT NOISE 7E PROPOSE A NOVEL METHOD OF ESTIMATING THE SPATIAL IMPULSE RESPONSE INDIRECTLY BY MEASURING THE ECHOES FROM SLIDING HALFPLANES AND INVERTING THEM 7ORKING WITH THIS METHOD WE FOUND A NEED FOR ESTIMATING SMALL TIME DELAY DIdERENCES IN THE RECEIVED ECHOES (ENCE A FAST AND SIMPLE METHOD WAS DEVELOPED WHICH IS BASED ON THE CORRELATION WITH (ILBERT TRANSFORMED ECHOES

#ONTENTS
0REFACE !CKNOWLEDGEMENTS 4HESIS )NTRODUCTION 3IGNAL -ODELS AND %STIMATION 0ROBLEMS IX XI 

-ULTICARRIER -ODULATION
 !N        !  -,      ! )NTRODUCTION TO /RTHOGONAL )NTRODUCTION           3YSTEM MODELS         3YSTEM ENVIRONMENTS      3YNCHRONIZATION         #HANNEL ESTIMATION       #HANNEL CODING         $ISCUSSION            4IME FREQUENCY LATTICE     &REQUENCY $IVISION -ULTIPLEXING                                                                                                                                                                                                               


                         

%STIMATION OF 4IMING AND &REQUENCY /dSET IN /&$- 3YSTEMS )NTRODUCTION                                  4HE /&$- SYSTEM MODEL                          -AXIMUM LIKELIHOOD ESTIMATION                       3IMULATIONS                                  $ISCUSSION                                   4HE LOG LIKELIHOOD FUNCTION                                                    

 /&$- #HANNEL %STIMATION BY 3INGULAR 6ALUE $ECOMPOSITION  )NTRODUCTION                                3YSTEM DESCRIPTION                            ,INEAR CHANNEL ESTIMATION ACROSS TONES                 %STIMATOR PERFORMANCE AND DESIGN                    'ENERIC LOW RANK ESTIMATOR                        #ONCLUSIONS                               ! /PTIMAL RANK REDUCTION                         " #HANNEL CORRELATION MATRICES                      # %STIMATOR MEAN SQUARED ERROR                    

 ! #OMPARATIVE 3TUDY OF 0ILOT BASED #HANNEL %STIMATORS FOR 7IRELESS /&$  )NTRODUCTION                                   

    ! "

3YSTEM DESCRIPTION   %STIMATORS        0ERFORMANCE EVALUATION #ONCLUSIONS       #OMPLEXITY OF LOW RANK #ORRELATION MATRICES 

                            ESTIMATORS        

     

     

     

     

     

     

     

     

     

     

     

     

     

     

     

     

     

     

     

     

     

     

 0ERFORMANCE !NALYSIS OF #ODED /&$- ON &ADING #HANNELS IDEAL )NTERLEAVING AND #HANNEL +NOWLEDGE  )NTRODUCTION                              3YSTEM DESCRIPTION                          0ERFORMANCE ANALYSIS                         3IMULATION                               #ONCLUSIONS                             ! #HANNEL CORRELATION                         " #HANNEL ESTIMATOR                         # %RROR EVENT ENUMERATION                      $ ,APLACE TRANSFORM h5 R                     

WITH .ON                                                                

5LTRASONIC )MAGING



 %STIMATION OF THE 3PATIAL )MPULSE 2ESPONSE OF AN 5LTRASONIC 4RANSDUCER 5SING A 4OMOGRAPHIC !PPROACH   )NTRODUCTION                                     4HEORY                                        %RROR ESTIMATES                                   %XPERIMENTS                                     2ESULTS                                        3UMMARY                                     ! 4HE INTEGRATION KERNEL                              " 36$ WITH WEIGHTED NORMS                             %STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES IN .ARROWBAND 5L TRASONIC %CHOES 5SING THE (ILBERT 4RANSFORM #ORRELATION   )NTRODUCTION                                     $ERIVATION OF THE ESTIMATOR                             0ERFORMANCE OF THE ESTIMATOR                           !PPLICATION EXAMPLE AVERAGING ALGORITHM                    %XPERIMENTS                                     2ESULTS                                        $ISCUSSION                                     ! 4HE EdECT OF BANDWIDTH                             " ,INEARIZING THE ESTIMATOR                            

0REFACE
4HIS THESIS DEALS WITH TWO AREAS OF APPLICATION MULTICARRIER MODULATION AND ULTRASONIC IMAGING ) STARTED AS A 0H$ STUDENT BY WORKING ON ULTRASONIC PULSE ECHO SYSTEMS 7E INVESTIGATED THE BEHAVIOR OF A TRANSDUCER FOR AIRBORNE ULTRASOUND TO SEE IF THE RESOLUTION COULD BE IMPROVED 4HIS WORK WAS PRESENTED IN MY LICENTIATE THESIS WHICH DEALT WITH THE SPATIAL DECOMPOSITION OF ULTRASONIC ECHOES "ASED ON THAT MATERIAL TWO ARTICLES WERE WRITTEN AND PUBLISHED IN INTERNATIONAL MAGAZINES 4HEY ARE PRESENTED HERE AS 0ARTS  AND  !FTER MY LICENTIATE THESIS ) CHANGED MY AREA OF RESEARCH AND STARTED TO LOOK AT MUL TICARRIER MODULATION IN PARTICULAR ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING /&$-  -Y WORK FOCUSED ON SYNCHRONIZATION AND CHANNEL ESTIMATION WHICH ARE PRESENTED HERE AS 0ARTS   4HE WORK ON 0ART  WAS MAINLY DONE DURING A VE MONTH VISIT TO 0URDUE 5NIVERSITY WHERE ) WORKED WITH 0ROF 3ARAH +ATE 7ILSON 4HE THESIS IS COMPRISED APART FROM THE INTRODUCTION OF THE FOLLOWING PARTS  / %DFORS - 3ANDELL * * VAN DE "EEK $ ,ANDSTRM & 3JBERG !N INTRODUC TION TO ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING 2ESEARCH REPORT   * * VAN DE "EEK - 3ANDELL 0 / "RJESSON -, ESTIMATION OF TIMING AND FRE QUENCY OdSET IN /&$- SYSTEMS  !CCEPTED FOR PUBLICATION IN )%%% 4RANS ACTIONS ON 3IGNAL 0ROCESSING  / %DFORS - 3ANDELL * * VAN DE "EEK 3 + 7ILSON 0 / "RJESSON /&$CHANNEL ESTIMATION BY SINGULAR VALUE DECOMPOSITION 2ESEARCH REPORT   - 3ANDELL / %DFORS ! COMPARATIVE STUDY OF PILOT BASED CHANNEL ESTIMATORS FOR WIRELESS /&$- 2ESEARCH 2EPORT   - 3ANDELL 3 + 7ILSON 0 / "RJESSON 0ERFORMANCE ANALYSIS OF CODED /&$ON FADING CHANNELS WITH NON IDEAL INTERLEAVING AND CHANNEL KNOWLEDGE 2ESEARCH 2EPORT   - 3ANDELL ! 'RENNBERG  %STIMATION OF THE SPATIAL IMPULSE RESPONSE OF AN ULTRASONIC TRANSDUCER USING A TOMOGRAPHIC APPROACH *OURNAL OF THE !COUSTICAL 3OCIETY OF !MERICA VOL  NO  /CT  PP   ! 'RENNBERG - 3ANDELL %STIMATION OF SUBSAMPLE TIME DELAY DIdERENCES IN NAR ROWBAND ULTRASONIC ECHOES USING THE (ILBERT TRANSFORM CORRELATION )%%% 4RANSAC TIONS ON 5LTRASONICS &ERROELECTRICS AND &REQUENCY #ONTROL VOL  NO  3EP  PP   IX

!CKNOWLEDGEMENTS
4HIS THESIS WAS WRITTEN AND CONCEIVED AT THE DIVISION OF 3IGNAL 0ROCESSING ,ULE 5NIVERSITY OF 4ECHNOLOGY )T HAS BEEN MY HOME FOR THE LAST TEN YEARS RST AS AN UNDER GRADUATE STUDENT AND THEN AS A 0H$ CANDIDATE 4HOUGH A DRASTIC MOVE FOR A SOUTHERN BOY ) HAVE NEVER REGRETTED MOVING UP HERE ) HAVE NEVER HAD ANY PROBLEMS WITH THE CLIMATE ALTHOUGH IT IS NINE MONTHS OF WINTER AND THREE MONTHS OF BAD SKIING WEATHER $URING MY TIME AS A 0H$ STUDENT ) HAVE HAD 0ROF 0ER /LA "RJESSON AS MY SUPERVISOR 4WO OTHER PEOPLE HAVE ALSO TAKEN ON A SENIOR ROLE $R !NDERS 'RENNBERG AND 0ROF 3ARAH +ATE 7ILSON !LL HAVE HELPED ME A LOT EACH WITH THEIR SPECIALTY IN MATHEMATICS !NDERS  SIGNAL PROCESSING 0ER /LA AND COMMUNICATIONS +ATIE  ) WOULD LIKE TO THANK THEM FOR THEIR SUPPORT AND GUIDANCE OVER THE YEARS !LSO A SPECIAL THANKS TO !NDERS FOR HIRING ME IN THE RST PLACE /UR DIVISION HAS BEEN IN THE MIDDLE OF THE MATHEMATICS AND THE APPLICATIONS THE LATTER THROUGH OUR COOPERATION WITH 4ELIA 2ESEARCH !" )T HAS BEEN A SUCCESSFUL SYMBIOSIS AND ) WOULD LIKE TO TAKE THE OPPORTUNITY TO THANK THEM 3EVERAL 0H$ STUDENTS FROM OUR DIVISION INCUDING ME HAVE HAD AN ObCE AT 4ELIA 2ESEARCH !" 4HAT TIME WAS VERY USEFUL AND ) LEARNED A LOT MY POOL GAME IMPROVED AS WELL  -OST OF MY WORK HAS BEEN DONE TOGETHER WITH MY FELLOW 0H$ STUDENTS /VE %DFORS AND *AN *AAP VAN DE "EEK $URING A FEW SWEATY SUMMER MONTHS /VE AND ) NISHED OUR THESES TOGETHER WHICH WAS BOTH MISERABLE AND REWARDING 4O PARAPHRASE 7INSTON #HURCHILL NEVER HAVE SO FEW WRITTEN SO MUCH IN SO LITTLE TIME *AN *AAP HAS HAD TO READ MOST OF THIS WHICH WE OWE HIM A BIG THANKS FOR "UT ) GUESS HE WILL GET HIS REVENGE WHEN HE WRITES HIS THESIS )T HAS BEEN A PLEASURE BOTH WORKING AND SOCIALIZING WITH THESE FELLOWS ) WOULD ALSO LIKE TO THANK THE REST OF THE DIVISION FOR CREATING AN OPEN MINDED AND SUPPORTIVE ATMOSPHERE AT WORK 4HIS CERTAINLY INCLUDES THE LOST SHEEP .ILS 3UNDSTRM AND 3VEN .ORDEBO ) WAS GIVEN THE OPPORTUNITY TO VISIT 0URDUE 5NIVERSITY AND 0ROF 3ARAH +ATE 7ILSON FOR VE MONTHS )T WAS A REWARDING EXPERIENCE AND ) WOULD LIKE TO THANK +ATIE FOR THE TIME THERE !LSO MY THANKS GOES TO 3HIAU HE 4SAI FOR HIS HELP AND ASSISTANCE !T 0URDUE ) ALSO HAD THE PLEASURE TO SEE 0AUL 0ETERSEN WHO HAS KEPT A LINGUISTIC EYE ON SEVERAL OF MY PAPERS AND MADE INVALUABLE SUGGESTIONS AND IMPROVEMENTS OF THE %NGLISH LANGUAGE IN THIS THESIS /N THE LIQUID SIDE OF LIFE ) WOULD LIKE TO SHOW MY APPRECIATION TO THE REGULARS AT THE 3TAd 0UB AND TO !KSEL STERLF FOR KEEPING IT ALIVE )T HAS BEEN VERY PLEASANT TO END THE WORKING WEEK WITH A FEW BEERS AND START THE WORKING WEEKEND WITH A HANGOVER 4HERE ARE SO MANY PEOPLE WHO HAVE HELPED ME THAT ) WISH ) COULD THANK BUT )M RUNNING OUT OF SPACE 3O ) NISH BY THANKING MY PARENTS FOR ENCOURAGING ME TO STUDY AT THE UNIVERSITY 7ITHOUT THEM ) WOULD PROBABLY NOT BE HERE

XI

4HESIS )NTRODUCTION 3IGNAL -ODELS AND %STIMATION 0ROBLEMS

4HESIS )NTRODUCTION 3IGNAL -ODELS AND %STIMATION 0ROBLEMS

)NTRODUCTION
!N IMPORTANT ELEMENT IN THE DESIGN AND ANALYSIS OF SIGNAL PROCESSING ALGORITHMS IS THE SIGNAL MODEL )T SHOULD NOT ONLY REECT THE ACTUAL SYSTEM BUT ALSO MAKE A TRACTABLE ANALYSIS POSSIBLE 4HESE TWO GOALS ARE OFTEN CONTRADICTORY REQUIRING A TRADE Od )N THIS THESIS WE USE SIGNAL MODELS THAT ARE INTENTED TO BE ADEQUATE FOR BOTH OF THESE PURPOSES IE THEY ARE ACCURATE WITH RESPECT TO THE SOUGHT PARAMETERS WHILE THEY MAY SIMPLIFY THE REST OF THE SYSTEM &ROM A SIGNAL PROCESSING POINT OF VIEW THESE MODELS ARE PREFERABLE SINCE FUNDAMENTAL ESTIMATION TECHNIQUES ARE APPLICABLE IN A RATHER STRAIGHTFORWARD MANNER )N THIS THESIS WE CONCENTRATE ON TWO ESTIMATION PROBLEMS SYNCHRONIZATION AND CHAN NEL ESTIMATION 7HILE THESE ARE THE TERMS FOR A COMMUNICATION SYSTEM THEY WOULD BE CALLED TIME DELAY ESTIMATION AND SYSTEM IDENTICATION IN ULTRASONIC IMAGING 4HIS THE SIS COVERS BOTH THOSE APPLICATIONS 3YNCHRONIZATION IN COMMUNICATION SYSTEMS SUCH AS A MULTICARRIER MODULATION SYSTEM SEEKS TO ALIGN TRANSMITTER AND RECEIVER IN TIME AND FREQUENCY 4YPICALLY IN A COHERENT RECEIVER THE RECEIVER NEEDS TO REGENERATE A COHER ENT REPLICA OF THE CARRIER GENERATE A PROPER SYNCHRONIZATION SYMBOL CLOCK AND COUNTER ACT CHANNEL DISPERSION 7E CALL THESE THREE OPERATIONS EACH OF WHICH COMPENSATES FOR SOME CHANNEL UNCERTAINTY CARRIER SYNCHRONIZATION SYMBOL SYNCHRONIZATION AND CHANNEL SYNCHRONIZATION 4HE LAST IS USUALLY CALLED CHANNEL EQUALIZATION 4HUS WE MAY VIEW THE ESTIMATION OF TIMING AND FREQUENCY OdSETS AND CHANNEL ATTENUATION AS TWO KINDS OF SYNCHRONIZATION 3IMILAR TASKS MAY BE RECOGNIZED IN ULTRASONIC IMAGING 4HE TYPE OF ULTRASONIC IMAGING CONSIDERED IN THIS THESIS IS PULSE ECHO IMAGING WHICH USES THE SAME PRINCIPLE AS SONAR ! PULSE IS TRANSMITTED FROM THE TRANSDUCER AND THE CHOES FROM THE SURFACE OF AN OBJECT ARE RECORDED 4HE TIME OF IGHT OF THE PULSE IS PROPORTIONAL TO THE DISTANCE BETWEEN THE TRANSDUCER AND THE OBJECT 5SUALLY THE RELATIVE DISTANCE BETWEEN TWO OBJECTS IS MORE INTERESTING THAN THE ABSOLUTE DISTANCE BETWEEN TRANSDUCER AND OBJECT SO THE RELEVANT PA RAMETER TO ESTIMATE IS THE TIME DELAY DIdERENCE BETWEEN THE TWO ECHOES !  DIMENSIONAL IMAGE OF AN OBJECT MAY NOW BE OBTAINED BY MOVING THE TRANSDUCER OVER THE OBJECT 3INCE THE TRANSMITTED ULTRASONIC PULSE CANNOT BE MADE INNITELY NARROW A SMALL AREA ON THE OBJECT WILL CAUSE THE ECHO AND NOT JUST ONE POINT 4HIS CAN BE VIEWED AS THE EQUIVALENT OF DISPERSION IN A COMMUNICATION SYSTEM AND HENCE CHANNEL ESTIMATION CORRESPONDS TO IDENTIFYING THIS BLURRING

-ULTICARRIER MODULATION
7E CONSIDER ONE TYPE OF MULTICARRIER MODULATION ORTHOGONAL FREQUENCY DIVISION MULTI PLEXING /&$-  )N %UROPE IT IS USED FOR BROADCASTING OF DIGITAL RADIO ;= AND HAS BEEN PROPOSED FOR DIGITAL VIDEO ;    = )N ADDITION WIRELESS MULTIUSER SYSTEMS

4HESIS )NTRODUCTION 3IGNAL -ODELS AND %STIMATION 0ROBLEMS

BASED ON /&$- ARE ALSO BEING CONSIDERED ;      = AS WELL AS WIRED COMMUNICATION APPLICATIONS OVER EXISTING COPPER NETWORKS ; = )N THE LATTER APPLICATION /&$- OFTEN GOES UNDER THE NAME OF DISCRETE MULTITONE $-4  4HE IDEAS BEHIND /&$- ARE NOT NEW ;   = BUT THE BREAKTHROUGH CAME ONLY A FEW YEARS AGO ; = )N /&$- DATA IS TRANSMITTED SIMULTANEOUSLY ON PARALLEL SUBCHANNELS EACH OCCUPYING A SMALL FRACTION OF THE AVAILABLE BANDWIDTH 4HUS IT IS A KIND OF FREQUENCY DIVISION WHERE THE FREQUENCY RESPONSE OF THE SUBCHANNELS ARE OVERLAPPING BUT ORTHOGONAL HENCE THE NAME /&$- ;= 4HIS DECOMPOSITION OF THE CHANNEL INTO NARROWBAND SUBCHANNELS HAS BEEN SHOWN TO BE GOOD FROM A THEORETICAL STANDPOINT ; = 4HE TRANSMITTED /&$SYMBOL IS RS 
E  8 t(

Wt Ds { Q 
t

  S  3
3fy



WHERE Wt ARE DATA SYMBOLS EG FROM A 1!- CONSTELLATION  3 IS THE SYMBOL TIME AND 3fy IS A GUARD INTERVAL BETWEEN CONSECUTIVE /&$- SYMBOLS .OTE THAT THE SUBCARRIERS  ARE SEPARATED BY Q AND THE GUARD SPACE PROTECTS AGAINST INTERSYMBOL INTERFERENCE 4HE MODULATION ON THE SUBCARRIERS CAN BE EbCIENTLY IMPLEMENTED IN A DIGITAL SYSTEM BY MEANS OF AN INVERSE DISCRETE &OURIER TRANSFORM )$&4 ;= $EMODULATION IS IMPLEMENTED SIMILARLY SEE &IG 

&IGURE  4HE /&$- SYSTEM TRANSMITTING SUBSEQUENT BLOCKS OF - COMPLEX DATA SYMBOLS $IGITAL IMPLEMENTATION A AND EQUIVALENT MODEL B  !LL SUBCHANNELS HAVE A NARROWBAND FREQUENCY RESPONSE AND EXPERIENCE APPROXIMATELY AT FADING 4HIS CAN BE OBTAINED EXACTLY IF THE GUARD SPACE CONSISTS OF A CYCLIC PREX ;= IE A COPY OF THE LAST PART OF THE SYMBOL WHICH IS AT LEAST AS LONG AS THE DURATION OF THE CHANNEL )N THIS CASE THE INPUT OUTPUT RELATION BETWEEN THE TRANSMITTED SYMBOL Wt AND THE RECEIVED SYMBOL Xt ON SUBCHANNEL J IS ;= Xt  Gt Wt
Mt  

(ERE Gt IS THE ATTENUATION AND Mt IS THE ADDITIVE CHANNEL NOISE AT SUBCARRIER J 4HIS MODEL OF AN /&$- SYSTEM IS COMMON AND USED FOR CHANNEL ESTIMATION AND BIT ERROR EVALUATION ;  = )T WILL BE USED THROUGHOUT THIS THESIS (OWEVER THE EXPRESSION

  ONLY HOLDS IF THE CHANNEL IS XED DURING ONE /&$- SYMBOL OTHERWISE INTER CARRIER INTERFERENCE )#) WILL OCCUR WHICH DESTROYS THE ORTHOGONALITY ; = 3INCE EACH SUBCHANNEL IS AT FADING IN /&$- ESTIMATES OF THE CHANNEL ATTENUATIONS Gt CAN BE OBTAINED BY TRANSMITTING KNOWN SYMBOLS ON CERTAIN SUBCHANNELS AND INTERPO LATING BETWEEN THEM ;= 4HIS IS A GENERALIZATION OF PILOT SYMBOL ASSISTED MODULATION FOR SINGLE CARRIER SYSTEMS ; = &INITE IMPULSE RESPONSE &)2 LTERS HAVE BEEN PRO POSED ; = BUT WE INVESTIGATE TRANSFORM BASED ESTIMATORS ; = 4HESE CAN BE VERY EbCIENT SINCE THE CHANNEL IMPULSE RESPONSE IS MUCH SHORTER THAN THE /&$- SYMBOL AND HENCE THE FREQUENCY RESPONSE IS OVERSAMPLED 4HIS MEANS THAT THE CHANNEL LIES IN A SUBSPACE OF THE OBSERVATIONS AND BY PERFORMING ESTIMATION IN THIS SUBSPACE WHICH HAS A CONSIDERABLY LOWER DIMENSION THAN THE OBSERVATIONS ESTIMATORS WITH LOW COMPLEXITY AND HIGH PERFORMANCE CAN BE CONSTRUCTED ;= 4HE CYCLIC PREX INTRODUCES REDUNDANCY INTO THE TRANSMITTED SIGNAL SINCE IT IS THE COPY OF THE LAST PART OF THE /&$- SYMBOL 4HIS REDUNDANCY CAN BE USED IN EG SYNCHRONIZA TION )NFORMATION ON BOTH TIME AND FREQUENCY OdSETS ARE CONTAINED WITHIN THE RECEIVED SIGNAL THUS REDUCING THE NEED FOR PILOTS (OW TO USE THIS REDUNDANCY CAN INTUITIVELY BE SEEN IN &IG  WHERE ONLY A TIME OdSET IS CONSIDERED .OTE THAT THE OBSERVATION INTERVAL

&IGURE  /&$- SYMBOL WITH AN UNKNOW TIME OdSET t 4HE CYCLIC PREX ) IS A COPY OF THE LAST PART OF THE /&$- SYMBOL )   IS CHOSEN SUCH THAT THERE CAN BE ONLY ONE COMPLETE /&$- SYMBOL IN IT )F WE ONLY CON SIDER ADDITIVE CHANNEL NOISE IE NO CHANNEL DISPERSION THE OBSERVED SIGNAL IS COMPOSED ALMOST ENTIRELY OF UNCORRELATED SAMPLES 4HE CORRELATIONS THAT DO EXIST ARE CAUSED BY THE CYCLIC PREX 4HE POSITION OF THE CYCLIC PREX WITHIN THE OBSERVATION INTERVAL CAN BE ESTIMATED BY LOOKING AT THE AUTOCORRELATION OF THE RECEIVED SIGNAL 7E DESIGNED A MAXIMUM LIKELIHOOD ESTIMATOR ;= AND SHOWED THAT IT COULD BE USED TO SIMULTANEOUSLY ESTIMATE TIME AND FREQUENCY OdSETS ;=

5LTRASONIC IMAGING
5LTRASOUND IS OFTEN USED FOR  DIMENSIONAL IMAGING ESPECIALLY WHEN THE MEASUREMENT EQUIPMENT IS NOT ALLOWED TO BE IN CONTACT WITH THE INVESTIGATED OBJECT 5LTRASONIC IMAGING

4HESIS )NTRODUCTION 3IGNAL -ODELS AND %STIMATION 0ROBLEMS

IS COMMON IN EG MEDICAL APPLICATIONS ;  = )N THIS THESIS WE CONSIDER THE PROBLEM OF OBTAINING A DETAILED SURFACE PROLE  DIMENSIONAL IMAGE OF A STATIONARY OBJECT BY USING AIRBORNE ULTRASOUND ;= 7E USE PULSE ECHO IMAGING WHERE THE ECHO FROM THE TRANSMITTED PULSE IS RECORDED 4HE TIME OF IGHT OF THE PULSE IS PROPORTIONAL TO THE DISTANCE BETWEEN THE TRANSDUCER AND THE OBJECT AND BY ESTIMATING THIS TIME ;  = THE DISTANCE CAN BE OBTAINED "Y MOVING THE TRANSDUCER OVER THE SURFACE OF THE OBJECT A NUMBER OF DISTANCES CAN BE COLLECTED AND A  DIMENSIONAL IMAGE FORMED 3INCE THE ABSOLUTE DISTANCE BETWEEN THE TRANSDUCER AND THE OBJECT IS NOT IMPORTANT IN THIS KIND OF APPLICATION ONLY THE RELATIVE TRAVEL TIMES BETWEEN THE ECHOES IS ESTIMATED ;= 4HE ULTRASONIC PULSE IS CREATED IN THE TRANSDUCER BY USING THE PIEZOELECTRIC EdECT ;= RST DESCRIBED BY 0IERRE AND *ACQUES #URIE IN  ;= "Y APPLYING AN ELECTRIC ELD CERTAIN MATERIALS CHANGE THEIR PHYSICAL DIMENSIONS 4HIS CAN BE USED TO CREATE A VIBRATION WHICH THROUGH A MATCHING LAYER IS LED TO THE SURFACE OF THE TRANSDUCER 4HE CREATED ULTRASONIC ELD IS THEN PROPAGATED THROUGH THE MEDIUM IN OUR CASE AIR 3INCE THE ULTRASONIC ATTENUATION IN AIR IS RATHER HIGH FOCUSED TRANSDUCERS ARE USED ;= 4HESE HAVE A CONCAVE SURFACE WHICH FOCUS THE ULTRASONIC ENERGY TO A SMALL AREA ;= 4HE WIDTH OF THE ULTRASONIC BEAM FOR OUR TRANSDUCER IS  MM AT THE FOCUS OF THE SYSTEM ;= 4HE PULSE ECHO SYSTEM CAN BE MODELLED ;= AS IN &IG 

&IGURE  -ODEL OF AN ULTRASONIC PULSE ECHO SYSTEM 4HE TRANSFER FUNCTION 3 S IS THE RELATION BETWEEN THE ELECTRICAL INPUT AND THE GENER ATED ACOUSTICAL ELD AT THE SURFACE OF THE TRANSDUCER 5SUALLY IT IS ASSUMED THAT THIS IS THE SAME FOR TRANSMISSION AND RECEPTION 4HE ACOUSTICAL ELD AT THE TRANSDUCERS SURFACE IS PROPAGATED THROUGH THE MEDIUM TO THE REECTING POINT WITH A TRANSFER FUNCTION G Q S  4HIS FUNCTION DEPENDS ON THE SPATIAL POSITION Q OF THE REECTING POINT AND IS ALSO ASSUMED TO BE THE SAME FOR TRANSMISSION AND RECEPTION )T HAS BEEN THEORETICALLY CALCULATED FOR A NUMBER OF TRANSDUCER SHAPES SUCH AS CIRCULAR AND RECTANGULAR PLANES ;= AND CONCAVE SURFACES ; = 4HE TRANSFER FUNCTION 3 S IS MORE DIbCULT TO CALCULATE ALTHOUGH SOME MODELS EXIST ;  = 5SUALLY IT HAS TO BE OBTAINED FROM MEASUREMENTS "Y ASSUM ING LINEARITY OF THE REECTING ECHOES THE ECHO FROM AN ARBITRARY SURFACE CAN BE MODELED AS THE SUM OF THE ECHOES FROM A INNITE NUMBER OF POINTS MAKING UP THE SURFACE ;= (ENCE AN INTERPRETATION OF THE REECTED ECHO IS AS AN OUTPUT OF A LINEAR SYSTEM ;= : R S  E Q S F Q S CQ 

 WHERE E Q S  3Q S c GQ Q S c GI Q S c 3I S IS THE SPATIAL IMPULSE RESPONSE OR POINT SPREAD FUNCTION OF THE TRANSDUCER AND F Q S IS THE RESPONSE OF THE OBJECT DETERMINED BY THE SHAPE OF ITS SURFACE ; =  4WO KINDS OF RESOLUTIONS EXIST IN AN ULTRASONIC IMAGING SYSTEM RANGE AXIAL AND LATERAL RADIAL  4HE FORMER IS USUALLY VERY GOOD WHILE THE LATTER CAN BE POOR )T IS MAINLY DETERMINED BY THE WIDTH OF THE PULSE !LTHOUGH THE ULTRASONIC BEAM IS VERY NARROW AT OR NEAR THE FOCAL LENGTH A NEIGHBORHOOD OF THE INSONIED SURFACE POINT WILL CAUSE ECHOES 4HIS CAN BE SEEN AS A SPATIAL SMOOTHING OR BLURRING (ENCE TO INCREASE THE LATERAL RESOLUTION BY SIGNAL PROCESSING IT IS VERY IMPORTANT TO KNOW THE SPATIAL IMPULSE RESPONSE ; = 3INCE DIRECT MEASUREMENTS ; = ARE SENSITIVE TO NOISE THE REECTOR MUST BE SMALL TO RESEMBLE A POINT  WE PROPOSE AN INDIRECT METHOD ;= "Y MEASURING THE ECHOES FROM SLIDING HALFPLANES THE SPATIAL IMPULSE RESPONSE CAN BE ESTIMATED 4HIS METHOD HAS A LOT IN COMMON WITH TOMOGRAPHY ;  = 4O ESTIMATE THE RELATIVE TIME DELAY OF ARRIVAL OF THE ECHOES THE MODEL SHOWN IN &IGURE  IS OFTEN USED ;= 4HE SIGNAL R J IS CALLED THE REFERENCE SIGNAL TO WHICH ALL

&IGURE  4IME DELAY MODEL OF ULTRASONIC ECHOES TIME DELAYS ARE RELATED M J IS THE OBSERVATION NOISE AND Q J IS THE OBSERVED SIGNAL )DEALLY THIS SHOULD BE A NOISELESS SIGNAL BUT OFTEN A MEASURED ECHO FROM AN ARBITRARY POINT ON THE OBJECTS SURFACE IS USED )F M J IS 'AUSSIAN NOISE THE MAXIMUM LIKELIHOOD ESTIMATOR OF t MAXIMIZES THE CROSS CORRELATION BETWEEN R J AND Q J ;= AND THIS METHOD OR VARIANTS THEREOF ;   = ARE COMMON FOR ESTIMATING THE TIME DELAY t )N THIS THESIS WE PRESENT AN ESTIMATOR ;= WHICH USES THE CORRELATION WITH THE (ILBERT TRANSFORM ;= OF THE REFERENCE ECHO TO ESTIMATE SMALL TIME DELAYS 4HIS ESTIMATOR IS SHOWN TO BE SIMPLE AND FAST

4HESIS OVERVIEW
4HIS THESIS CONSISTS OF SEVEN PARTS COVERING TWO AREAS 0ARTS   DEAL WITH MULTICARRIER MODULATION AND 0ARTS   WITH ULTRASONIC IMAGING !LL PARTS HAVE BEEN PUBLISHED ELSEWHERE AND ARE REPRODUCED HERE IN THEIR ORIGINAL FORM WITH THE FOLLOWING EXCEPTIONS !LL EQUATIONS HAVE BEEN NUMBERED TO MAKE REFERENCING EASY 4HE NUMBERS ON SECTIONS EQUATIONS GURES ETC HAVE BEEN CHANGED TO ALLOW A COMMON NUMBERING THROUGHOUT THE THESIS !LSO A COMMON BIBLIOGRAPHY IS USED FOR THE WHOLE THESIS INSTEAD OF A REFERENCE LIST FOR EACH PART "ELOW IS AN OVERVIEW OF THE THESIS WHICH PLACES THE PARTS INTO A LARGER PERSPECTIVE 0ART  IS AN OVERVIEW OF THOSE AREAS OF /&$- THAT OUR SIGNAL PROCESSING DIVISION IS CURRENT DOING RESEARCHING IN SYNCHRONIZATION CHANNEL ESTIMATION AND CODING 4HIS

4HESIS )NTRODUCTION 3IGNAL -ODELS AND %STIMATION 0ROBLEMS

MAY BE VIEWED AS A GENERAL INTRODUCTION TO /&$- )N 0ART  A SIMULTANEOUS MAXIMUM LIKELIHOOD -, ESTIMATOR IS PRESENTED FOR TIMING AND FREQUENCY OdSETS 4HIS ESTIMATOR USES THE REDUNDANCY IN THE CYCLIC PREX TO PERFORM SYNCHRONIZATION WITHOUT REQUIRING ADDITIONAL PILOTS #HANNEL ESTIMATION IS CONSIDERED IN 0ART  'OOD BUT HIGH COMPLEXITY ESTIMATORS ARE SIMPLIED TO MAKE THEM MORE ATTRACTIVE FROM AN IMPLEMENTATIONAL POINT OF VIEW "Y USING OPTIMAL RANK REDUCTION NEARLY SIMILAR PERFORMANCE LEVELS CAN BE MAINTAINED WITH SIGNICANT REDUCTIONS IN COMPLEXITY 4HIS IS FURTHER ELABORATED ON IN 0ART  WHERE THE MENTIONED ESTIMATOR IS SIMULATED IN A CODED SYSTEM AND COMPARED TO OTHER ESTIMATORS 0ART  INVESTIGATES AN ANALYTICAL METHOD OF CALCULATING THE BIT ERROR PROBABILITY IN A CODED /&$- SYSTEM 4HE ESTIMATION OF THE SPATIAL IMPULSE RESPONSE OF AN ULTRASONIC TRANSDUCER IS DEALT WITH IN 0ART  4HIS IMPORTANT FUNCTIONS CAN BE USED TO ENHANCE THE SPATIAL RESOLUTION IN ULTRASONIC IMAGES 4HIS IS RELATED TO A METHOD FOR QUICKLY ESTIMATING SMALL TIME DELAYS WHICH IS DESCRIBED IN 0ART 

3UMMARY OF CONTRIBUTIONS
"ELOW IS A LIST OF THE MOST IMPORTANT CONTRIBUTIONS OF THIS THESIS q 3YNCHRONIZATION IN TIME AND FREQUENCY CAN BE PERFORMED WITHOUT PILOTS IN /&$SYSTEMS "Y USING THE REDUNDANCY IN THE CYCLIC PREX WE HAVE DERIVED A NOVEL SIMULTANEOUS MAXIMUM LIKELIHOOD -, ESTIMATOR !LTHOUGH THIS ESTIMATOR IS DE RIVED FOR A NON DISPERSIVE CHANNEL SIMULATIONS SHOW THAT FOR A DISPERSIVE CHANNEL THE FREQUENCY SYNCHRONIZER HAS ADEQUATE PERFORMANCE WHILE THE TIME SYNCHRONIZER CAN BE USED IN AN ACQUISITION MODE 0ART  q 3INCE THE CHANNEL IMPULSE RESPONSE IS MUCH SHORTER THAN THE /&$- SYMBOL THE CHANNEL ATTENUATIONS CAN BE VIEWED AS AN OVER SAMPLED STOCHASTIC PROCESS 7E HAVE USED THIS TO DERIVE A NOVEL LOW COMPLEXITY CHANNEL ESTIMATOR BY USING OPTIMAL RANK REDUCTION OF THE LINEAR MINIMUM MEAN SQUARED ERROR ,--3% ESTIMATOR /UR ESTIMATOR MAINTAINS GOOD PERFORMANCE WHILE DRASTICALLY REDUCING THE IMPLE MENTATIONAL COMPLEXITY 0ARTS  AND  q 4O INCREASE LATERAL RESOLUTION IN ULTRASONIC PULSE ECHO IMAGING IT IS IMPORTANT TO KNOW THE SPATIAL IMPULSE RESPONSE OR POINT SPREAD FUNCTION OF THE TRANSDUCER 4O AVOID NOISE SENSITIVE DIRECT MEASUREMENTS OF THIS FUNCTION WE PROPOSE A NOVEL INDIRECT METHOD "Y MEASURING THE ECHOES FROM SLIDING HALFPLANES WHICH ARE MORE ROBUST TO MEASUREMENT NOISE THE SPATIAL IMPULSE RESPONSE CAN BE ESTIMATED 4HIS METHOD IS SIMILAR TO TOMOGRAPHY 0ART 

-ULTICARRIER -ODULATION

0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING

4HIS PART HAS BEEN PUBLISHED AS / %DFORS - 3ANDELL ** VAN DE "EEK $ ,ANDSTRM & 3JBERG !N )NTRODUCTION TO /RTHOG ONAL &REQUENCY $IVISION -ULTIPLEXING 2ESEARCH 2EPORT 45,%!  $IVISION OF 3IGNAL 0ROCESSING ,ULE 5NIVERSITY OF 4ECHNOLOGY



!N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING

!BSTRACT 4HIS REPORT IS AN INTRODUCTION TO ORTHOGONAL FREQUENCY DIVISION MUL TIPLEXING /&$-  4HE FOCUS IS ON SIGNAL PROCESSING AREAS PURSUED BY OUR RESEARCH GROUP 7E PRESENT A HISTORICAL BACKGROUND AND SOME FREQUENTLY USED SYSTEM MOD ELS 4YPICAL AREAS OF APPLICATIONS ARE ALSO DESCRIBED BOTH WIRED AND WIRELESS )N ADDITION TO THE GENERAL OVERVIEW THE ADDRESSED AREAS INCLUDE SYNCHRONIZATION CHANNEL ESTIMATION AND CHANNEL CODING "OTH TIME AND FREQUENCY SYNCHRONIZATION IS DESCRIBED AND THE EdECTS OF SYNCHRONIZATION ERRORS ARE PRESENTED $IdERENT TYPES OF CHANNEL ESTIMATORS ARE DESCRIBED WHERE THE FOCUS IS ON LOW COMPLEXITY ALGORITHMS AND IN THIS CONTEXT ADVANTAGES AND DISADVANTAGES OF COHERENT AND DIdERENTIAL MODULATION ARE ALSO DISCUSSED #HANNEL CODING IS DESCRIBED BOTH FOR WIRELESS AND WIRED SYSTEMS AND POINTERS ARE INCLUDED TO EVALUATION TOOLS AND BIT LODING ALGORITHMS !N EXTENSIVE BIBLIOGRAPHY IS ALSO INCLUDED



)NTRODUCTION

4HE AIM OF THIS REPORT IS TWOFOLD 4HE RST AIM IS TO PROVIDE AN INTRODUCTION TO ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING /&$- SYSTEMS AND SELECTED PARTS OF ITS THEORETICAL BACKGROUND 4HE SECOND AIM IS TO DESCRIBE THE AREAS OF RESEARCH WITHIN /&$- THAT ARE PURSUED AT THE $IVISION OF 3IGNAL 0ROCESSING ,ULE 5NIVERSITY 4HIS ALSO INCLUDES A BY NO MEANS COMPLETE DESCRIPTION OF RELATED WORK THAT MAY BE OF INTEREST 4HE PRESENTATION IS IN THE FORM OF A SINGLE BODY WHERE WE DO NOT SEPARATE OUR OWN WORK FROM THAT BY OTHERS 4HE TECHNOLOGY WE CALL /&$- IN THIS REPORT IS USUALLY VIEWED AS A COLLECTION OF TRANSMISSION TECHNIQUES 7HEN APPLIED IN A WIRELESS ENVIRONMENT SUCH AS RADIO BROAD CASTING IT IS USUALLY REFERRED TO AS /&$- (OWEVER IN A WIRED ENVIRONMENT SUCH AS IN ASYMMETRIC DIGITAL SUBSCRIBER LINES !$3,  THE TERM DISCRETE MULTITONE $-4 IS MORE APPROPRIATE 4HROUGHOUT THIS REPORT WE ONLY USE THE TERM $-4 WHEN EXPLICITLY AD DRESSING THE WIRED ENVIRONMENT &URTHER THE TWO TERMS SUBCARRIER AND SUBCHANNEL WILL BE USED INTERCHANGEABLY 4HE HISTORY OF /&$- HAS BEEN ADDRESSED SEVERAL TIMES IN THE LITERATURE SEE EG ; = WHICH WE HAVE CONDENSED TO THE BRIEF OVERVIEW BELOW 4HE HISTORY OF /&$- DATES BACK TO THE MID S WHEN #HANG PUBLISHED HIS PAPER ON THE SYNTHESIS OF BANDLIMITED SIGNALS FOR MULTICHANNEL TRANSMISSION ;= (E PRESENTS A PRINCIPLE FOR TRANSMITTING MESSAGES SIMULTANEOUSLY THROUGH A LINEAR BANDLIMITED CHAN NEL WITHOUT INTERCHANNEL )#) AND INTERSYMBOL INTERFERENCE )3)  3HORTLY AFTER #HANG PRESENTED HIS PAPER 3ALTZBERG PERFORMED AN ANALYSIS OF THE PERFORMANCE ;= WHERE HE CONCLUDED THAT THE STRATEGY OF DESIGNING AN EbCIENT PARALLEL SYSTEM SHOULD CONCENTRATE MORE ON REDUCING CROSSTALK BETWEEN ADJACENT CHANNELS THAN ON PERFECTING THE INDIVIDUAL CHANNELS THEMSELVES SINCE THE DISTORTIONS DUE TO CROSSTALK TEND TO DOMINATE 4HIS IS AN IMPORTANT CONCLUSION WHICH HAS PROVEN CORRECT IN THE DIGITAL BASEBAND PROCESSING THAT



0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING

EMERGED A FEW YEARS LATER ! MAJOR CONTRIBUTION TO /&$- WAS PRESENTED IN  BY 7EINSTEIN AND %BERT ;= WHO USED THE DISCRETE &OURIER TRANSFORM $&4 TO PERFORM BASEBAND MODULATION AND DEMODULATION 4HIS WORK DID NOT FOCUS ON PERFECTING THE INDIVIDUAL CHANNELS BUT RATHER ON INTRODUCING EbCIENT PROCESSING ELIMINATING THE BANKS OF SUBCARRIER OSCILLATORS 4O COMBAT )3) AND )#) THEY USED BOTH A GUARD SPACE BETWEEN THE SYMBOLS AND RAISED COSINE WINDOWING IN THE TIME DOMAIN 4HEIR SYSTEM DID NOT OBTAIN PERFECT ORTHOGONALITY BETWEEN SUBCARRIERS OVER A DISPERSIVE CHANNEL BUT IT WAS STILL A MAJOR CONTRIBUTION TO /&$- !NOTHER IMPORTANT CONTRIBUTION WAS DUE TO 0ELED AND 2UIZ IN  ;= WHO IN TRODUCED THE CYCLIC PREX #0 OR CYCLIC EXTENSION SOLVING THE ORTHOGONALITY PROBLEM )NSTEAD OF USING AN EMPTY GUARD SPACE THEY LLED THE GUARD SPACE WITH A CYCLIC EXTENSION OF THE /&$- SYMBOL 4HIS EdECTIVELY SIMULATES A CHANNEL PERFORMING CYCLIC CONVOLU TION WHICH IMPLIES ORTHOGONALITY OVER DISPERSIVE CHANNELS WHEN THE #0 IS LONGER THAN THE IMPULSE RESPONSE OF THE CHANNEL 4HIS INTRODUCES AN ENERGY LOSS PROPORTIONAL TO THE LENGTH OF THE #0 BUT THE ZERO )#) GENERALLY MOTIVATES THE LOSS /&$- SYSTEMS ARE USUALLY DESIGNED WITH RECTANGULAR PULSES BUT RECENTLY THERE HAS BEEN AN INCREASED INTEREST IN PULSE SHAPING ;  = "Y USING PULSES OTHER THAN RECTANGULAR THE SPECTRUM CAN BE SHAPED TO BE MORE WELL LOCALIZED IN FREQUENCY WHICH IS BENECIAL FROM AN INTERFERENCE POINT OF VIEW /&$- IS CURRENTLY USED IN THE %UROPEAN DIGITAL AUDIO BROADCASTING $!" STANDARD ;= 3EVERAL $!" SYSTEMS PROPOSED FOR .ORTH !MERICA ARE ALSO BASED ON /&$- ;= AND ITS APPLICABILITY TO DIGITAL 46 BROADCASTING IS CURRENTLY BEING INVESTIGATED ;    = /&$- IN COMBINATION WITH MULTIPLE ACCESS TECHNIQUES ARE SUBJECT TO SIGNICANT INVESTIGATION SEE EG ;    = /&$- UNDER THE NAME $-4 HAS ALSO ATTRACTED A GREAT DEAL OF ATTENTION AS AN EbCIENT TECHNOLOGY FOR HIGH SPEED TRANSMISSION ON THE EXISTING TELEPHONE NETWORK SEE EG ;   = 4HIS REPORT IS ORGANIZED AS FOLLOWS )N 3ECTION  WE PRESENT COMMON /&$- MODELS INCLUDING CONTINUOUS TIME AND DISCRETE TIME %NVIRONMENTS IN WHICH /&$- SYSTEMS ARE EXPECTED TO WORK ARE SUMMARIZED IN 3ECTION  3YNCHRONIZATION PROBLEMS AND PROPOSED SOLUTION ARE PRESENTED IN 3ECTION  #HANNEL ESTIMATION IS ELABORATED ON IN 3ECTION  AND CODING IN BOTH WIRELESS AND WIRED /&$- SYSTEMS IS DISCUSSED IN 3ECTION  &INALLY IN 3ECTION  WE DISCUSS AND SUMMARIZE THE CONTENTS OF THIS REPORT



3YSTEM MODELS

4HE BASIC IDEA OF /&$- IS TO DIVIDE THE AVAILABLE SPECTRUM INTO SEVERAL SUBCHANNELS SUBCARRIERS  "Y MAKING ALL SUBCHANNELS NARROWBAND THEY EXPERIENCE ALMOST AT FADING WHICH MAKES EQUALIZATION VERY SIMPLE 4O OBTAIN A HIGH SPECTRAL EbCIENCY THE FREQUENCY RESPONSE OF THE SUBCHANNELS ARE OVERLAPPING AND ORTHOGONAL HENCE THE NAME /&$- 4HIS ORTHOGONALITY CAN BE COMPLETELY MAINTAINED EVEN THOUGH THE SIGNAL PASSES THROUGH A TIME DISPERSIVE CHANNEL BY INTRODUCING A CYCLIC PREX 4HERE ARE SEVERAL VERSIONS OF /&$- SEE EG ;  = BUT WE FOCUS ON SYSTEMS USING SUCH A CYCLIC PREX ;= ! CYCLIC PREX IS A COPY OF THE LAST PART OF THE /&$- SYMBOL WHICH IS PREPENDED TO THE

 TRANSMITTED SYMBOL SEE &IGURE  4HIS MAKES THE TRANSMITTED SIGNAL PERIODIC WHICH

&IGURE  4HE CYCLIC PREX IS A COPY OF THE LAST PART OF THE /&$- SYMBOL PLAYS A DECISIVE ROLL IN AVOIDING INTERSYMBOL AND INTERCARRIER INTERFERENCE ;= 4HIS IS EXPLAINED LATER IN THIS SECTION !LTHOUGH THE CYCLIC PREX INTRODUCES A LOSS IN SIGNAL TO NOISE RATIO 3.2  IT IS USUALLY A SMALL PRICE TO PAY TO MITIGATE INTERFERENCE ! SCHEMATIC DIAGRAM OF A BASEBAND /&$- SYSTEM IS SHOWN IN &IGURE 
4RANSMITTER
WK WK

#HANNEL
E S M

2ECEIVER
QS QXJ`

3 % # (

7 4 ,

4G

RXJ`

52

RS

F~ & S

25

4G

7 4 , $ #

XK XK

3 % #
X- ` K

W- ` K

&IGURE  ! DIGITAL IMPLEMENTATION OF A BASEBAND /&$- SYSTEM #0 AND #0 DENOTE THE INSERTION AND DELETION OF THE CYCLIC PREX RESPECTIVELY &OR THIS SYSTEM WE EMPLOY THE FOLLOWING ASSUMPTIONS q ! CYCLIC PREX IS USED q 4HE IMPULSE RESPONSE OF THE CHANNEL IS SHORTER THAN THE CYCLIC PREX q 4RANSMITTER AND RECEIVER ARE PERFECTLY SYNCHRONIZED q #HANNEL NOISE IS ADDITIVE WHITE AND COMPLEX 'AUSSIAN q 4HE FADING IS SLOW ENOUGH FOR THE CHANNEL TO BE CONSIDERED CONSTANT DURING ONE /&$- SYMBOL INTERVAL 4HE DIbCULTIES IN A COMPLETE ANALYSIS OF THIS SYSTEM MAKE IT RATHER AWKWARD FOR THEORETICAL STUDIES 4HEREFORE IT IS COMMON PRACTICE TO USE SIMPLIED MODELS RESULTING IN A TRACTABLE ANALYSIS 7E CLASSIFY THESE /&$- SYSTEM MODELS INTO TWO DIdERENT CLASSES CONTINUOUS TIME AND DISCRETE TIME



#ONTINUOUS TIME MODEL

4HE RST /&$- SYSTEMS DID NOT EMPLOY DIGITAL MODULATION AND DEMODULATION (ENCE THE CONTINUOUS TIME /&$- MODEL PRESENTED BELOW CAN BE CONSIDERED AS THE IDEAL /&$-



0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING

SYSTEM WHICH IN PRACTICE IS DIGITALLY SYNTHESIZED 3INCE THIS IS THE RST MODEL DESCRIBED WE MOVE THROUGH IT IN A STEP BY STEP FASHION 7E START WITH THE WAVEFORMS USED IN THE TRANSMITTER AND PROCEED ALL THE WAY TO THE RECEIVER 4HE BASEBAND MODEL IS SHOWN IN &IGURE 
4RANSMITTER
W W W
K

#HANNEL
E S M

K

S S
`  S

RS

F~ & S

QS

2ECEIVER K + 3  S  S

`  S

X X X

K

K

` K

` K

&IGURE  "ASE BAND /&$- SYSTEM MODEL q 4RANSMITTER !SSUMING AN /&$- SYSTEM WITH - SUBCARRIERS A BANDWIDTH OF 6 (Z AND SYMBOL LENGTH OF 3 SECONDS OF WHICH 3fy SECONDS IS THE LENGTH OF THE CYCLIC PREX THE TRANSMITTER USES THE FOLLOWING WAVEFORMS t S  P 
Q Qfy


Ds { E tQfy IF S  : 3 <

T

OTHERWISE



WHERE 3  -6
3fy  .OTE THAT t S  t S
-6 WHEN S IS WITHIN THE CYCLIC PREX : 3fy < 3INCE t S IS A RECTANGULAR PULSE MODULATED ON THE CARRIER FREQUENCY J6-  THE COMMON INTERPRETATION OF /&$- IS THAT IT USES - SUBCARRIERS EACH CARRYING A LOW BIT RATE 4HE WAVEFORMS t S ARE USED IN THE MODULATION AND THE TRANSMITTED BASE BAND SIGNAL FOR /&$- SYMBOL NUMBER K IS Ru S 
E  8 t(

Wt&u t S ` K3 



WHERE W&u  W&u     WE &u ARE COMPLEX NUMBERS FROM A SET OF SIGNAL CONSTELLATION POINTS 7HEN AN INNITE SEQUENCE OF /&$- SYMBOLS IS TRANSMITTED THE OUTPUT FROM THE TRANSMITTER IS A JUXTAPOSITION OF INDIVIDUAL /&$- SYMBOLS RS 

 8
u(

Ru S 

 E  8 8
u(

 t

Wt&u t S ` K3 



(

q 0HYSICAL CHANNEL 7E ASSUME THAT THE SUPPORT OF THE POSSIBLY TIME VARIANT IMPULSE RESPONSE F ~  S OF THE PHYSICAL CHANNEL IS RESTRICTED TO THE INTERVAL ~  : 3fy < IE TO THE LENGTH OF THE CYCLIC PREX 4HE RECEIVED SIGNAL BECOMES : Qfy Q S  FcR S  F ~  S R S ` ~ C~
M ES 


 WHERE M E S IS ADDITIVE WHITE AND COMPLEX 'AUSSIAN CHANNEL NOISE q 2ECEIVER 4HE /&$- RECEIVER CONSISTS OF A LTER BANK MATCHED TO THE LAST PART :3fy  3 < OF THE TRANSMITTER WAVEFORMS t S  IE | t S  t 3 `S  IF S  : 3 ` 3fy <  OTHERWISE 

%dECTIVELY THIS MEANS THAT THE CYCLIC PREX IS REMOVED IN THE RECEIVER 3INCE THE CYCLIC PREX CONTAINS ALL )3) FROM THE PREVIOUS SYMBOL THE SAMPLED OUTPUT FROM THE RECEIVER LTER BANK CONTAINS NO )3) (ENCE WE CAN IGNORE THE TIME INDEX K WHEN CALCULATING THE SAMPLED OUTPUT AT THE J TH MATCHED LTER "Y USING    AND   WE GET : Xt   Q c t S J(Q  : :
Q Qfy

: Q

Qfy Qfy

F ~ S

 Q  E 8


S t 3 ` S CS
 

M E 3 ` S t S CS

t (

Wt t S ` ~

C~

t S CS


7E CONSIDER THE CHANNEL TO BE XED OVER THE /&$- SYMBOL INTERVAL AND DENOTE IT BY F ~  WHICH GIVES Xt 
E  8

: Q

t ( Qfy

Wt

: Q t: Qfy


u F ~ t S ` ~ C~


Qfy

t S CS


M E 3 ` S t S CS

4HE INTEGRATION INTERVALS ARE 3fy  S  3 AND   ~  3fy WHICH IMPLIES THAT   S ` ~  3 AND THE INNER INTEGRAL CAN BE WRITTEN AS : Qfy


Ds {t ~ Qfy T(E P F ~ t S ` ~ C~  F ~ C~ 3 ` 3fy  : Ds {t Qfy T(E Qfy P  F ~ Ds {t ~T(E C~ 3 ` 3fy  : Qfy
   

3fy  S   3

4HE LATTER PART OF THIS EXPRESSION IS THE SAMPLED FREQUENCY RESPONSE OF THE CHANNEL AT FREQUENCY E  J  6-  IE AT THE J  TH SUBCARRIER FREQUENCY t u : Qfy 6  Gt  & J  F ~ Ds {t ~T(E C~ 
 





0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING WHERE & E IS THE &OURIER TRANSFORM OF F ~  5SING THIS NOTATION THE OUTPUT FROM THE RECEIVER LTER BANK CAN BE SIMPLIED TO : Q s {t Qfy T(E : Q D P Xt  Wt Gt t S CS
M E 3 ` S t S CS 3 ` 3 Qfy Qfy fy t ( : Q E  8  Wt Gt t S t S CS
Mt 
E  8
     

t (


Qfy



2Q WHERE Mt  Qfy M E 3 ` S t S CS 3INCE THE TRANSMITTER LTERS t S ARE ORTHOGO NAL : Q
Qfy

t S


: Q s {t Qfy T(E s {tQfy T(E D D P P S CS  CS  p :J ` J <   3 ` 3 3 ` 3 Qfy fy fy



WHERE p :J < IS THE +RONECKER DELTA FUNCTION ;= WE CAN SIMPLIFY  AND OBTAIN Xt  Gt Wt
Mt  WHERE Mt IS ADDITIVE WHITE 'AUSSIAN NOISE !7'.  4HE BENET OF A CYCLIC PREX IS TWOFOLD IT AVOIDS BOTH )3) SINCE IT ACTS AS A GUARD SPACE AND )#) SINCE IT MAINTAINS THE ORTHOGONALITY OF THE SUBCARRIERS  "Y RE INTRODUCING THE TIME INDEX K WE MAY NOW VIEW THE /&$- SYSTEM AS A SET OF PARALLEL 'AUSSIAN CHANNELS ACCORDING TO &IGURE  

G W
K

K

M

K

X G ` M ` 
K K

K

W `
-

K

X `
-

K

&IGURE  4HE CONTINUOUS TIME /&$- SYSTEM INTERPRETED AS PARALLEL 'AUSSIAN CHANNELS !N EdECT TO CONSIDER AT THIS STAGE IS THAT THE TRANSMITTED ENERGY INCREASES WITH THE LENGTH OF THE CYCLIC PREX WHILE THE EXPRESSIONS FOR THE RECEIVED AND SIG 2 SAMPLED  NALS  STAY THE SAME 4HE TRANSMITTED ENERGY PER SUBCARRIER IS Jt S J CS  3  3 ` 3fy  AND THE 3.2 LOSS BECAUSE OF THE DISCARDED CYCLIC PREX IN THE RECEIVER BECOMES 2-1ux  ` KNF  ` o  

 WHERE o  3fy 3 IS THE RELATIVE LENGTH OF THE CYCLIC PREX 4HE LONGER THE CYCLIC PREX THE LARGER THE 3.2 LOSS 4YPICALLY THE RELATIVE LENGTH OF THE CYCLIC PREX IS SMALL AND THE )#) AND )3) FREE TRANSMISSION MOTIVATES THE 3.2 LOSS LESS THAN  D" FOR o    &IGURE  DISPLAYS A SCHEMATIC PICTURE OF THE FREQUENCY RESPONSE OF THE INDIVIDUAL SUBCHANNELS IN AN /&$- SYMBOL )N THIS GURE THE INDIVIDUAL SUBCHANNELS OF THE SYSTEM ARE SEPARATED 4HE RECTANGULAR WINDOWING OF THE TRANSMITTED PULSES RESULTS IN A SINC
- RTAB@QQHDQR 2O@BHMF aE  6-

&IGURE  ! SYMBOLIC PICTURE OF THE INDIVIDUAL SUBCHANNELS FOR AN /&$- SYSTEM WITH - TONES OVER A BANDWIDTH 6  SHAPED FREQUENCY RESPONSE FOR EACH CHANNEL 4HUS THE POWER SPECTRUM OF THE /&$SYSTEM DECAYS AS E   )N SOME CASES THIS IS NOT SUbCIENT AND METHODS HAVE BEEN PROPOSED TO SHAPE THE SPECTRUM )N ;= A RAISED COSINE PULSE IS USED WHERE THE ROLL Od REGION ALSO ACTS AS A GUARD SPACE SEE &IGURE  )F THE AT PART IS THE /&$- SYMBOL INCLUDING THE

&IGURE  0ULSE SHAPING USING THE RAISED COSINE FUNCTION 4HE GRAY PARTS OF THE SIGNAL INDICATE THE EXTENSIONS CYCLIC PREX BOTH )#) AND )3) ARE AVOIDED 4HE SPECTRUM WITH THIS KIND OF PULSE SHAPING IS SHOWN IN &IGURE  WHERE IT IS COMPARED WITH A RECTANGULAR PULSE 4HE OVERHEAD INTRODUCED BY AN EXTRA GUARD SPACE WITH A GRACEFUL ROLL Od CAN BE A GOOD INVESTMENT SINCE THE SPECTRUM FALLS MUCH MORE QUICKLY AND REDUCES THE INTERFERENCE TO ADJACENT FREQUENCY BANDS /THER TYPES OF PULSE SHAPING SUCH AS OVERLAPPING ;= AND WELL LOCALIZED PULSES ; = HAVE ALSO BEEN INVESTIGATED



0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING

&IGURE  3PECTRUM WITH RECTANGULAR PULSE SOLID AND RAISED COSINE PULSE DASHED 



$ISCRETE TIME MODEL

!N ENTIRELY DISCRETE TIME MODEL OF AN /&$- SYSTEM IS DISPLAYED IN &IGURE  #OM PARED TO THE CONTINUOUS TIME MODEL THE MODULATION AND DEMODULATION ARE REPLACED BY AN INVERSE $&4 )$&4 AND A $&4 RESPECTIVELY AND THE CHANNEL IS A DISCRETE TIME CONVOLUTION 4HE CYCLIC PREX OPERATES IN THE SAME FASHION IN THIS SYSTEM AND THE CAL CULATIONS CAN BE PERFORMED IN ESSENTIALLY THE SAME WAY 4HE MAIN DIdERENCE IS THAT ALL INTEGRALS ARE REPLACED BY SUMS
4RANSMITTER
WK WK

#HANNEL
E :J< M

2ECEIVER
Q:J<

3 % # (

7 4 ,

"/

R:J<

F:LJ<

"/

7 4 , $ #

XK XK

3 % #
X- ` K

W- ` K

&IGURE  $ISCRETE TIME /&$- SYSTEM &ROM THE RECEIVERS POINT OF VIEW THE USE OF A CYCLIC PREX LONGER THAN THE CHANNEL WILL TRANSFORM THE LINEAR CONVOLUTION IN THE CHANNEL TO A CYCLIC CONVOLUTION $ENOTING CYCLIC CONVOLUTION BY ] WE CAN WRITE THE WHOLE /&$- SYSTEM AS Eu Xu  #%3 (#%3 Wu ] Fu
M  #%3 (#%3 Wu ] Fu
Mu  

WHERE Xu CONTAINS THE - RECEIVED DATA POINTS Wu THE - TRANSMITTED CONSTELLATION POINTS Fu THE CHANNEL IMPULSE RESPONSE OF THE CHANNEL PADDED WITH ZEROS TO OBTAIN A LENGTH OF Eu THE CHANNEL NOISE 3INCE THE CHANNEL NOISE IS ASSUMED WHITE AND 'AUSSIAN -  AND M Eu REPRESENTS UNCORRELATED 'AUSSIAN NOISE &URTHER WE USE THAT THE THE TERM Mu  #%3 M $&4 OF TWO CYCLICALLY CONVOLVED SIGNALS IS EQUIVALENT TO THE PRODUCT OF THEIR INDIVIDUAL $&4S $ENOTING ELEMENT BY ELEMENT MULTIPLICATION BY a THE ABOVE EXPRESSION CAN BE WRITTEN Xu  Wu a #%3 Fu
Mu  Wu a Gu
Mu  

 WHERE Gu  #%3 Fu IS THE FREQUENCY RESPONSE OF THE CHANNEL 4HUS WE HAVE OBTAINED THE SAME TYPE OF PARALLEL 'AUSSIAN CHANNELS AS FOR THE CONTINUOUS TIME MODEL 4HE ONLY DIdERENCE IS THAT THE CHANNEL ATTENUATIONS Gu ARE GIVEN BY THE - POINT $&4 OF THE DISCRETE TIME CHANNEL INSTEAD OF THE SAMPLED FREQUENCY RESPONSE AS IN  



! TIME FREQUENCY INTERPRETATION

4HE MODELS DESCRIBED ABOVE ARE TWO CLASSICAL MODELS OF /&$- WITH A CYCLIC PREX ! MORE GENERAL MODEL SUITABLE FOR EG PULSE SHAPING IS TO VIEW /&$- AS TRANSMISSION OF DATA IN A LATTICE IN THE TIME FREQUENCY PLANE #ONSIDER RST A TRANSMITTED /&$- SIGNAL RS 8 RS  Wt&u t&u S  
t&u

WHERE THE FUNCTIONS t&u S ARE TRANSLATIONS IN TIME BY ~ AND IN FREQUENCY BY y OF THE PROTOTYPE FUNCTION O S  IE t&u S  O S ` K~ Ds {ty   4HIS CREATES A TWO DIMENSIONAL  $ LATTICE IN THE TIME FREQUENCY PLANE ; = SEE &IGURE  5SUALLY THE PROTOTYPE FUNCTION IS CHOSEN AS THE RECTANGULAR WINDOW O S 

&IGURE  ,ATTICE IN THE TIME FREQUENCY PLANE 4HE DATA SYMBOLS Wt&u ARE TRANSMITTED AT THE LATTICE POINTS v S v ~  4HE SPACING IN THE FREQUENCY DIRECTION IS THEN y   ~ ` 3fy  WHERE 3fy IS THE LENGTH OF THE CYCLIC PREX &OR A DISCUSSION ON THE IMPACT OF PROTOTYPE FUNCTIONS SEE !PPENDIX ! %ACH TRANSMITTED DATA SYMBOL IN THE LATTICE EXPERIENCES AT FADING SEE   WHICH SIMPLIES EQUALIZATION AND CHANNEL ESTIMATION 4HE CHANNEL ATTENUATIONS AT THE LATTICE POINTS ARE CORRELATED AND BY TRANSMITTING KNOWN SYMBOLS AT SOME POSTIONS THE CHANNEL ATTENUATIONS CAN BE ESTIMATED WITH AN INTERPOLATION LTER ;  = 4HIS IS A  $ VERSION OF PILOT SYMBOL ASSISTED MODULATION WHICH HAS BEEN PROPOSED FOR SEVERAL WIRELESS /&$- SYSTEMS SEE EG ;  = ! MORE DETAILED DESCRIPTION OF /&$- CHANNEL ESTIMATION IS GIVEN IN 3ECTION 


y~  



0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING



)MPERFECTIONS

$EPENDING ON THE ANALYZED SITUATION IMPERFECTIONS IN A REAL /&$- SYSTEM MAY BE IGNORED OR EXPLICITLY INCLUDED IN THE MODEL "ELOW WE MENTION SOME OF THE IMPERFECTIONS AND THEIR CORRESPONDING EdECTS q $ISPERSION "OTH TIME AND FREQUENCY DISPERSION OF THE CHANNEL CAN DESTROY THE ORTHOGONALITY OF THE SYSTEM IE INTRODUCE BOTH )3) AND )#) ;= )F THESE EdECTS ARE NOT SUbCIENTLY MITIGATED BY EG A CYCLIC PREX AND A LARGE INTER CARRIER SPACING THEY HAVE TO BE INCLUDED IN THE MODEL /NE WAY OF MODELLING THESE EdECTS IS AN INCREASE OF THE ADDITIVE NOISE ;= q .ONLINEARITIES AND CLIPPING DISTORTION /&$- SYSTEMS HAVE HIGH PEAK TO AVERAGE POWER RATIOS AND HIGH DEMANDS ON LINEAR AMPLIERS ;= .ONLINEARITIES IN AMPLIERS MAY CAUSE BOTH )3) AND )#) IN THE SYSTEM %SPECIALLY IF THE AMPLIERS ARE NOT DESIGNED WITH PROPER OUTPUT BACK Od /"/  THE CLIPPING DISTORTION MAY CAUSE SEVERE DEGRADATION 4HESE EdECTS HAVE BEEN ADDRESSED IN EG ;   = 3PECIAL CODING STRATEGIES WITH THE AIM TO MINIMIZE PEAK TO AVERAGE POWER RATIOS HAVE ALSO BEEN SUGGESTED SEE EG ;  = q %XTERNAL INTERFERENCE "OTH WIRELESS AND WIRED /&$- SYSTEMS SUdER FROM EXTERNAL INTERFERENCE )N WIRE LESS SYSTEMS THIS INTERFERENCE USUALLY STEMS FROM RADIO TRANSMITTERS AND OTHER TYPES ELECTRONIC EQUIPMENT IN THE VINCINITY OF THE RECEIVER )N WIRED SYSTEMS THE LIMIT ING FACTOR IS USUALLY CROSSTALK WHICH IS DISCUSSED IN MORE DETAIL IN 3ECTION  )NTERFERENCE CAN BE INCLUDED IN THE MODEL AS EG COLOURED NOISE



3YSTEM ENVIRONMENTS

4WO MAJOR GROUPS OF COMMUNICATION SYSTEMS ARE THOSE WHO OPERATE IN WIRELESS AND WIRED ENVIRONMENTS &OR INSTANCE WHEN DESIGNING A WIRELESS /&$- SYSTEM THE FADING CHANNEL IS USUALLY A MAJOR OBSTACLE WHILE FOR A WIRED /&$- AKA $-4 SYSTEM CROSSTALK AND IMPULSIVE NOISE ARE MORE DIbCULT TO HANDLE )N THE FOLLOWING SECTIONS WE BRIEY DISCUSS THE WIRELESS AND WIRED ENVIRONMENTS



7IRELESS SYSTEMS

)N WIRELESS SYSTEMS RADIO SYSTEMS  CHANGES IN THE PHYSICAL ENVIRONMENT CAUSE THE CHAN NEL TO FADE 4HESE CHANGES INCLUDE BOTH RELATIVE MOVEMENT BETWEEN TRANSMITTER AND RECEIVER AND MOVING SCATTERERSREECTORS IN THE SURROUNDING SPACE 7HEN DEVELOPING NEW STANDARDS FOR WIRELESS SYSTEMS CHANNEL MODELS ARE USUALLY CLAS SIED ACCORDING TO THE ENVIRONMENT IN WHICH THE RECEIVER OPERATES 4HESE ENVIRONMENTS

 ARE OFTEN DESCRIBED IN TERMS LIKE 2URAL AREA "USINESS INDOOR ETC -ODELS OF THIS TYPE ARE SPECIED BY EG THE %UROPEAN TELECOMMUNICATIONS STANDARDS INSTITUTE %43) ; = )N THEORETICAL STUDIES OF WIRELESS SYSTEMS THE CHANNEL MODELS ARE USUALLY CHOSEN SO THAT THEY RESULT IN A TRACTABLE ANALYSIS 4HE TWO MAJOR CLASSES OF FADING CHARACTERISTICS ARE KNOWN AS 2AYLEIGH AND 2ICIAN ;= ! 2AYLEIGH FADING ENVIRONMENT ASSUMES NO LINE OF SIGHT AND NO XED REECTORSSCATTERERS 4HE EXPECTED VALUE OF THE FADING IS ZERO )F THERE IS A LINE OF SIGHT THIS CAN BE MODELLED BY 2ICIAN FADING WHICH HAS THE SAME CHARACTERISTICS AS THE 2AYLEIGH FADING EXCEPT FOR A NON ZERO EXPECTED VALUE /FTEN PROPERTIES OF A THEORETICAL MODEL ARE CHARACTERIZED BY ONLY A FEW PARAMETERS SUCH AS POWER DELAY PROLE AND MAXIMAL $OPPLER FREQUENCY 4HE POWER DELAY PROLE | a DEPENDS ON THE ENVIRONMENT AND A COMMON CHOICE IS THE EXPONENTIALLY DECAYING PROLE | ~  D~(~v  

WHERE ~ IS THE TIME DELAY AND ~v IS THE ROOT MEAN SQUARED 2-3 VALUE OF THE POWER DELAY PROLE 3EVERAL OTHER CHOICES ARE POSSIBLE SEE EG ;= 4HE MAXIMAL $OPPLER FREQUENCY Eg&vd CAN BE DETERMINED BY U Eg&vd  Ef  B 

WHERE THE CARRIER FREQUENCY IS Ef (Z THE SPEED OF THE RECEIVER IS U MS AND THE SPEED OF LIGHT IS B {  b # MS )SOTROPIC SCATTERING IS COMMONLY ASSUMED IE THE RECEIVED SIGNAL POWER IS SPREAD UNIFORMLY OVER ALL ANGLES OF ARRIVAL WHICH RESULTS IN A 5 SHAPED $OPPLER SPECTRUM 4HIS IS USUALLY REFERRED TO AS A *AKES SPECTRUM ;= AND IS DETERMINED BY THE MAXIMAL $OPPLER FREQUENCY "EFORE WE START DISCUSSING THE DIdERENT SCENARIOS ENCOUNTERED IN WIRELESS SYSTEMS THERE ARE A FEW THINGS THAT MAY BE SAID ABOUT /&$- ON FADING CHANNELS IN GENERAL  4HE INTER CARRIER SPACING OF THE SYSTEM HAS TO BE CHOSEN LARGE COMPARED TO THE MAXIMAL $OPPLER FREQUENCY OF THE FADING CHANNEL TO KEEP THE )#) SMALL ; = 4HIS IS FURTHER DISCUSSED IN !PPENDIX !  )F THE ORTHOGONALITY OF THE SYSTEM IS MAINTAINED THE BASIC /&$- STRUCTURE DOES NOT NECESSITATE TRADITIONAL EQUALIZING (OWEVER TO EXPLOIT THE DIVERSITY OF THE CHANNEL PROPER CODING AND INTERLEAVING IS REQUIRED ;= 7E HAVE CHOSEN TO DISCUSS THE WIRELESS ENVIRONMENT IN TWO CONTEXTS THE TRANSMISSION FROM A BASE STATION TO MOBILE TERMINALS DOWNLINK AND THE TRANSMISSION FROM MOBILE TERMINALS TO A BASE STATION UPLINK  4HE REASON FOR THE CHOSEN CONTEXTS IS THAT ONE OR BOTH USUALLY ARE REPRESENTED IN EVERY WIRELESS SYSTEM AND THEY REQUIRE QUITE DIdERENT DESIGN STRATEGIES 4HE MOST FREQUENTLY DISCUSSED WIRELESS /&$- SYSTEMS ARE FOR BROADCASTING EG DIGITAL AUDIO AND DIGITAL VIDEO AND ONLY CONTAIN A DOWNLINK SINCE THERE IS NO RETURN CHANNEL #ELLULAR SYSTEMS ON THE OTHER HAND HAVE BOTH A DOWNLINK AND AN UPLINK

 

0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING $OWNLINK

! SCHEMATIC PICTURE OF THE DOWNLINK ENVIRONMENT IS SHOWN IN &IGURE  )N THIS CASE MOBILE TERMINAL NUMBER M RECEIVES THE SIGNAL R S TRANSMITTED FROM THE BASE STATION THROUGH ITS OWN CHANNEL Fw S  AND THE RECEIVED SIGNAL Qw S IS GIVEN BY Qw S  R c Fw S  

#H AN NE L

NEL  #HAN

#HAN NEL +

4ERMINAL 

"ASE STATION

4ERMINAL +

4ERMINAL 

&IGURE  4HE WIRELESS DOWNLINK ENVIRONMENT 4HIS ENVIRONMENT IMPLIES THAT EACH RECEIVER TERMINAL ONLY HAS TO SYNCHRONIZE TO THE BASE STATION AND FROM ITS POINT OF VIEW THE OTHER TERMINALS DO NOT EXIST 4HIS MAKES SYNCHRONIZATION RELATIVELY EASY AND ALL PILOT INFORMATION TRANSMITTED FROM THE BASE STATION CAN BE USED FOR CHANNEL ESTIMATION AND SYNCHRONIZATION 4HE DOWNLINK ENVIRONMENT HAS BEEN THOROUGHLY INVESTIGATED SEE EG ;    =  ,ARGE PORTIONS OF WORK PRESENTED ON SYSTEMS OF THIS KIND HAVE BEEN CONCERNED WITH DIGITAL AUDIO SEE EG ;    = AND DIGITAL VIDEO SEE EG ;      = BROADCASTING  5PLINK

! SCHEMATIC PICTURE OF THE UPLINK ENVIRONMENT IS SHOWN IN &IGURE  )N THIS CASE THE BASE STATION RECEIVES THE TRANSMITTED SIGNAL Rw S FROM MOBILE TERMINAL M THROUGH CHANNEL Fw S  AND THE TOTAL RECEIVED SIGNAL Q S AT THE BASE STATION IS A SUPERPOSITION Q S 
B 8 w(

Rw c Fw S



OF SIGNALS FROM ALL MOBILE TERMINALS 4HE MAJOR PROBLEM HERE IS THE SUPERPOSITION OF SIGNALS ARRIVING THROUGH DIdERENT CHANNELS &OR THE BASE STATION TO BE ABLE TO SEPARATE THE SIGNALS FROM EACH RECEIVER A SUbCIENT ORTHOGONALITY BETWEEN RECEIVED SIGNALS FROM DIdERENT TERMINALS HAS TO BE ACHIEVED 3EVERAL METHODS FOR OBTAINING THIS HAVE BEEN PROPOSED 4HESE INCLUDE COMBI NATIONS OF /&$- AND CODE DIVISION TIME DIVISION AND FREQUENCY DIVISION MULTIPLE ACCESS #$-! 4$-! AND &$-! RESPECTIVELY  !LL THREE HAVE BEEN PROPOSED IN ;= AND &$-!/&$- IS CURRENTLY UNDER INVESTIGATION IN EG ;  =


NEL  #HAN

#HAN NEL

#H AN NE L

4ERMINAL 

"ASE STATION

4ERMINAL +

4ERMINAL 

&IGURE  4HE WIRELESS UPLINK ENVIRONMENT )NDEPENDENT OF THE METHOD CHOSEN TO SEPARATE SIGNALS FROM DIdERENT TERMINALS THE SYSTEM SYNCHRONIZATION IS ONE OF THE MAJOR DESIGN ISSUES 4O AVOID INTERFERENCE ALL MO BILE TERMINALS HAVE TO BE JOINTLY SYNCHRONIZED TO THE BASE STATION &URTHER IF COHERENT MODULATION IS USED AS IN ;  = THE DIdERENT CHANNELS FROM THE USERS HAVE TO BE ESTIMATED SEPARATELY



7IRED SYSTEMS

7HEN STUDYING WIRED COMMUNICATION SYSTEMS AND TRANSMISSION CHARACTERISTICS OF CABLES A DISTINCTION IS OFTEN MADE BETWEEN SHIELDED CABLES LIKE COAXIAL CABLES AND UNSHIELDED CABLES LIKE TWISTED WIRE PAIRS  #OAXIAL CABLES HAVE MUCH BETTER TRANSMISSION PROPERTIES FOR BROADBAND SIGNALS THAN DO WIRE PAIRS %XCEPT FOR COMPUTER NETWORKS COAXIAL CABLES AND WIRE PAIRS CURRENTLY EXIST IN TWO BASICALLY DIdERENT NETWORK TOPOLOGIES 7IRE PAIRS ARE THE DOMINATING CABLE TYPE IN TELEPHONE ACCESS NETWORKS THAT ARE BUILT FOR POINT TO POINT AND TWO WAY COMMUNICATION #OAXIAL CABLES ARE USUALLY FOUND IN CABLE 46 SYSTEMS A NETWORK TOPOLOGY THAT IS PRIMARILY INTENDED FOR BROADCASTING AND NOT FOR POINT TO POINT COMMUNICATION 4HE CABLE 46 SYSTEMS SOMETIMES CONTAIN AMPLIERS THAT MAKE BIDIRECTIONAL COMMUNICATION ALMOST IMPOSSIBLE (OWEVER CABLE 46 NETWORKS ARE CURRENTLY BEING UPGRADED TO SUPPORT BIDIRECTIONAL COMMUNICATION 7E FOCUS ON WIRE PAIRS AND WILL NOT DISCUSS COAXIAL CABLES FURTHER 4HE COPPER WIRE PAIR DOES NOT CHANGE ITS PHYSICAL BEHAVIOR SIGNICANTLY WITH TIME AND IS THEREFORE CONSIDERED A STATIONARY CHANNEL ;= 4HIS MAKES IT POSSIBLE TO USE A TECHNIQUE CALLED BIT LOADING ;= SEE 3ECTION   WHICH MAKES GOOD USE OF THE SPECTRALLY SHAPED CHANNEL 7HEN BIT LOADING IS USED IN A WIRED /&$- SYSTEM IT IS OFTEN REFERRED TO AS $-4 3INCE /&$- IN COMBINATION WITH BIT LOADING MAKES EbCIENT USE OF AVAILABLE BAND WIDTH IT HAS BECOME A GOOD CANDIDATE FOR DIGITAL SUBSCRIBER LINE $3, SYSTEMS $3, IS ANOTHER NAME FOR DIGITAL HIGH SPEED COMMUNICATION IN THE TELEPHONE ACCESS NETWORK 7HEN THE BIT RATE OdERED IN DOWNSTREAM DIRECTION TO THE SUBSCRIBER IS LARGER THAN THE BIT RATE IN UPSTREAM DIRECTION TO THE BASE  IT IS CALLED AN ASYMMETRIC DIGITAL SUBSCRIBER LINE !$3,  !$3, IS SUITABLE FOR APPLICATIONS LIKE VIDEO ON DEMAND GAMES VIRTUAL SHOPPING INTERNET SURNG ETC WHERE MOST OF THE DATA GOES FROM THE BASE TO THE SUB SCRIBER )N THE 53! THERE EXISTS AN !$3, STANDARD THAT SUPPORTS DOWNSTREAM BIT RATES FROM  TO  -BITSS ;= 4HE BIT RATES OF THE UPSTREAM RETURN PATH USUALLY RANGES



0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING

BETWEEN  AND  KBITSS ;= 3TANDARDS FOR SYMMETRICAL $3,S HAVE ALSO EMERGED TO SUPPORT VIDEO CONFERENCING AND OTHER SERVICES WITH HIGH DATA RATE IN THE UPSTREAM DIRECTION 4HE RST SYMMETRIC $3, SYSTEM WAS CALLED HIGH BIT RATE DIGITAL SUBSCRIBER LINE ($3, ;= WHICH CURRENTLY SUPPORTS BIT RATES BETWEEN  AND  -BITSSEC IN BOTH DIRECTIONS ; = &OR DIGITAL SUBSCRIBER LINES WITH HIGHER BIT RATES THAN ($3, AND !$3, THE TERM VERY HIGH BIT RATE DIGITAL SUBSCRIBER LINE 6$3, IS USED  3UBSCRIBER LINE TRANSFER FUNCTION

4HE CHARACTERISTICS OF THE WIRE PAIR CHANNEL HAVE BEEN STUDIED IN A NUMBER OF PAPERS ;  = ! THOROUGH DESCRIPTION OF THE TRANSFER FUNCTION OF COPPER WIRES AND NOISE SOURCES IS GIVEN BY 7ERNER IN ;= &OR $3,S USING A LARGE FREQUENCY RANGE SEVERAL -(Z OR HIGHER THE ATTENUATION FUNCTION CAN BE APPROXIMATED AS

y J' E C J  Dgt i 



WHERE C IS THE LENGTH OF THE CABLE AND J IS A CABLE CONSTANT 4HIS MODEL IS OFTEN USED WHEN 6$3, AND ($3, SYSTEMS ARE ANALYZED ; =  .OISE AND CROSSTALK

4HE MOST IMPORTANT NOISE SOURCES IN THE SUBSCRIBER LINE ENVIRONMENT ARE CROSSTALK FROM OTHER WIRE PAIRS IN THE SAME CABLE RADIO FREQUENCY 2& NOISE FROM NEARBY RADIO TRANS MITTERS AND IMPULSE NOISE FROM RELAYS SWITCHES ELECTRICAL MACHINES ETC !7'. IS GENERALLY NOT A LIMITING FACTOR IN DIGITAL SUBSCRIBER LINES FOR SHORT CABLES BUT BECOMES MORE IMPORTANT WITH INCREASING CABLE LENGTH )N EG ;= !'7. P IS INCLUDED IN THE CHANNEL MODEL WITH A SPECTRAL DENSITY OF ` D"M(Z x6 (Z  )MPULSE NOISE IS DIbCULT TO CHARACTERIZE COMPLETELY BUT SOME EdORTS HAS BEEN MADE TO MODEL THIS KIND OF DISTURBANCES ; = 4HE NORMAL WAY TO MITIGATE THE EdECTS OF IMPULSE NOISE ON A $-4 SYSTEM IS TO ADD   D" TO THE SYSTEM MARGIN ;= AND TO USE SPECIALLY DESIGNED CODES ;= )T SHOULD BE NOTED THAT $-4 IS MORE RESISTANT TO IMPULSE NOISE THAN SINGLE CARRIER SYSTEMS SUCH AS CARRIERLESS AMPLITUDEPHASE #!0 MODULATION ;= 4HE IMPACT OF 2& NOISE ON A $3, SYSTEM CAN BE REDUCED SIGNICANTLY WITH /&$AND BIT LOADING ;= 2& NOISE CAN BE MODELLED AS NARROWBAND DISTURBANCE WITH KNOWN SPECTRAL DENSITY AND THE BIT ERROR RATE "%2 CAN BE PRESERVED BY TRANSMITTING FEWER SOMETIMES ZERO BITS ON THE DISTURBED SUBCHANNELS 4HERE ARE BASICALLY TWO DIdERENT FORMS OF CROSSTALK NEAR END CROSSTALK .%84 AND FAR END CROSSTALK &%84  .%84 OCCURS AT THE CENTRAL ObCE BASE STATION WHEN THE WEAK UPSTREAM SIGNAL Q S  IS DISTURBED BY STRONG DOWNSTREAM SIGNALS R S  SEE &IGURE  &%84 IS CROSSTALK FROM ONE TRANSMITTED SIGNAL R S  TO ANOTHER Q S  IN THE SAME DIRECTION SEE &IGURE  AND APPEARS AT BOTH ENDS OF THE WIRE LOOP 4HE SPECTRAL DENSITY OF .%84 IS MODELLED IN ;= AS /E E  / E JE E (  



&IGURE  .EAR END CROSSTALK .%84 

&IGURE  &AR END CROSSTALK &%84  AND THE SPECTRAL DENSITY OF &%84 AS

y /7 E C  / E J7 E  J' E C J C  / E J7 E  Dgt i C



WHERE / E IS THE SPECTRAL DENSITY OF THE TRANSMITTED SIGNALS JE AND J7 ARE CONSTANTS DEPENDING ON THE TYPE OF CABLE HOW WELL BALANCED THE CABLES ARE AND THE NUMBER OF DISTURBING COPPER PAIRS ;= .OTE THAT .%84 DOES NOT DEPEND ON THE LENGTH OF THE WIRE PAIR )N &IGURE  WE DISPLAY AN EXAMPLE OF THE SPECTRAL DENSITY OF A RECEIVED SIGNAL .%84 AND &%84


/2 J '


E  CJ

//%

E E C

NEXT FEXT

=  " D N I Z  (  7 ; Y T  I S N E D  L A R T C E P S  R E W O 0 




&REQUENCY ;-(Z=



&IGURE  0OWER SPECTRAL DENSITY OF ATTENUATED SIGNAL .%84 AND &%84



0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING



3YNCHRONIZATION

/NE OF THE ARGUMENTS AGAINST /&$- IS THAT IT IS HIGHLY SENSITIVE TO SYNCHRONIZATION ERRORS IN PARTICULAR TO FREQUENCY ERRORS ;= (ERE WE GIVE AN OVERVIEW OF THREE SYNCHRO NIZATION PROBLEMS SYMBOL CARRIER FREQUENCY AND SAMPLING FREQUENCY SYNCHRONIZATION !LSO THE EdECTS OF PHASE OdSETS AND PHASE NOISE ARE DISCUSSED




3YMBOL SYNCHRONIZATION
4IMING ERRORS

! GREAT DEAL OF ATTENTION IS GIVEN TO SYMBOL SYNCHRONIZATION IN /&$- SYSTEMS (OWEVER BY USING A CYCLIC PREX THE TIMING REQUIREMENTS ARE RELAXED SOMEWHAT 4HE OBJECTIVE IS TO KNOW WHEN THE SYMBOL STARTS 4HE IMPACT OF TIMING ERRORS HAS BEEN ANALYZED IN ; = ! TIMING OdSET GIVES RISE TO A PHASE ROTATION OF THE SUBCARRIERS 4HIS PHASE ROTATION IS LARGEST ON THE EDGES OF THE FREQUENCY BAND )F A TIMING ERROR IS SMALL ENOUGH TO KEEP THE CHANNEL IMPULSE RESPONSE WITHIN THE CYCLIC PREX THE ORTHOGONALITY IS MAINTAINED )N THIS CASE A SYMBOL TIMING DELAY CAN BE VIEWED AS A PHASE SHIFT INTRODUCED BY THE CHANNEL AND THE PHASE ROTATIONS CAN BE ESTIMATED BY A CHANNEL ESTIMATOR )F A TIME SHIFT IS LARGER THAN THE CYCLIC PREX )3) WILL OCCUR 4HERE ARE TWO MAIN METHODS FOR TIMING SYNCHRONIZATION BASED ON PILOTS OR ON THE CYCLIC PREX !N ALGORITHM OF THE FORMER KIND WAS SUGGESTED BY 7ARNER AND ,EUNG IN ;= 4HEY USE A SCHEME WHERE THE /&$- SIGNAL IS TRANSMITTED BY FREQUENCY MODULATION &-  4HE TRANSMITTER ENCODES A NUMBER OF RESERVED SUBCHANNELS WITH KNOWN PHASES AND AMPLITUDES 4HE SYNCHRONIZATION TECHNIQUE WITH MODICATIONS IS APPLICABLE TO /&$SIGNALS TRANSMITTED BY AMPLITUDE MODULATION 4HEIR ALGORITHM CONSISTS OF  PHASES POWER DETECTION COARSE SYNCHRONIZATION AND NE SYNCHRONIZATION 4HE RST PHASE POWER DETECTION DETECTS WHETHER OR NOT AN /&$- SIGNAL IS PRESENT BY MEASURING THE RECEIVED POWER AND COMPARE IT TO A THRESHOLD 4HE SECOND PHASE COARSE SYNCHRONIZATION IS USED TO ACQUIRE SYNCHRONIZATION ALIGNMENT TO WITHIN f SAMPLES 4HIS PERFORMANCE IS NOT ACCEPTABLE BUT THIS PHASE SERVES TO SIMPLIFY THE TRACKING ALGO RITHM WHICH CAN ASSUME THAT THE TIMING ERROR IS SMALL  4HE COARSE SYNCHRONIZATION IS DONE BY CORRELATING THE RECEIVED SIGNAL TO A COPY OF THE TRANSMITTED SYNCHRONIZATION SIGNAL 4O ND THE PEAK OF THIS CORRELATION WITH ENOUGH ACCURACY A DIGITAL LTER IS USED TO PROVIDE INTERPOLATED DATA VALUES AT FOUR TIMES THE ORIGINAL DATA RATE )N THE LAST PHASE OF THE SYNCHRONIZATION NE SYNCHRONIZATION  THE SUBCHANNELS WITH PILOTS ARE EQUALIZED WITH THE ESTIMATED CHANNEL OBTAINED FROM PILOTS 3INCE THE COARSE SYNCHRONIZATION GUAR ANTEES THAT THE TIMING ERROR IS LESS THAN f THE CHANNEL IMPULSE RESPONSE IS WITHIN THE CYCLIC PREX 4HE REMAINING PHASE ERRORS ON THE PILOT SUBCHANNELS ARE DUE TO TIMING ERROR AND CAN BE ESTIMATED BY LINEAR REGRESSION 4HERE ARE ALSO SYNCHRONIZATION ALGORITHMS BASED ON THE CYCLIC PREX )N ;= THE DIdERENCE BETWEEN RECEIVED SAMPLES SPACED - SAMPLES APART IS FORMED Q J ` Q J
-  7HEN ONE OF THE SAMPLES BELONGS TO THE CYCLIC PREX AND THE OTHER ONE TO THE /&$SYMBOL FROM WHICH IT IS COPIED THE DIdERENCE SHOULD BE SMALL /THERWISE THE DIdERENCE BETWEEN TWO UNCORRELATED RANDOM VARIABLES WILL HAVE TWICE THE POWER AND HENCE ON

 AVERAGE WILL BE LARGER "Y WINDOWING THIS DIdERENCE WITH A RECTANGULAR WINDOW OF THE SAME LENGTH AS THE CYCLIC PREX THE OUTPUT SIGNAL HAS A MINIMUM WHEN A NEW /&$SYMBOL STARTS 4HIS IDEA IS MORE FORMALLY ELABORATED IN ;  = 4HE LIKELIHOOD FUNCTION GIVEN THE OBSERVED SIGNAL Q J WITH A TIMING AND FREQUENCY ERROR IS DERIVED IN ; = 4HIS FUNCTION IS MAXIMIZED TO SIMULTANEOUSLY OBTAIN ESTIMATES OF BOTH TIMING AND FREQUENCY OdSETS 7ITH NO FREQUENCY OdSET THE LIKELIHOOD FUNCTION WITH RESPECT TO A TIMING OdSET t IS c t 
t C  8

 2-1 1D FQ J Q J
- G ` JQ J ` Q J
- J 2-1
 2-1
 t (t



&OR MEDIUM AND HIGH 3.2S 2-1   A MAXIMUM LIKELIHOOD -, ESTIMATOR BASED ON c t ESSENTIALLY APPLIES A MOVING AVERAGE TO THE TERM JQ J ` Q J
- J  IE THE SAME AS THE ESTIMATOR IN ;= (OWEVER FOR SMALL 3.2 VALUES THE CROSSCORRELATION Q J Q J
ALSO HAS TO BE TAKEN INTO ACCOUNT ! SIMILAR PROCEDURE IS USED IN ;= WITH THE DIdERENCE THAT THE INPHASE AND QUADRATURE PARTS OF THE OBSERVED SIGNAL Q J ARE QUANTIZED TO  BIT BEFORE t IS ESTIMATED 4HIS YIELDS A SYMBOL SYNCHRONIZER WITH A LOW COMPLEXITY THAT CAN BE USED IN AN ACQUISITION MODE 3YNCHRONIZATION IN THE UPLINK IS MORE DIbCULT THAN IN THE DOWNLINK OR IN BROADCASTING 4HIS IS DUE TO THE FACT THAT THERE WILL BE A SEPARATE OdSET FOR EACH USER 4HIS PROBLEM HAS NOT YET BEEN GIVEN MUCH ATTENTION IN THE LITERATURE (OWEVER A RANDOM ACCESS SEQUENCE IS USED TO SYNCHRONIZE THE MOBILE AND THE BASE STATION IN ;= )NTERFERENCE DUE TO NON SYNCHRONIZED TRANSMISSION HAS BEEN INVESTIGATED IN ;=  #ARRIER PHASE NOISE

#ARRIER PHASE NOISE IS CAUSED BY IMPERFECTIONS IN THE TRANSMITTER AND RECEIVER OSCILLATORS &OR A FREQUENCY SELECTIVE CHANNEL NO DISTINCTION CAN BE MADE BETWEEN THE PHASE ROTATION INTRODUCED BY A TIMING ERROR AND A CARRIER PHASE OdSET ;= !N ANALYSIS OF THE IMPACT OF CARRIER PHASE NOISEh IS DONE IN ;= 4HERE i IT IS MODELLED AS A 7IENER PROCESS t S WITH $ Ft S G   AND $ t S
S ` t S   {n JSJ WHERE n IN (Z DENOTES THE ONE SIDED  D" LINEWIDTH OF THE ,ORENTZIAN POWER DENSITY SPECTRUM OF THE FREE RUNNING CARRIER GENERATOR 4HE DEGRADATION IN 3.2 IE THE INCREASE IN 3.2 NEEDED TO COMPENSATE FOR THE ERROR CAN BE APPROXIMATED BY t u n $  # D" { {  KM  6 - 

WHERE 6 IS THE BANDWIDTH AND $ - IS THE PER SYMBOL 3.2 .OTE THAT THE DEGRADATION INCREASES WITH THE NUMBER OF SUBCARRIERS $UE TO THE RAPID VARIATIONS OF THE PHASE NOISE IT MAY CAUSE LARGE PROBLEMS !NALYSIS OF THE IMPACT OF PHASE NOISE IN CODED SYSTEMS HAS BEEN DONE IN ;=



0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING



3AMPLING FREQUENCY SYNCHRONIZATION

4HE RECEIVED CONTINUOUS TIME SIGNAL IS SAMPLED AT INSTANTS DETERMINED BY THE RECEIVER CLOCK 4HERE ARE TWO TYPES OF METHODS OF DEALING WITH THE MISMATCH IN SAMPLING FREQUENCY )N SYNCHRONIZED SAMPLING SYSTEMS A TIMING ALGORITHM CONTROLS A VOLTAGE CONTROLLED CRYSTAL OSCILLATOR IN ORDER TO ALIGN THE RECEIVER CLOCK WITH THE TRANSMITTER CLOCK 4HE OTHER METHOD IS NON SYNCHRONIZED SAMPLING WHERE THE SAMPLING RATE REMAINS XED WHICH REQUIRES POST PROCESSING IN THE DIGITAL DOMAIN 4HE EdECT OF A CLOCK FREQUENCY OdSET IS TWOFOLD THE USEFUL SIGNAL COMPONENT IS ROTATED AND ATTENUATED AND IN ADDITION )#) IS INTRODUCED )N ;= THE BIT ERROR RATE PERFORMANCE OF A NON SYNCHRONIZED SAMPLED /&$SYSTEM HAS BEEN INVESTIGATED )T IS SHOWN THAT NON SYNCHRONIZED SAMPLING SYSTEMS ARE MUCH MORE SENSITIVE TO A FREQUENCY OdSET COMPARED WITH A SYNCHRONIZED SAMPLING SYS TEM &OR NON SYNCHRONIZED SAMPLING SYSTEMS IT WAS SHOWN THAT THE DEGRADATION IN D" DUE TO A FREQUENCY SAMPLING OdSET DEPENDS ON THE SQUARE OF THE CARRIER INDEX AND ON THE SQUARE OF THE RELATIVE FREQUENCY OdSET %RRORS IN THE SAMPLING FREQUENCY FOR $-4 SYSTEMS HAVE BEEN ANALYZED IN ;=




#ARRIER FREQUENCY SYNCHRONIZATION

&REQUENCY ERRORS

&REQUENCY OdSETS ARE CREATED BY DIdERENCES IN OSCILLATORS IN TRANSMITTER AND RECEIVER $OPPLER SHIFTS OR PHASE NOISE INTRODUCED BY NON LINEAR CHANNELS 4HERE ARE TWO DESTRUC TIVE EdECTS CAUSED BY A CARRIER FREQUENCY OdSET IN /&$- SYSTEMS /NE IS THE REDUCTION OF SIGNAL AMPLITUDE THE SINC FUNCTIONS ARE SHIFTED AND NO LONGER SAMPLED AT THE PEAK AND THE OTHER IS THE INTRODUCTION OF )#) FROM THE OTHER CARRIERS SEE &IGURE  4HE

&IGURE  %dECTS OF A FREQUENCY OdSET a%  REDUCTION IN SIGNAL AMPLITUDE p AND INTERCARRIER INTERFERENCE q  LATTER IS CAUSED BY THE LOSS OF ORTHOGONALITY BETWEEN THE SUBCHANNELS )N ;= 0OLLET ET AL ANALYTICALLY EVALUATE THE DEGRADATION OF THE "%2 CAUSED BY THE PRESENCE OF CARRIER

 FREQUENCY OdSET AND CARRIER PHASE NOISE FOR AN !7'. CHANNEL )T IS FOUND THAT A MUL TICARRIER SYSTEM IS MUCH MORE SENSITIVE THAN A SINGLE CARRIER SYSTEM $ENOTE THE RELATIVE 7 FREQUENCY OdSET NORMALIZED BY THE SUBCARRIER SPACING BY aE  T(E  WHERE a% IS THE FREQUENCY OdSET AND - THE NUMBER OF SUBCARRIERS 4HE DEGRADATION # IN 3.2 IN D" CAN THEN BE APPROXIMATED BY  # D" { {aE  KM 


t u  - a a% $ $   { -  KM  6 -



.OTE THAT THE DEGRADATION IN D" INCREASES WITH THE SQUARE OF THE NUMBER OF SUBCARRIERS IF a% AND 6 ARE XED )N ;= -OOSE DERIVES THE SIGNAL TO INTERFERENCE RATIO 2(1 ON A FADING AND DISPER SIVE CHANNEL 4HE 2(1 IS DENED AS THE RATIO OF THE POWER OF THE USEFUL SIGNAL TO THE POWER OF THE INTERFERENCE SIGNAL )#) AND h ADDITIVE i NOISE  (E ASSUMED THAT ALL CHANNEL ATTENUATIONS Gt HAVE THE SAME POWER $ JGt J  !N UPPER BOUND ON THE DEGRADATION IS # D" v  KNF
6 RHM { aE 
 E 

RHMB aE



WHERE RHMB W  RHM {W  {W  4HE FACTOR  IS FOUND FROM A LOWER BOUND OF THE SUMMATION OF ALL INTERFERING SUBCARRIERS )N &IGURE  THE DEGRADATION IS PLOTTED AS A FUNCTION OF THE NORMALIZED FREQUENCY OdSET aE  IE RELATIVE TO THE SUBCARRIER SPAC ING 4HE SYNCHRONIZATION REQUIREMENTS FOR AN /&$- SYSTEM HAVE BEEN INVESTIGATED IN

&IGURE  $EGRADATION IN 3.2 DUE TO A FREQUENCY OdSET NORMALIZED TO THE SUBCARRIER SPACING  !NALYTICAL EXPRESSION FOR !7'. DASHED AND FADING CHANNELS SOLID  ;= 4HE CONCLUSION THEREIN IS THAT IN ORDER TO AVOID SEVERE DEGRADATION THE FREQUENCY SYNCHRONIZATION ACCURACY SHOULD BE BETTER THAN  

 

0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING &REQUENCY ESTIMATORS

3EVERAL CARRIER SYNCHRONIZATION SCHEMES HAVE BEEN SUGGESTED IN THE LITERATURE !S WITH SYMBOL SYNCHRONIZATION THEY CAN BE DIVIDED INTO TWO CATEGORIES BASED ON PILOTS OR ON THE CYCLIC PREX "ELOW FOLLOWS A SHORT OVERVIEW OF SOME OF THEM 0ILOT AIDED ALGORITHMS HAVE BEEN ADDRESSED IN ;= )N THAT WORK SOME SUBCARRIERS ARE USED FOR THE TRANSMISSION OF PILOTS USUALLY A PSEUDO NOISE 0. SEQUENCE  5SING THESE KNOWN SYMBOLS THE PHASE ROTATIONS CAUSED BY THE FREQUENCY OdSET CAN BE ESTIMATED 5NDER THE ASSUMPTION THAT THE FREQUENCY OdSET IS LESS THAN HALF THE SUBCARRIER SPACING THERE IS A ONE TO ONE CORRESPONDENCE BETWEEN THE PHASE ROTATIONS AND THE FREQUENCY OdSET 4O ASSURE THIS AN ACQUISITION ALGORITHM MUST BE APPLIED )N ;= SUCH AN ALGORITHM IS B   CONSTRUCTED BY FORMING A FUNCTION WHICH IS SINC SHAPED AND HAS A PEAK FOR E ` E )T WAS FOUND THAT BY EVALUATING THIS FUNCTION IN POINTS 3 APART AN ACQUISITION COULD BE OBTAINED BY MAXIMIZING THAT FUNCTION 4HIS ACQUISITION SCHEME WAS CONRMED BY COMPUTER SIMULATIONS TO WORK WELL BOTH FOR AN !7'. CHANNEL AND A FADING CHANNEL ! RELATED TECHNIQUE IS TO USE THE CYCLIC PREX WHICH TO SOME EXTENT CAN BE VIEWED AS PILOTS 4HE REDUNDANCY OF THE CYCLIC PREX CAN BE USED IN SEVERAL WAYS EG BY CREATING A FUNCTION THAT PEAKS AT ZERO OdSET AND NDING ITS MAXIMIZING VALUE ; = OR BY DOING MAXIMUM LIKELIHOOD ESTIMATION ;   = )N ;= IT IS ASSUMED THAT THE CYCLIC PREX HAS THE SAME SIZE AS THE /&$- SYMBOL IE THE USEFUL SYMBOL IS TRANSMITTED TWICE  IN ;= AVERAGING IS PERFORMED TO REMOVE THE DATA DEPENDENCE AND IN ;= DECISION DIRECTION IS USED )N ;= THE LIKELIHOOD FUNCTION FOR BOTH TIMING AND FREQUENCY OdSETS IS DERIVED BY ASSUMING A NON DISPERSIVE CHANNEL AND BY CONSIDERING THE TRANSMITTED DATA SYMBOLS Wt UNCORRELATED "Y MAXIMIZING THIS FUNCTION A SIMULTANEOUS ESTIMATION OF THE TIMING AND FREQUENCY OdSETS CAN BE OBTAINED )F THE FREQUENCY ERROR IS SLOWLY VARYING COMPARED THE /&$- SYMBOL RATE A PHASE LOCKED LOOP 0,, ;= CAN BE USED TO REDUCE THE ERROR FURTHER )T IS INTERESTING TO NOTE THE RELATIONSHIP BETWEEN TIME AND FREQUENCY SYNCHRONIZATION )F THE FREQUENCY SYNCHRONIZATION IS A PROBLEM IT CAN BE REDUCED BY LOWERING THE NUMBER OF SUBCARRIERS WHICH WILL INCREASE THE SUBCARRIER SPACING 4HIS WILL HOWEVER INCREASE THE DEMANDS ON THE TIME SYNCHRONIZATION SINCE THE SYMBOL LENGTH GETS SHORTER IE A LARGER RELATIVE TIMING ERROR WILL OCCUR 4HUS THE SYNCHRONIZATIONS IN TIME AND FREQUENCY ARE CLOSELY RELATED TO EACH OTHER



#HANNEL ESTIMATION

-ODULATION CAN BE CLASSIED AS DIdERENTIAL OR COHERENT 7HEN USING DIdERENTIAL MODU LATION THERE IS NO NEED FOR A CHANNEL ESTIMATE SINCE THE INFORMATION IS ENCODED IN THE DIdERENCE BETWEEN TWO CONSECUTIVE SYMBOLS 4HIS IS A COMMON TECHNIQUE IN WIRELESS SYS TEMS WHICH SINCE NO CHANNEL ESTIMATOR IS NEEDED REDUCES THE COMPLEXITY OF THE RECEIVER $IdERENTIAL MODULATION IS USED IN THE %UROPEAN $!" STANDARD ;= 4HE DRAWBACKS ARE ABOUT A  D" NOISE ENHANCEMENT ;= AND AN INABILITY TO USE EbCIENT MULTIAMPLITUDE CONSTELLATIONS (OWEVER DIdERENTIAL SCHEMES CAN BENET FROM ASSISTANCE BY A CHANNEL ES TIMATOR ;= !N INTERESTING ALTERNATIVE TO COHERENT MODULATION IS DIdERENTIAL AMPLITUDE

 AND PHASE SHIFT KEYING $!03+ ;   = WHERE A SPECTRAL EbCIENCY GREATER THAN THAT OF $03+ IS ACHIEVED BY USING A DIdERENTIAL CODING OF AMPLITUDE AS WELL 4HIS REQUIRES A NONUNIFORM AMPLITUDE DISTRIBUTION #OHERENT MODULATION HOWEVER ALLOWS ARBITRARY SIGNAL CONSTELLATIONS AND IS AN OBVIOUS CHOICE IN WIRED SYSTEMS WHERE THE CHAN NEL HARDLY CHANGES WITH TIME )N WIRELESS SYSTEMS THE EbCIENCY OF COHERENT MODULATION MAKES IT AN INTERESTING CHOICE WHEN THE BIT RATE IS HIGH AS IN DIGITAL VIDEO BROADCAST $6" ; = 4HE CHANNEL ESTIMATION IN WIRED SYSTEMS IS FAIRLY STRAIGHTFORWARD AND IS NOT DISCUSSED IN DETAIL BELOW 7E CONCENTRATE ON CHANNEL ESTIMATION IN WIRELESS SYSTEMS WHERE THE COMPLEXITY OF THE ESTIMATOR IS AN IMPORTANT DESIGN CRITERION 4HERE ARE MAINLY TWO PROBLEMS IN THE DESIGN OF CHANNEL ESTIMATORS FOR WIRELESS /&$SYSTEMS 4HE RST PROBLEM CONCERNS THE CHOICE OF HOW PILOT INFORMATION DATASIGNALS KNOWN AT THE RECEIVER SHOULD BE TRANSMITTED 4HIS PILOT INFORMATION IS NEEDED AS A REFERENCE FOR CHANNEL ESTIMATION 4HE SECOND PROBLEM IS THE DESIGN OF AN ESTIMATOR WITH BOTH LOW COMPLEXITY AND GOOD CHANNEL TRACKING ABILITY 4HESE TWO PROBLEMS ARE INTERCONNECTED SINCE THE PERFORMANCE OF THE ESTIMATOR DEPENDS ON HOW PILOT INFORMATION IS TRANSMITTED



0ILOT INFORMATION

#HANNEL ESTIMATORS USUALLY NEED SOME KIND OF PILOT INFORMATION AS A POINT OF REFERENCE ! FADING CHANNEL REQUIRES CONSTANT TRACKING SO PILOT INFORMATION HAS TO BE TRANSMITTED MORE OR LESS CONTINUOUSLY $ECISION DIRECTED CHANNEL ESTIMATION CAN ALSO BE USED ;= BUT EVEN IN THESE TYPES OF SCHEMES PILOT INFORMATION HAS TO BE TRANSMITTED REGULARLY TO MITIGATE ERROR PROPAGATION 4O THE AUTHORS KNOWLEDGE THERE IS VERY LITTLE PUBLISHED ON HOW TO TRANSMIT PILOT INFORMATION IN WIRELESS /&$- (OWEVER AN EbCIENT WAY OF ALLOWING A CONTINUOUSLY UPDATED CHANNEL ESTIMATE IS TO TRANSMIT PILOT SYMBOLS INSTEAD OF DATA AT CERTAIN LOCATIONS OF THE /&$- TIME FREQUENCY LATTICE 4HIS CAN BE VIEWED AS A GENERALIZATION OF PILOT SYMBOL ASSISTED MODULATION 03!- IN THE SINGLE CARRIER CASE 03!- IN THE SINGLE CARRIER CASE WAS INTRODUCED IN ;= AND THOROUGHLY ANALYZED IN ;= !N EXAMPLE OF THIS IS SHOWN IN &IGURE  WHERE BOTH SCATTERED AND CONTINUAL PILOT SYMBOLS ARE SHOWN )N A PRELIMINARY DRAFT OF THE %UROPEAN $6" STANDARD ;= PILOT INFORMATION IS SPECIED TO BE TRANSMITTED ON BOOSTED SUBCARRIERS BOTH SCATTERED AND AS CONTINUAL PILOT CARRIERS "OOSTED SUBCARRIERS MEANS THAT PILOT INFORMATION IS TRANSMITTED AT HIGHER POWER THAN THE DATA )N GENERAL THE FADING CHANNEL CAN BE VIEWED AS A  $ SIGNAL TIME AND FREQUENCY  WHICH IS SAMPLED AT PILOT POSITIONS AND THE CHANNEL ATTENUATIONS BETWEEN PILOTS ARE ESTIMATED BY INTERPOLATION 4HIS ENABLES US TO USE THE  $ SAMPLING THEOREM TO PUT LIMITS THE DENSITY OF THE PILOT PATTERN ;= (OWEVER AS IN THE SINGLE CARRIER CASE ;= THE PILOT PATTERN SHOULD BE DESIGNED SO THAT THE CHANNEL IS OVERSAMPLED AT THE RECEIVER



0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING

&IGURE  !N EXAMPLE OF PILOT INFORMATION TRANSMITTED BOTH SCATTERED AND CONTINUAL ON CERTAIN SUBCARRIERS



%STIMATOR DESIGN

!SSUMING THAT THE PILOT PATTERN IS CHOSEN THE OPTIMAL LINEAR CHANNEL ESTIMATOR IN TERMS OF MEAN SQUARED ERROR -3% IS A  $ 7IENER LTER +NOWING THE STATISTICAL PROPERTIES OF THE CHANNEL SUCH AN ESTIMATOR CAN BE DESIGNED USING STANDARD TECHNIQUES ;= 4HE COMBINATION OF HIGH DATA RATES AND LOW BIT ERROR RATES NECESSITATES THE USE OF ESTIMATORS THAT HAVE BOTH LOW COMPLEXITY AND HIGH ACCURACY 4HESE TWO CONSTRAINTS ON THE ESTI MATORS WORK AGAINST EACH OTHER -OST ESTIMATORS WITH HIGH ACCURACY SUCH AS THE  $ 7IENER LTER HAVE A LARGE COMPUTATIONAL COMPLEXITY WHILE ESTIMATORS OF LOWER COMPLEX ITY USUALLY PRODUCE A LESS ACCURATE ESTIMATE 4HE ART IN DESIGNING CHANNEL ESTIMATORS IS NDING A GOOD TRADE Od BETWEEN COMPLEXITY AND PERFORMANCE 4HE ISSUE OF REDUCING THE COMPUTATIONAL COMPLEXITY WHILE MAINTAINING MOST OF THE PERFORMANCE HAS BEEN ADDRESSED IN SEVERAL PUBLICATIONS )N ;= SEPARABLE LTERS ARE APPLIED INSTEAD OF A  $ NITE IMPULSE RESPONSE &)2 LTER 4HE USE OF SEPARABLE LTERS INSTEAD OF FULL  $ LTERS IS A STANDARD TECHNIQUE USED TO REDUCE COMPUTATIONAL COMPLEXITY IN MULTIDIMENSIONAL SIGNAL PROCESSING ;= 5SING THIS TECHNIQUE THE ESTIMATION IS RST PERFORMED IN THE FREQUENCY DIRECTION USING A  $ &)2 LTER AND THEN IN THE TIME DIRECTION USING A SECOND  $ &)2 LTER 4HIS RESTRICTS THE OBTAINABLE  $ IMPULSE RESPONSES TO THOSE THAT ARE THE OUTER PRODUCT OF TWO  $ LTERS 4HIS RESULTS IN A SMALL PERFORMANCE LOSS BUT THE GREATLY REDUCED COMPLEXITY USUALLY MOTIVATES THE USE OF SEPARABLE LTERS ; = ! SECOND APPROACH IN THE REDUCTION OF COMPUTATIONAL COMPLEXITY IS BASED ON USING TRANSFORMS THAT CONCENTRATE THE CHANNEL POWER TO A FEW TRANSFORM COEbCIENTS THUS AL

 LOWING EbCIENT CHANNEL ESTIMATION TO BE PERFORMED WITH LITTLE EdORT IN THE TRANSFORM DOMAIN ,OW COMPLEXITY ESTIMATORS OF THIS TYPE BASED ON BOTH THE $&4 ;  = AND ON OPTIMAL RANK REDUCTION ; = HAVE BEEN PROPOSED 4HIS TECHNIQUE USUALLY YIELD ES TIMATORS OF HIGH PERFORMANCE AND LOW COMPLEXITY BUT MAY RESULT IN AN IRREDUCIBLE ERROR OOR UNLESS SPECIAL CARE IS TAKEN 4HESE EdECTS ARE ANALYZED IN DETAIL IN ; = ! COMPARATIVE ANALYSIS OF PILOT BASED ESTIMATORS PRESENTED IN ;= SHOWS THAT THE COMBINATION OF SEPARABLE LTERS AND LOW RANK APPROXIMATIONS CAN GIVE HIGH PERFORMANCE ESTIMATORS OF LOW COMPLEXITY WHERE THE GREATEST PORTION OF THE REDUCED COMPLEXITY STEMS FROM THE USE OF SEPARABLE LTERS



0ERFORMANCE EXAMPLE

&IGURE  SHOWS AN EXAMPLE ON THE DIdERENCE IN CODED BIT ERROR RATE BETWEEN COHERENT AND DIdERENTIAL MODULATION 4HE SIMULATIONS ARE PERFORMED FOR A WIRELESS  SUBCARRIER /&$- SYSTEM WITH A  SAMPLE CYCLIC PREX 4HE CHANNEL IS 2AYLEIGH FADING WITH  RELATIVE $OPPLER FREQUENCY AND THE SYSTEM USES TRELLIS CODED MODULATION WITH  PHASE SHIFT KEYING  03+ ACCORDING TO ;= 4HE DIdERENTIAL MODULATION IS PERFORMED IN THE FREQUENCY DIRECTION SINCE THE FREQUENCY CORRELATION IS GREATER THAN THE TIME CORRELATION ;= 4HE PERFORMANCE OF COHERENT MODULATION IS PRESENTED FOR BOTH KNOWN CHANNEL AND WITH A LOW COMPLEXITY ESTIMATOR ;= 4HE LOW COMPLEXITY CHANNEL ESTIMATOR USES A  PILOT SYMBOL DENSITY AND REQUIRES ONLY  MULTIPLICATIONS PER ESTIMATED CHANNEL ATTENUATION

&IGURE  !N EXAMPLE ON THE DIdERENCE BETWEEN COHERENT AND DIdERENTIAL  03+ IN A 2AYLEIGH FADING ENVIRONMENT 4HIS GURE ILLUSTRATES THAT COHERENT MODULATION WITH LOW COMPLEXITY CHANNEL ESTIMA TION CAN OUTPERFORM DIdERENTIAL MODULATION %VEN THOUGH THE CHANNEL CORRELATION IN THE FREQUENCY DIRECTION IS LARGE IN THIS CASE THE BEGINNING OF AN ERROR OOR FOR THE DIdERENTIAL MODULATION IS CLEARLY VISIBLE FOR $e - GREATER THAN ABOUT  D" 4HIS ERROR OOR IS OF



0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING

THE SAME TYPE AS THE ONE EXPERIENCED IN THE SINGLE CARRIER CASE WHEN THE CHANNEL IS FADING ;=



#HANNEL CODING

4HIS CHAPTER DESCRIBES CODING IN /&$- SYSTEMS 4HE CODING PROBLEM IS QUITE DIdERENT FOR THE WIRELESS AND THE WIRED CASE )N THE LATTER CASE THE CHANNEL IS STATIC AND TECHNIQUES LIKE BIT LOADING AND MULTIDIMENSIONAL CODING ARE APPROPRIATE /N A FADING CHANNEL THE MAIN DIdERENCE BETWEEN AN /&$- SYSTEM AND A SINGLE CARRIER SYSTEM IS THE INTERLEAVING 7E WILL GIVE A SHORT OVERVIEW OF CODING ON BOTH WIRELESS AND THE WIRED CHANNELS



7IRELESS SYSTEMS

5SING A TIME DOMAIN EQUALIZER IT IS POSSIBLE TO OBTAIN AN , BRANCH DIVERSITY IF THE CHANNEL CONSISTS OF , RESOLVABLE PATHS ;= )N /&$- THE EQUALIZER DOES NOT GIVE YOU ANY DIVERSITY SINCE ALL SUBCHANNELS ARE NARROWBAND AND EXPERIENCE AT FADING ;= (OWEVER THE STRUCTURE OF /&$- OdERS THE OPPORTUNITY TO CODE ACROSS THE SUBCARRIERS )N ;= IT IS SHOWN THAT WITH AN , PATH CHANNEL IT IS POSSIBLE TO OBTAIN AN , BRANCH DIVERSITY THROUGH CODING (ENCE IN THIS DIVERSITY CONTEXT A MULTI CARRIER SYSTEM IS COMPARABLE TO A SINGLE CARRIER SYSTEM "ESIDES COMBATING FADING CODING HAS ALSO BEEN PROPOSED TO DEAL WITH LONG ECHOES THAT CAUSES )3) BETWEEN SUBSEQUENT /&$- SYMBOLS ;= 4HE DESIGN OF CODES FOR /&$- SYSTEMS ON FADING CHANNELS FOLLOWS MANY OF THE STAN DARD TECHNIQUES 7HAT IS SPECIAL ABOUT /&$- IS THE TIME FREQUENCY LATTICE AND THE POSSIBILITY TO USE TWO DIMENSIONS FOR INTERLEAVING AND CODING 4O ILLUSTRATE HOW TO USE THIS STRUCTURE WE GIVE AN OVERVIEW OF TWO SYSTEMS THE %UROPEAN $!" STANDARD AND A TRELLIS CODED SYSTEM BY (HER 4HE $!" SYSTEM USES DIdERENTIAL MODULATION WHICH AVOIDS CHANNEL ESTIMATION WHILE THE OTHER SYSTEM USES A MULTIAMPLITUDE SIGNAL CONSTEL LATION WHICH REQUIRES CHANNEL ESTIMATION  $IGITAL !UDIO "ROADCASTING

$IGITAL BROADCASTING TO MOBILE RECEIVERS WAS UNDER CONSIDERABLE INVESTIGATION IN THE LATE S ; = 4HE STANDARD FOR $!" WAS SET IN %UROPE TO USE /&$- ;= WHILE IT IS STILL UNDER INVESTIGATION IN THE 53! ;= 4HE %UROPEAN $!" SYSTEM USES DIdERENTIAL QUADRATURE PHASE SHIFT KEYING $103+ TO AVOID CHANNEL ESTIMATION !N ARGUMENT FOR THIS IS BASED ON SIMPLE AND INEXPENSIVE RECEIVERS FOR CONSUMERS 4HE CHANNEL ENCODING PROCESS IS BASED ON PUNCTURED CONVOLUTIONAL CODING WHICH ALLOWS BOTH EQUAL AND UNEQUAL ERROR PROTECTION ;= !S A MOTHER CODE A RATE  CONVOLUTIONAL CODE WITH CONSTRAINT LENGTH  AND OCTAL POLYNOMIALS  IS USED 4HE PUNCTURING PROCEDURE ALLOWS THE EdECTIVE CODE RATE TO VARY BETWEEN  AND  )NTERLEAVING IS PERFORMED IN BOTH TIME AND FREQUENCY 4HE FORMER IS A KIND OF BLOCK IN TERLEAVING AFTER WHICH THE BITS ARE MAPPED TO 103+ SYMBOLS 4HE FREQUENCY INTERLEAVER WORKING WITH THE 103+ SYMBOLS FOLLOWS A PERMUTATION RULE OF THE  SUBCARRIERS )N

 ONE OF THE THREE TRANSMISSIONS MODES THIS PERMUTATION RULE IS e  e M    e M ` 

 LNC   M         

4HIS PERMUTATION DENES THE SET Fe   e   e       e  G  F       G 

ACCORDING TO WHICH THE INTERLEAVING PATTERN IS CHOSEN !FTER THE FREQUENCY INTERLEAVING THE 103+ SYMBOLS ARE DIdERENTIALLY MODULATED ON EACH SUBCARRIER  4RELLIS CODED /&$-

!S AN EXAMPLE OF A MULTIAMPLITUDE COHERENT AND CODED /&$- SYSTEM WE GIVE A SHORT OVERVIEW OF A SYSTEM INVESTIGATED BY (HER 4HE ORIGINAL INVESTIGATION ;= ADDRESSES A DIGITAL AUDIO BROADCASTING SCENARIO BUT THE CONCEPT IS MORE GENERAL (HERS INVESTI GATION IS THEREFORE ONE OF THE MOST REFERENCED PAPERS ON /&$- (IS SYSTEM IS A POWER AND BANDWIDTH EbCIENT CONCATENATED CODING SYSTEM FOR DATA TRANSMISSION ON TIME AND FREQUENCY SELECTIVE MOBILE FADING CHANNELS ! CONCATENATED CODING IS USED IN CONJUNCTION WITH DOUBLE INTERLEAVING AND SLOW FREQUENCY HOPPING TO PROVIDE DIVERSITY !N OVERVIEW OF THE SYSTEM IS DEPICTED IN &IGURE  4HE OUTER CODES ARE RATE COMPATIBLE PUNCTURED

&IGURE  /VERVIEW OF THE SYSTEM INVESTIGATED BY (HER ;= CODES 2#0# DERIVED FROM THE RATE  CODE   WITH CONSTRAINT LENGTH  4HE OUTER INTERLEAVING SCHEME IS APPLIED TO BREAK UP ERROR BURSTS FROM THE INNER CODING SYS TEM 3INCE THESE BURSTS ARE TYPICALLY MUCH SHORTER THAN THE FADING BURSTS THE OUTER INTERLEAVING CAN BE MUCH SIMPLER THAN THE INNER INTERLEAVING 4HE INNER CODE IS BINARY TRELLIS CODED MODULATION 4#- WITH ONE DIMENSIONAL SIGNAL CONSTELLATION 4HE REASON FOR THIS CHOICE IS THAT THEY WERE FOUND TO PROVIDE A GOOD DIVER SITY FACTOR AT A VERY LOW DECODER COMPLEXITY ;= 4HE CODE USED IS A ONE DIMENSIONAL  STATE CODE WITH  LEVEL UNIFORM PULSE AMPLITUDE MODULATION  0!-  4HE  0!OUTPUT SYMBOLS ARE THEN COMBINED TO A  POINT QUADRATURE AMPLITUDE MODULATION  1!- CONSTELLATION AND INTERLEAVED TO BREAK UP CHANNEL MEMORY )N THE RECEIVER THE



0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING

6ITERBI ALGORITHM 6! IS USED FOR DECODING 4HIS ALGORITHM IS CAPABLE OF USING THE CHAN NEL STATE INFORMATION OBTAINED FROM A PILOT SEQUENCE SEE 3ECTION  4HE DECODING IS PERFORMED BY MINIMIZING THE METRIC n 8n nB n  )  nGw n JXw ` Ww J 
w

WHERE B Gw IS THE CHANNEL ESTIMATE Xw IS THE DEINTERLEAVED OBSERVATION AFTER EQUALIZATION THE REAL OR IMAGINARY PART AND Ww IS A POTENTIAL CODEWORD "ECAUSE OF THE OUTER CODE THE 6! SHOULD BE MODIED TO PROVIDE RELIABILITY ESTIMATES TOGETHER WITH THE DECODED SEQUENCE 4HIS ENABLES SOFT DECODING OF THE OUTER CODE AS WELL "Y APPLYING A SOFT OUTPUT 6ITERBI ALGORITHM 3/6!  AN IMPROVEMENT OF ABOUT  D" IS OBTAINED ;= 4OGETHER WITH INNER CODING MULTICARRIER SIGNALLING AND SLOW FREQUENCY HOPPING THE INTERLEAVER PROVIDES DUAL TIMEFREQUENCY DIVERSITY !N EXAMPLE OF SLOW FREQUENCY HOPPING IS DEPICTED IN &IGURE  %ACH PROGRAM CONSISTS OF A NUMBER OF SUBCARRIERS AND /&$- SYMBOLS THE

&IGURE  3LOW FREQUENCY HOPPING %ACH PROGRAM /r USES A BANDWIDTH ! AND CHANGES FREQUENCY BAND AFTER 3qxy  PARAMETERS ! AND 3qxy ARE CHOSEN TO MAXIMIZE THE DIVERSITY OF THE SYSTEM ! SIMILAR BUT SOMEWHAT MORE GENERALIZED HOPPING SCHEME IS PRESENTED IN ;=  /THER SYSTEMS

4HERE HAVE BEEN OTHER CODED /&$- SYSTEMS PROPOSED AND ANALYZED IN THE LITERATURE )N ; = AN /&$-&- SYSTEM IS INVESTIGATED AND SIMULATED /&$- IS PROPOSED IN %UROPE AS THE TRANSMISSION TECHNIQUE FOR THE NEW $6" SYSTEM ;= WHERE A MULTIRESO LUTION SCHEME IS USED TOGETHER WITH JOINT SOURCECHANNEL CODING ; = 4HIS ALLOWS SEVERAL BIT RATES AND THEREBY A GRACEFUL DEGRADATION OF IMAGE QUALITY IN THE FRINGES OF THE BROADCAST AREA  #ODING ON FADING CHANNELS

#ODES FOR FADING CHANNELS HAVE BEEN INVESTIGATED FOR A LONG TIME AND THE SEARCH FOR GOOD CODES IS STILL GOING ON !N OVERVIEW OF THIS SUBJECT IS FOUND IN ;= $ESIGN OF CODES

 HAS BEEN ANALYZED IN ;   = AND FOR TRELLIS CODED MULTIPLE 03+ IN ; = !SYMMETRIC 03+ CONSTELLATIONS ARE CONSIDERED IN ;= AND CODE DESIGN FOR 2ICIAN FADING CHANNELS IS INVESTIGATED IN ;= 0ERFORMANCE BOUNDS HAVE BEEN DERIVED FOR 2AYLEIGH FADING CHANNELS IN ;      = AND FOR 2ICIAN FADING CHANNELS IN ;  = 5SUALLY PERFORMANCE ANALYSIS OF CODES ASSUMES PERFECT KNOWLEDGE OF THE CHANNEL (OWEVER IN ; = AN ANALYTICAL METHOD WAS INTRODUCED THAT ALLOWS NON IDEAL CHANNEL INFORMATION 4HIS WAS LATER GENERALIZED TO INCLUDE NON IDEAL INTERLEAVING ; = 4HIS METHOD HAS BEEN USED TO ANALYZE A CODED /&$- SYSTEM WITH PILOT BASED CHANNEL ESTI MATION ON 2AYLEIGH FADING CHANNELS ;= ! MAJOR BENET OF USING ANALYTICAL METHODS FOR EVALUATION OF CODED SYSTEMS IS THAT A CODED BIT ERROR RATE CAN BE OBTAINED QUICKLY AFTER SYSTEM MODICATIONS WITHOUT TIME CONSUMING SIMULATIONS



7IRED SYSTEMS

!N IMPORTANT DIdERENCE BETWEEN A WIRED AND A WIRELESS SYSTEM IS THE CHARACTERISTICS OF THE CHANNEL )N THE WIRED CASE THE CHANNEL IS OFTEN CONSIDERED STATIONARY WHICH FACILI TATES A NUMBER OF TECHNIQUES TO IMPROVE THE COMMUNICATION SYSTEM #HANNEL CODING IN COMBINATION WITH A TECHNIQUE CALLED BIT LOADING IS OFTEN EMPLOYED FOR THIS PURPOSE -UL TIDIMENSIONAL TRELLIS CODES ARE WELL SUITED FOR THE CHANNEL CODING 7HEN USING BIT LOADING THE SUBCHANNELS ARE ASSIGNED INDIVIDUAL NUMBERS OF BITS ACCORDING TO THEIR RESPECTIVE 3.2S !N /&$- BASED COMMUNICATION SYSTEM USING BIT LOADING IS OFTEN REFERRED TO AS A $-4 SYSTEM  "IT LOADING

"IT LOADING IS A TECHNIQUE THAT IS USED FOR MULTICARRIER SYSTEMS OPERATING ON STATIONARY CHANNELS ;= ! STATIONARY CHANNEL MAKES IT POSSIBLE TO MEASURE THE 3.2 ON EACH SUB CHANNEL AND ASSIGN INDIVIDUAL NUMBERS OF TRANSMITTED BITS ! SUBCHANNEL WITH HIGH 3.2 THUS TRANSMITS MORE BITS THAN A SUBCHANNEL WITH LOW 3.2 &IGURE  SHOWS A SCHEMATIC PICTURE OF 3.2 AND HOW THE NUMBER OF BITS ON EACH SUBCHANNEL VARY ACCORDINGLY
3.2 "ITS

3UBCARRIER

3UBCARRIER

&IGURE  #HANNEL 3.2 LEFT AND CORRESPONDING NUMBER OF BITS ON EACH SUBCARRIER RIGHT  7HEN PERFORMING BIT LOADING ONE USUALLY OPTIMIZES FOR EITHER HIGH DATA RATE LOW AVERAGE TRANSMITTING ENERGY OR LOW ERROR PROBABILITY 4YPICALLY TWO OF THESE ARE KEPT



0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING

CONSTANT AND THE THIRD IS THE GOAL FOR THE OPTIMIZATION 7HICH PARAMETER SHOULD BE OPTIMIZED DEPENDS ON THE SYSTEM ITS ENVIRONMENT AND ITS APPLICATION 7HEN THERE IS ONLY ONE SYSTEM OPERATING ON A CABLE THIS SYSTEM NEITHER INTERFERES WITH NOR IS INTERFERED BY OTHER SYSTEMS 4HIS MEANS THAT CONTROLLING THE TRANSMITTING POWER TO REDUCE CROSSTALK IS NOT NECESSARY 'IVEN A DATA RATE AND A BIT ERROR PROBABILITY WHATEVER TRANSMISSION ENERGY NEEDED TO ACHIEVE THESE GOALS CAN BE USED WITHIN REASONABLE LIMITS  )N A MULTI SYSTEM ENVIRONMENT WHERE THERE ARE SEVERAL SYSTEMS TRANSMITTING IN THE SAME CABLE THE PROBLEM IS MORE COMPLICATED SINCE THE SYSTEMS EXPERIENCE CROSSTALK 4HE LEVEL OF CROSSTALK IS PROPORTIONAL TO THE TRANSMITTING POWER IN THE SYSTEMS SEE  AND   )T IS THEREFORE DESIRABLE TO HAVE AN EQUAL TRANSMISSION POWER IN ALL SYSTEMS TO OBTAIN EQUAL DISTURBANCE SITUATIONS )N A MULTI SYSTEM ENVIRONMENT THE AVERAGE TRANSMITTING POWER IS USUALLY XED AND THE OPTIMIZATION IS FOR EITHER HIGH DATA RATE OR LOW BIT ERROR RATE  "IT LOADING ALGORITHMS

4HERE ARE SEVERAL TECHNIQUES FOR BIT LOADING IN $-4 SYSTEMS AND SOME OF THESE ARE DESCRIBED IN ;   = !S MENTIONED EARLIER THERE ARE SEVERAL PARAMETERS THAT ONE CAN OPTIMIZE FOR -OST ALGORITHMS OPTIMIZE FOR HIGH DATA RATE OR LOW BIT ERROR RATE 'IVEN A CERTAIN DATA RATE AND AN ENERGY CONSTRAINT THE (UGHES (ARTOGS ALGORITHM PROVIDES THE BIT LOADING FACTORS THAT YIELD MINIMAL BIT ERROR RATE SEE EG ;= 4HE IDEA BEHIND THE (UGHES (ARTOGS ALGORITHM IS TO ASSIGN ONE BIT AT A TIME TO THE SUBCHANNELS 4HE ALGORITHM CALCULATES THE ENERGY COST TO SEND ONE BIT MORE ON EACH SUBCHANNEL 4HE SUBCHANNEL WITH THE SMALLEST ENERGY COST IS THEN ASSIGNED THE BIT 4HIS PROCEDURE IS REPEATED UNTIL A DESIRED BIT RATE IS OBTAINED #HOW HAS SHOWED THAT COMPLEXITY OF THE (UGHES (ARTOGS ALGORITHM IS PROPORTIONAL TO THE NUMBER OF SUBCHANNELS AND THE NUMBER OF BITS TRANSMITTED IN A $-4 FRAME ;= (E ALSO SUGGESTS A SUBOPTIMAL ALGORITHM OF LOWER COMPLEXITY IN ;= !N ALGORITHM THAT MAINTAINS AN EQUAL BIT ERROR PROBABILITY OVER ALL SUBCHANNELS GIVEN A DATA RATE AND AN ENERGY CONSTRAINT IS PRESENTED BY &ISCHER IN ;= ! SUBOPTIMAL WAY OF PERFORMING BIT LOADING TO ACHIEVE A HIGH DATA RATE WHILE MAIN TAINING A CONSTANT SYMBOL ERROR PROBABILITY ACROSS ALL SUBCHANNELS IS PRESENTED BY 4U ;= )N HIS ALGORITHM THE BIT LOADING FACTORS ARE CALCULATED ACCORDING TO t u $t Ft og
 ` KNF "  At  KNF  *}t WHERE At IS THE NUMBER OF BITS CARRIED BIT LOADING FACTOR ON SUBCARRIER J  $t THE AVERAGE  SYMBOL TRANSMISSION ENERGY Ft THE CHANNEL ATTENUATION AND }t THE NOISE VARIANCE 4HE CODING GAIN IS DENOTED og AND THE CONSTELLATION EXPANSION FACTOR DUE TO CODING IS DENOTED "  &URTHER TO OBTAIN A DESIRED SYMBOL ERROR RATE OF /h  THE DESIGN CONSTANT * IS CHOSEN TO v t uw /h  * 0   -h WHERE -h IS THE NUMBER OF NEAREST NEIGHBORS

 %XPRESSION  CAN BE VIEWED AS THE UNION BOUND FOR A 1!- CONSTELLATION WITH SOME MODICATIONS FOR CODING WHERE * IS THE 3.2 REQUIRED TO OBTAIN AN ERROR PROBABILITY  /h  4HE CHANNEL 3.2 INCLUDING CODING  $t Ft og }t  IS DIVIDED BY THE 3.2 REQUIRED TO TRANSMIT ONE BIT &INALLY THE NUMBER OF BITS NEEDED IN THE CODING KNF "  IS SUBTRACTED TO GET THE NUMBER OF BITS CARRIED BY SUBCHANNEL J )F THE NUMBER OF SYSTEMS TRANSMITTING IN A CABLE VARY THE AMOUNT OF CROSSTALK WILL VARY ACCORDINGLY 4O HANDLE THE SITUATION WHERE THE NUMBER OF TRANSMITTING SYSTEMS VARY ONE CAN EITHER DO THE BIT LOADING FOR A WORST CASE OR EMPLOY ADAPTIVE BIT LOADING #HOW ;= PRESENTS SUCH AN ADAPTIVE ALGORITHM CALLED THE BIT SWAP ALGORITHM WHICH IS DESIGNED FOR THE CASE WHEN A XED DATA RATE IS SPECIED 7HEN TRYING TO MAXIMIZE THE DATA RATE WITH A CONSTANT TRANSMITTING POWER IT IS OPTIMAL TO ALLOW BIT LOADING FACTORS TO SPAN A CONTINUOUS RANGE OF VALUES 4U PRESENTS SOME RESULTS IN ;= ON HOW THE GRANULARITY OF BIT LOADING FACTORS AdECTS THE OBTAINED DATA RATE IN SUCH A SYSTEM /NE WAY TO GET NON INTEGER BIT LOADING FACTORS IS TO USE MULTIDIMENSIONAL CODES -ULTIDIMENSIONAL CODES ALLOW A FRACTIONAL NUMBER OF BITS PER  $ SYMBOL TO BE TRANSMITTED ON EACH SUBCHANNEL &OR  $  $ AND  $ CODES THE GRANULARITIES BECOME  BIT  BITS AND  BITS PER  $ SYMBOL RESPECTIVELY !NOTHER TECHNIQUE REFERRED TO AS ENERGY LOADING IS TO ALLOW SOME SORT OF NE TUNING OF THE TRANSMITTED ENERGY ON THE SUBCHANNELS IE ADJUSTING THE ENERGY $t IN  SO THAT IT CORRESPONDS TO ONE OF THE SUPPORTED BIT LOADING FACTORS (OWEVER ENERGY LOADING ONLY WORKS IF THE TUNING IS SMALL WHICH REQUIRES MANY BIT LOADING FACTORS AND A SIDE EdECT IS THAT A MORE COMPLEX SIGNAL CONSTELLATION MAPPERDEMAPPER IS REQUIRED



#HANNEL CODING

#ODING IN $-4 HAS BEEN ANALYZED IN EG ; = ! TYPICAL CODING SCHEME FOR $-4 CONSISTS OF AN OUTER CODE AN INTERLEAVER AND AN INNER CODE $UE TO THE TYPE OF APPLI CATIONS THAT $-4 IS DESIGNED TO CARRY SEE 3ECTION  IT IS APPEALING TO KEEP A LOW DELAY BETWEEN TRANSMITTER AND RECEIVER 4HIS LIMITS THE INTERLEAVING DEPTH AND AdECTS THE CHOICE OF ERROR CORRECTING CODES 4HE CODING SCHEME ANALYZED IN ;= USES AN OUTER 2EED 3OLOMON CODE AND AN INNER TRELLIS CODE 4HE 2EED 3OLOMON CODE AND INTERLEAVING ARE DESIGNED TO REDUCE ERRORS DUE TO IMPULSE NOISE "Y USING ONLY ONE CODER THAT CODES ACROSS SUBCARRIERS THE DELAY IS SMALL COMPARED TO THE CASE WHERE ONE TRELLIS CODE IS USED FOR EACH SUBCARRIER 4HIS IS DUE TO THE 6ITERBI DECODERS NEED FOR A CERTAIN DECISION DEPTH TO MAKE A GOOD DECISION )N THE INVESTIGATED SYSTEM THE AMOUNT OF DATA SENT IN ONE $-4 FRAME IS ENOUGH FOR THE 6ITERBI ALGORITHM TO MAKE A DECISION -ULTIDIMENSIONAL CODES ARE WELL SUITED FOR $-4 SYSTEMS "Y USING SEVERAL  $ CON STELLATIONS ON DIdERENT SUBCHANNELS IT IS EASY TO CREATE MULTIDIMENSIONAL CONSTELLATIONS !S DESCRIBED EARLIER THE MULTIDIMENSIONAL CODES ALLOW FRACTIONAL BITS TO BE TRANSMITTED WHICH REDUCES THE GRANULARITY OF THE BIT LOADING FACTORS



0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING



$ISCUSSION

4HIS SECTION IS BOTH A DISCUSSION AND A SUMMARY OF THE MATERIAL PRESENTED EARLIER IN THIS REPORT /NE OF THE MAJOR ADVANTAGES OF /&$- IS ITS ROBUSTNESS AGAINST MULTIPATH PROP AGATION (ENCE ITS TYPICAL APPLICATIONS ARE IN TOUGH RADIO ENVIRONMENTS /&$- IS ALSO SUITABLE IN SINGLE FREQUENCY NETWORKS SINCE THE SIGNALS FROM OTHER TRANSMITTERS CAN BE VIEWED AS ECHOES IE MULTIPATH PROPAGATION 4HIS MEANS THAT IT IS FAVORABLE TO USE /&$- IN BROADCASTING APPLICATIONS SUCH AS $!" AND $6" 4HE USE OF /&$- IN MULTIUSER SYSTEMS HAS GAINED AN INCREASING INTEREST THE LAST FEW YEARS 4HE DOWNLINK IN THOSE SYSTEMS IS SIMILAR TO BROADCASTING WHILE THE UPLINK PUTS HIGH DEMANDS ON EG SYNCHRONIZATION 4HE FUTURE OF /&$- AS A TRANSMISSION TECHNIQUE FOR MULTIUSER SYSTEMS DEPENDS ON HOW WELL THESE PROBLEMS CAN BE SOLVED )N WIRED SYSTEMS THE STRUCTURE OF /&$- OdERS THE POSSIBILITY OF EbCIENT BITLOADING "Y ALLOCATING A DIdERENT NUMBER OF BITS TO DIdERENT SUBCHANNELS DEPENDING ON THEIR INDIVIDUAL 3.2S EbCIENT TRANSMISSION CAN BE ACHIEVED !LTHOUGH OTHER SYSTEMS HAVE BEEN PROPOSED /&$- IS THE DOMINATING TECHNIQUE ON EG DIGITAL SUBSCRIBER LINES .OTE THAT /&$- OFTEN GOES UNDER THE NAME $-4 WHEN USED IN WIRED SYSTEMS WITH BITLOADING 4HERE ARE ALSO PROBLEMS ASSOCIATED WITH /&$- SYSTEM DESIGN 4HE TWO MAIN OBSTA CLES WHEN USING /&$- ARE THE HIGH PEAK TO AVERAGE POWER RATIO AND SYNCHRONIZATION 4HE FORMER PUTS HIGH DEMANDS ON LINEARITY IN AMPLIERS 3YNCHRONIZATION ERRORS IN BOTH TIME AND FREQUENCY DESTROY THE ORTHOGONALITY AND CAUSE INTERFERENCE "Y USING A CYCLIC PREX THE TIMING REQUIREMENTS ARE SOMEWHAT RELAXED SO THE BIGGEST PROBLEMS ARE DUE TO HIGH FREQUENCY SYNCHRONIZATION DEMANDS $EGRADATION DUE TO FREQUENCY ERRORS CAN BE CAUSED BOTH BY DIdERENCES IN LOCAL OSCILLATORS AND BY $OPPLER SHIFTS ! GREAT DEAL OF EdORT IS THEREFORE SPENT ON DESIGNING ACCURATE FREQUENCY SYNCHRONIZERS FOR /&$- !S IN ANY DIGITAL COMMUNICATION SYSTEM THERE ARE TWO ALTERNATIVES FOR MODULATION COHERENT OR DIdERENTIAL 4HE %UROPEAN $!" SYSTEM USES DIdERENTIAL 103+ WHILE THE PROPOSED SCHEME FOR $6" IS COHERENT  1!- $IdERENTIAL 03+ IS SUITABLE FOR LOW DATA RATES AND GIVES SIMPLE AND INEXPENSIVE RECEIVERS WHICH IS IMPORTANT FOR PORTABLE CON SUMER PRODUCTS LIKE $!" RECEIVERS (OWEVER IN $6" THE DATA RATE IS MUCH HIGHER AND LOW BIT ERROR RATES ARE DIbCULT TO OBTAIN WITH DIdERENTIAL 03+ ! NATURAL CHOICE FOR $6" IS THEREFORE MULTIAMPLITUDE SCHEMES $UE TO THE STRUCTURE IN /&$- IT IS EASY TO DESIGN EbCIENT CHANNEL ESTIMATORS AND EQUALIZERS 4HIS IS ONE OF THE APPEALING PROPERTIES OF /&$- WHICH SHOULD BE EXPLOITED TO ACHIEVE HIGH SPECTRAL EbCIENCY #ODING IN WIRELESS /&$- SYSTEMS DOES NOT DIdER MUCH FROM CODING IN WIRELESS SINGLE CARRIER SYSTEMS 4HE MAIN DIdERENCE IS THAT INTERLEAVING IN /&$- ALLOWS SYMBOLS TO BE SPREAD IN BOTH TIME AND FREQUENCY 4HE POSSIBILITY TO INTERLEAVE IN FREQUENCY OVERCOMES THE DRAWBACK OF NOT OBTAINING DIVERSITY FROM THE EQUALIZER 3INCE EACH SUBCHANNEL EX PERIENCES AT FADING CODE DESIGN AND PERFORMANCE ANALYSIS DEVELOPED FOR AT FADING CHANNELS CAN BE USED $ECODING CAN BE PERFORMED WITH A 6ITERBI DECODER WHERE THE METRIC DEPENDS ON THE ESTIMATED CHANNEL ATTENUATIONS 4HIS MEANS THAT SYMBOLS ARE WEIGHED WITH THEIR RESPECTIVE CHANNEL STRENGTH 4HIS REDUCES THE EdECT OF ERRORS CAUSED BY SYMBOLS TRANSMITTED DURING A FADE



!CKNOWLEDGEMENTS
4HE AUTHORS WOULD LIKE TO EXTEND THEIR GRATITUDE TO THE STAd AT 4ELIA 2ESEARCH !" ,ULE AND THE COLLEAGUES AT THE $IVISION OF 3IGNAL 0ROCESSING ,ULE 5NIVERSITY OF 4ECHNOLOGY FOR PROVIDING VALUABLE COMMENTS AND CORRECTIONS

!

4IME FREQUENCY LATTICE

)N THIS APPENDIX WE GIVE A BRIEF OVERVIEW OF THE EdECTS OF PULSE SHAPING 7E ONLY CONSIDER TIME AND FREQUENCY DISPERSION AND EXCLUDE CHANNEL NOISE IN THIS ANALYSIS 7E DESCRIBE THE /&$- SIGNAL AS 8 RS  Wt&u t&u S  
t&u

WHERE THE FUNCTIONS t&u S ARE TRANSLATIONS OF A PROTOTYPE FUNCTION O S  t&u S  O S ` K~ Ds {ty  

4HIS ALLOWS US TO INTERPRET THE PULSE SHAPING PROBLEM 4HE RECEIVER USES THE FUNCTIONS t&u S THAT ARE TRANSLATIONS OF A POSSIBLY DIdERENT PROTOTYPE FUNCTION O S  t&u S  O S ` K~ Ds {ty  )N /&$- THESE FUNCTIONS FULL Ht&u S  t &u S I  p :J ` J   K ` K < 
 





WHERE Ha aI DENOTES THE %UCLIDEAN INNER PRODUCT ;= (ENCE THE TRANSMITTER AND RECEIVER FUNCTIONS ARE BI ORTHOGONAL ;= 4HIS SIMPLIES THE RECEIVER SINCE :  S CS  Wt&u  HR S  t&u S I  R S t&u 

(OWEVER A TIME OR FREQUENCY DISPERSIVE CHANNEL DESTROYS THIS ORTHOGONALITY "Y CAREFULLY CHOOSING O S AND O S  THE EdECTS OF THE LOSS OF ORTHOGONALITY CAN BE KEPT LOW !N /&$- SYSTEM MUST BE SUbCIENTLY RESISTANT TO BOTH TIME AND FREQUENCY DISPERSION 4HE FORMER CAN DEALT WITH BY INTRODUCING A GUARD SPACE USUALLY IN THE FORM OF A CYCLIC PREX  WHILE THE LATTER IS OFTEN APPROACHED BY PULSE SHAPING )N SYSTEMS WITH A CYCLIC PREX AND NO PULSE SHAPING THE FUNCTIONS O S AND O S ARE CHOSEN AS THE RECTANGULAR PULSE ALTHOUGH OF DIdERENT LENGTHS 4HE RECEIVER PROTOTYPE FUNCTION O S IN THIS CASE IS SHORTER THAN THE TRANSMITTER PROTOTYPE FUNCTION O S  WHICH CORRESPONDS TO THE REMOVAL OF THE CYCLIC PREX ! COMMON PROPAGATION MODEL IS OBTAINED BY ASSUMING THAT THE CHANNEL CONSISTS OF A NUMBER OF ELEMENTARY PATHS ;= WHERE EACH PATH IS DESCRIBED BY A DELAY A FREQUENCY OdSET AND A COMPLEX ATTENUATION 4HUS BY INVESTIGATING THE EdECTS OF A STATIC DELAY AND FREQUENCY OdSET THE SENSITIVITY TO A FADING MULTIPATH CHANNEL CAN BE EVALUATED ;= 4HIS



0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING

ANALYSIS CAN BE MADE WITH THE CROSS AMBIGUITY FUNCTION ;= OF THE PROTOTYPE FUNCTIONS O S AND O S :  s{iCS ~ E  O S O  S`~ D

WHICH CAN BE VIEWED AS A CROSSCORRELATION FUNCTION IN THE TIME FREQUENCY PLANE 4HE BI ORTHOGONALITY OF t&u AND t&u S REQUIRES THAT :  s {vu~ s{vyCS O S O Ht&u S  tv&uw S I  D S ` M~ D  Ds {vu~


M~  Ly  p :M L< 



4HIS IS A CONDITION ON THE SAMPLES OF THE CROSS AMBIGUITY FUNCTION AT POSITIONS M~  Ly  4HE CROSS AMBIGUITY FUNCTION SHOULD BE ZERO FOR ALL M L     BUT A DELAY OR FREQUENCY OdSET WILL DESTROY THE ORTOGONALITY SINCE ~ E IS NOT SAMPLED AT ITS ZEROS 7ITH A DELAY a~ AND A FREQUENCY OdSET ay  THE SIGNAL POWER IS J a~ ay J AND THE INTERFERENCE POWER CAN BE UPPER BOUNDED BY  ` J a~ ay J ;= 4HIS BOUNDS THE SIGNAL TO INTERFERENCE RATIO 3)2 FROM BELOW AS J a~ ay J  3)2 w  ` J a~ ay J 

4HUS THE CROSS AMBIGUITY FUNCTION IS A MEASURE OF THE INTERFERENCE IN THE SYSTEM CAUSED BY A DELAY OR A FREQUENCY OdSET )N ;= A PROTOTYPE FUNCTION IS CREATED WHICH IS CLAIMED TO HAVE A NEAR OPTIMUM CROSS AMBIGUITY FUNCTION 4HE CROSS AMBIGUITY FUNCTION FOR A RECTANGULAR PULSE IN A SYSTEM WITH CYCLIC PREX IS SHOWN IN &IGURE  .OTE THAT

&IGURE  !MBIGUITY FUNCTION FOR A RECTANGULAR PULSE AND CYCLIC PREX WITH LENGTHS ~   AND 3fy   RESPECTIVELY THE SYSTEM IS INSENSITIVE TO A TIME DELAY LESS THAN 3fy   SINCE THE CROSS AMBIGUITY FUNCTION IS AT AT THE TOP

 4HE MAIN PROBLEM WITH USING A RECTANGULAR PULSE O S IS THAT IT IS NOT WELL LOCALIZED IN FREQUENCY $ENOTE THE TIME AND FREQUENCY WIDTHS RESPECTIVELY OF THE UNIT ENERGY SIGNAL O S BY :   aS  S JO S J CS   :   aE  E  J/ E J CE  P WHERE / E IS THE &OURIER TRANSFORM OF O S  4HEN aS  ~   AND aE   FOR THE RECTANGULAR PULSE 4HIS FREQUENCY SPREAD OF ENERGY IS THE REASON FOR )#) IN THE CASE OF TRANSMISSION OVER FREQUENCY DISPERSIVE CHANNELS ;= 4HUS OTHER PULSES HAVE BEEN SOUGHT  TO OVERCOME THIS PROBLEM 3INCE THE 'AUSSIAN PULSE O S  D{ HAS A MINIMAL TIME BANDWIDTH PRODUCT ;= IT HAS BEEN USED TO FORM SUITABLE FUNCTIONS ; = 0ROLATE SPHEROIDAL WAVE FUNCTIONS ;= HAVE ALSO BEEN USED TO MINIMIZE OUT OF BAND ENERGY IN PULSES UNDER CERTAIN CONDITIONS ; =



0ART  !N )NTRODUCTION TO /RTHOGONAL &REQUENCY $IVISION -ULTIPLEXING

0ART  -, %STIMATION OF 4IMING AND &REQUENCY /dSET IN /&$- 3YSTEMS

4HIS PART HAS BEEN ACCEPTED FOR PUBLICATION IN )%%% 4RANSACTIONS ON 3IGNAL 0ROCESSING AS ** VAN DE "EEK - 3ANDELL 0/ "RJESSON -, %STIMATION OF 4IMING AND &REQUENCY /dSET IN /&$- 3YSTEMS



-, %STIMATION OF 4IMING AND &REQUENCY /dSET IN /&$- 3YSTEMS

!BSTRACT 7E PRESENT THE JOINT -AXIMUM ,IKELIHOOD SYMBOL TIMING AND CARRIER FREQUENCY OdSET ESTIMATOR IN /&$- SYSTEMS 2EDUNDANT INFORMATION CONTAINED WITHIN THE CYCLIC PREX ENABLES THIS ESTIMATION WITHOUT ADDITIONAL PILOTS 3IMULA TIONS SHOW THAT THE FREQUENCY ESTIMATOR MAY BE USED IN A TRACKING MODE AND THE TIMING ESTIMATOR IN AN ACQUISITION MODE



)NTRODUCTION

/RTHOGONAL FREQUENCY DIVISION MULTIPLEXING /&$- SYSTEMS HAVE RECENTLY GAINED IN CREASED INTEREST /&$- IS USED IN THE %UROPEAN DIGITAL BROADCAST RADIO SYSTEM AND IS BEING INVESTIGATED FOR OTHER WIRELESS APPLICATIONS SUCH AS DIGITAL BROADCAST TELEVISION AND MOBILE COMMUNICATION SYSTEMS AS WELL AS FOR BROADBAND DIGITAL COMMUNICATION ON EXISTING COPPER NETWORKS 3EE ; = AND THE REFERENCES THEREIN 7E ADDRESS TWO PROBLEMS IN THE DESIGN OF /&$- RECEIVERS /NE PROBLEM IS THE UN KNOWN /&$- SYMBOL TIMING 3ENSITIVITY TO A TIMING OdSET IS HIGHER IN MULTI CARRIER SYSTEMS THAN IN SINGLE CARRIER SYSTEMS AND HAS BEEN DISCUSSED IN ; = ! SECOND PROBLEM IS THE MISMATCH OF THE OSCILLATORS IN THE TRANSMITTER AND RECEIVER 4HE DEMOD ULATION OF A SIGNAL WITH AN OdSET IN THE CARRIER FREQUENCY CAN CAUSE A HIGH BIT ERROR RATE AND MAY DEGRADE THE PERFORMANCE OF A SYMBOL SYNCHRONIZER ; = ! SYMBOL CLOCK AND A FREQUENCY OdSET ESTIMATE MAY BE GENERATED AT THE RECEIVER WITH THE AID OF PILOT SYMBOLS KNOWN TO THE RECEIVER ; = OR AS IN ;= BY MAXIMIZING THE AVERAGE LOG LIKELIHOOD FUNCTION (OWEVER THE STRUCTURE OF THE TRANSMITTED /&$- SIGNAL ITSELF OdERS THE OPPORTUNITY FOR SYNCHRONIZATION 3UCH AN APPROACH IS FOUND IN ;  = FOR A TIMING OdSET AND IN ;  = FOR A FREQUENCY OdSET 7E PRESENT AND EVALUATE THE JOINT MAXIMUM LIKELIHOOD -, ESTIMATION OF THE TIMING AND CARRIER FREQUENCY OdSET IN /&$- SYSTEMS 4HE KEY ELEMENT THAT WILL RULE THE DISCUSSION IS THAT THE /&$DATA SYMBOLS ALREADY CONTAIN SUbCIENT INFORMATION TO PERFORM SYNCHRONIZATION /UR NOVEL ALGORITHM EXPLOITS THE CYCLIC PREX PRECEDING THE /&$- SYMBOLS THUS REDUCING THE NEED FOR PILOTS



4HE /&$- SYSTEM MODEL

&IGURE  ILLUSTRATES THE BASEBAND DISCRETE TIME /&$- SYSTEM MODEL WE INVESTIGATE 4HE COMPLEX DATA SYMBOLS ARE MODULATED BY MEANS OF AN INVERSE DISCRETE &OURIER TRANS FORM )$&4 ON - PARALLEL SUBCARRIERS 4HE RESULTING /&$- SYMBOL IS SERIALLY TRANS MITTED OVER A DISCRETE TIME CHANNEL WHOSE IMPULSE RESPONSE WE ASSUME IS SHORTER THAN + SAMPLES !T THE RECEIVER THE DATA ARE RETRIEVED BY MEANS OF A DISCRETE &OURIER TRANSFORM $&4



0ART  -, %STIMATION OF 4IMING AND &REQUENCY /dSET IN /&$- 3YSTEMS

&IGURE  4HE /&$- SYSTEM TRANSMITTING SUBSEQUENT BLOCKS OF - COMPLEX DATA SYMBOLS !N ACCEPTED MEANS OF AVOIDING INTERSYMBOL INTERFERENCE )3) AND PRESERVING ORTHOG ONALITY BETWEEN SUBCARRIERS IS TO COPY THE LAST + SAMPLES OF THE BODY OF THE /&$SYMBOL - SAMPLES LONG AND APPEND THEM AS A PREAMBLE THE CYCLIC PREX TO FORM THE COMPLETE /&$- SYMBOL ; = 4HE EdECTIVE LENGTH OF THE /&$- SYMBOL AS TRANSMITTED IS THIS CYCLIC PREX PLUS THE BODY +
- SAMPLES LONG  4HE INSERTION OF A CYCLIC PREX CAN BE SHOWN TO RESULT IN AN EQUIVALENT PARALLEL ORTHOGONAL CHANNEL STRUCTURE WHICH ALLOWS FOR SIMPLE CHANNEL ESTIMATION AND EQUALIZATION ;= )N SPITE OF THE LOSS OF TRANSMISSION POWER AND BANDWIDTH ASSOCIATED WITH THE CYCLIC PREX THESE PROPERTIES GENERALLY MOTIVATE ITS USE ; = )N THE FOLLOWING ANALYSIS WE ASSUME THAT THE CHANNEL IS NON DISPERSIVE AND THAT THE TRANSMITTED SIGNAL R J IS ONLY AdECTED BY COMPLEX ADDITIVE WHITE 'AUSSIAN NOISE !7'. M J  7E WILL HOWEVER EVALUATE OUR ESTIMATORS PERFORMANCE FOR BOTH THE !7'. CHANNEL AND A TIME DISPERSIVE CHANNEL #ONSIDER TWO UNCERTAINTIES IN THE RECEIVER OF THIS /&$- SYMBOL THE UNCERTAINTY IN THE ARRIVAL TIME OF THE /&$- SYMBOL SUCH AMBIGUITY GIVES RISE TO A ROTATION OF THE DATA SYMBOLS AND THE UNCERTAINTY IN CARRIER FREQUENCY A DIdERENCE IN THE LOCAL OSCILLATORS IN THE TRANSMITTER AND RECEIVER GIVES RISE TO A SHIFT OF ALL THE SUBCARRIERS  4HE RST UNCERTAINTY IS MODELLED AS A DELAY IN THE CHANNEL IMPULSE RESPONSE IE p J ` t  WHERE t IS THE INTEGER VALUED UNKNOWN ARRIVAL TIME OF A SYMBOL 4HE LATTER IS MODELLED AS A COMPLEX MULTIPLICATIVE DISTORTION OF THE RECEIVED DATA IN THE TIME DOMAIN Ds {t(E  WHERE DENOTES THE DIdERENCE IN THE TRANSMITTER AND RECEIVER OSCILLATORS AS A FRACTION OF THE INTER CARRIER SPACING - IN NORMALIZED FREQUENCY  .OTICE THAT ALL SUBCARRIERS EXPERIENCE THE SAME SHIFT  4HESE TWO UNCERTAINTIES AND THE !7'. THUS YIELD THE RECEIVED SIGNAL Q J  R J ` t Ds {t(E
M J   4WO OTHER SYNCHRONIZATION PARAMETERS ARE NOT ACCOUNTED FOR IN THIS MODEL &IRST

 AN OdSET IN THE CARRIER PHASE MAY AdECT THE SYMBOL ERROR RATE IN COHERENT MODULATION )F THE DATA IS DIdERENTIALLY ENCODED HOWEVER THIS EdECT IS ELIMINATED !N OdSET IN THE SAMPLING FREQUENCY WILL ALSO AdECT THE SYSTEM PERFORMANCE 7E ASSUME THAT SUCH AN OdSET IS NEGLIGIBLE 4HE EdECT OF NON SYNCHRONIZED SAMPLING IS INVESTIGATED IN ;= .OW CONSIDER THE TRANSMITTED SIGNAL R J  4HIS IS THE $&4 OF THE DATA SYMBOLS Wt  WHICH WE ASSUME ARE INDEPENDENT (ENCE R J IS A LINEAR COMBINATION OF INDEPENDENT IDENTICALLY DISTRIBUTED RANDOM VARIABLES )F THE NUMBER OF SUBCARRIERS IS SUbCIENTLY LARGE WE KNOW FROM THE CENTRAL LIMIT THEOREM THAT R J APPROXIMATES A COMPLEX 'AUSSIAN PROCESS WHOSE REAL AND IMAGINARY PARTS ARE INDEPENDENT 4HIS PROCESS HOWEVER IS NOT WHITE SINCE THE APPEARANCE OF A CYCLIC PREX YIELDS A CORRELATION BETWEEN SOME PAIRS OF SAMPLES SPACED - SAMPLES APART (ENCE Q J IS NOT A WHITE PROCESS EITHER BUT BECAUSE OF ITS PROBABILISTIC STRUCTURE IT CONTAINS INFORMATION ABOUT THE TIMING OdSET t AND CARRIER FREQUENCY OdSET  4HIS IS THE CRUCIAL OBSERVATION THAT OdERS THE OPPORTUNITY FOR JOINT ESTIMATION OF THESE PARAMETERS BASED ON Q J  ! SYNCHRONIZER CANNOT DISTINGUISH BETWEEN PHASE SHIFTS INTRODUCED BY THE CHANNEL AND THOSE INTRODUCED BY SYMBOL TIMING DELAYS ;= 4IMING ERROR REQUIREMENTS MAY RANGE FROM THE ORDER OF ONE SAMPLE WIRELESS APPLICATIONS WHERE THE CHANNEL PHASE IS TRACKED AND CORRECTED BY THE CHANNEL EQUALIZER TO A FRACTION OF A SAMPLE IN EG ($3, WHERE THE CHANNEL IS STATIC AND ESSENTIALLY ESTIMATED ONLY DURING START UP  7ITHOUT A FREQUENCY OdSET THE FREQUENCY RESPONSE OF EACH SUBCHANNEL IS ZERO AT ALL OTHER SUBCARRIER FREQUENCIES IE THE SUBCHANNELS DONT INTERFERE WITH ONE OTHER ;= 4HE EdECT OF A FREQUENCY OdSET IS A LOSS OF ORTHOGONALITY BETWEEN THE TONES 4HE RESULTING INTERCARRIER INTERFERENCE )#) HAS BEEN INVESTIGATED IN ;= 4HE EdECTIVE SIGNAL TO NOISE RATIO 2-1 DUE TO BOTH ADDITIVE NOISE AND )#) IS SHOWN TO BE LOWER BOUNDED BY 2-1h 2-1 w 
 a 2-1 RHM { t RHM { { u 

  }w  4HE DIdERENCE BETWEEN THE 2-1 AND THE 2-1h IS A MEASURE OF WHERE 2-1  } THE SENSITIVITY TO A FREQUENCY OdSET  .OTICE THAT IN THE ABSENCE OF ADDITIVE NOISE THE FREQUENCY OdSET MUST SATISFY J J v  a   IN ORDER TO OBTAIN AN 2-1h OF  D" OR HIGHER 4HIS RESULT AGREES WELL WITH THE ANALYSIS OF MULTIUSER /&$- SYSTEMS IN ;= WHICH STATES THAT A FREQUENCY ACCURACY OF   OF THE INTER CARRIER SPACING IS NECESSARY



-AXIMUM LIKELIHOOD ESTIMATION

!SSUME THAT WE OBSERVE -

+ CONSECUTIVE SAMPLES OF Q J  CF &IGURE  AND THAT THESE SAMPLES CONTAIN ONE COMPLETE -
+ SAMPLE /&$- SYMBOL 4HE POSITION OF THIS SYMBOL WITHIN THE OBSERVED BLOCK OF SAMPLES HOWEVER IS UNKNOWN BECAUSE THE CHANNEL DELAY t IS UNKNOWN TO THE RECEIVER $ENE THE INDEX SETS ) s Ft     t
+ ` G AND )  s Ft
-     t
-
+ ` G  SEE &IGURE  4HE SET )  THUS CONTAINS THE INDEXES OF THE DATA SAMPLES THAT ARE COPIED INTO THE CYCLIC PREX AND THE SET ) CONTAINS THE INDEXES OF THIS PREX #OLLECT THE OBSERVED SAMPLES IN THE -
+ b  VECTOR Q s :Q     Q -
+ <Q  .OTICE THAT THE SAMPLES IN THE CYCLIC PREX AND THEIR COPIES



0ART  -, %STIMATION OF 4IMING AND &REQUENCY /dSET IN /&$- 3YSTEMS

&IGURE  3TRUCTURE OF /&$- SIGNAL WITH CYCLICLY EXTENDED SYMBOLS R J  4HE SET ) CONTAINS THE CYCLIC PREX IE THE COPIES OF THE + DATA SAMPLES IN )   Q J  J  ) ; )   ARE PAIRWISE CORRELATED IE    L  }
}w  s { } D LJ  )  $ FQ J Q J
L G     OTHERWISE h i h i   WHERE } s $ JR J J AND }w s $ JM J J  WHILE THE REMAINING SAMPLES Q J  J    ) ; ) ARE MUTUALLY UNCORRELATED 4HE LOG LIKELIHOOD FUNCTION FOR t AND  c t  IS THE LOGARITHM OF THE PROBABILITY DENSITY FUNCTION E QJt OF THE -
+ OBSERVED SAMPLES IN Q GIVEN THE ARRIVAL TIME t AND THE CARRIER FREQUENCY OdSET  )N THE FOLLOWING WE WILL DROP ALL ADDITIVE AND POSITIVE MULTIPLICATIVE CONSTANTS THAT SHOW UP IN THE EXPRESSION OF THE LOG LIKELIHOOD FUNCTION SINCE THEY DONT AdECT THE MAXIMIZING ARGUMENT -OREOVER WE DROP THE CONDITIONING ON t FOR NOTATIONAL CLARITY 5SING THE CORRELATION PROPERTIES OF THE OBSERVATIONS Q THE LOG LIKELIHOOD FUNCTION CAN BE WRITTEN AS c t  KNF E Q Jt 9  KNF E Q J Q J
9
t(

@
t

 KNF

9 E Q J Q J
E Q J E Q J
t@

@X@

E Q J


9
t

E Q J



WHERE E a DENOTES THE PROBABILITY DENSITY FUNCTION OF THE VARIABLES IN ITS ARGUMENT .OTICE THAT IT IS USED FOR BOTH ONE AND TWO DIMENSIONAL DISTRIBUTIONS 4HE PRODUCT 1 E Q J IN  IS INDEPENDENT OF t SINCE THE PRODUCT IS OVER ALL J AND SINCE t THE DENSITY E Q J IS ROTATIONALLY INVARIANT  3INCE THE -, ESTIMATION OF t AND IS THE ARGUMENT MAXIMIZING c t  WE MAY OMIT THIS FACTOR 5NDER THE ASSUMPTION THAT Q IS A JOINTLY 'AUSSIAN VECTOR  IS SHOWN IN THE !PPENDIX TO BE c t  Jo t J BNR {
;o t ` |h t  

 WHERE ; DENOTES THE ARGUMENT OF A COMPLEX NUMBER o L s

v8 C  t (v

Q J Q J
- 



vC  8 h L s JQ J J
JQ J
- J   t(v



AND

n n n n  n n J
G $ F Q J Q 2-1 n  } Q |sn  h i h i n n   2-1
 n $ JQ J J $ JQ J
- J n }
}w



IS THE MAGNITUDE OF THE CORRELATION COEbCIENT BETWEEN Q J AND Q J

-  4HE RST TERM IN  IS THE WEIGHTED MAGNITUDE OF o t  A SUM OF + CONSECUTIVE CORRELATIONS BETWEEN PAIRS OF SAMPLES SPACED - SAMPLES APART 4HE WEIGHTING FACTOR DEPENDS ON THE FREQUENCY OdSET 4HE TERM h t IS AN ENERGY TERM INDEPENDENT OF THE FREQUENCY OdSET  .OTICE THAT ITS CONTRIBUTION DEPENDS ON THE 3.2 BY THE WEIGHTING FACTOR |  4HE MAXIMIZATION OF THE LOG LIKELIHOOD FUNCTION CAN BE PERFORMED IN TWO STEPS L@W c t
t&

 L@W L@W c t
t

 L@W c t B- , t 
t



4HE MAXIMUM WITH RESPECT TO THE FREQUENCY OdSET IS OBTAINED WHEN THE COSINE TERM IN  EQUALS ONE 4HIS YIELDS THE -, ESTIMATION OF B- , t  `  ;o t
M { 

WHERE M IS AN INTEGER ! SIMILAR FREQUENCY OdSET ESTIMATOR HAS BEEN DERIVED IN ;= UNDER DIdERENT ASSUMPTIONS .OTICE THAT BY THE PERIODICITY OF THE COSINE FUNCTION SEVERAL MAX IMA ARE FOUND 7E ASSUME THAT AN ACQUISITION OR ROUGH ESTIMATE OF THE FREQUENCY OdSET HAS BEEN PERFORMED AND THAT J J   THUS M   3INCE BNR { B- , t
;o t   THE LOG LIKELIHOOD FUNCTION OF t THE COMPRESSED LOG LIKELIHOOD FUNCTION WITH RESPECT TO BECOMES c t B- , t  Jo t J ` |h t BECOMES   

AND THE JOINT -, ESTIMATION OF t AND

B- ,  @QF L@W FJo t J ` |h t G  t t r s  B- ,  B- ,  ` ;o t {

.OTICE THAT ONLY TWO QUANTITIES AdECT THE LOG LIKELIHOOD FUNCTION AND THUS THE PERFOR MANCE OF THE ESTIMATOR  THE NUMBER OF SAMPLES IN THE CYCLIC PREX + AND THE CORRELATION COEbCIENT | GIVEN BY THE 2-1 4HE FORMER IS KNOWN AT THE RECEIVER AND THE LATTER CAN BE XED "ASICALLY THE QUANTITY o t PROVIDES THE ESTIMATES OF t AND  )TS MAGNITUDE



0ART  -, %STIMATION OF 4IMING AND &REQUENCY /dSET IN /&$- 3YSTEMS

B- ,  WHILE ITS PHASE AT THIS TIME COMPENSATED BY AN ENERGY TERM PEAKS AT TIME INSTANT t INSTANT IS PROPORTIONAL TO B- ,  )F IS A PRIORI KNOWN TO BE ZERO THE LOG LIKELIHOOD FUNC B- , IS ITS MAXIMIZING ARGUMENT 4HIS TION FOR t BECOMES c t  1DFo t G ` |h t AND t ESTIMATOR AND A LOW COMPLEX VARIANT IS ANALYZED IN ;= )N AN /&$- RECEIVER THE QUANTITY o t  DENED IN   IS CALCULATED ON LINE &IG B- , URE  SHOWS c t B- , t WHOSE MAXIMIZING ARGUMENTS ARE THE TIMING ESTIMATES t B- , YIELD THE FREQUENCY ESTIMATES  AND ` { ;o t WHOSE VALUES AT THE TIME INSTANTS t .OTICE THAT  AND  DESCRIBE AN OPEN LOOP STRUCTURE #LOSED LOOP IMPLEMEN

&IGURE  4HE SIGNALS THAT GENERATE THE -, ESTIMATES -   +     B- ,  AND 3.2  D"  4HE MAXIMUMS OF c t B- , t TOP GIVE THE TIMING ESTIMATE t !T THESE TIME INSTANTS THE ARGUMENT OF o t BOTTOM GIVES B- , 

TATIONS BASED ON  AND  MAY ALSO BE CONSIDERED )N SUCH STRUCTURES THE SIGNAL c t B- , t IS TYPICALLY FED BACK IN A PHASE LOCKED LOOP 0,,  )F WE CAN ASSUME THAT t IS CONSTANT OVER A CERTAIN PERIOD THE INTEGRATION IN THE 0,, CAN SIGNICANTLY IMPROVE THE PERFORMANCE OF THE ESTIMATORS





3IMULATIONS

7E USE -ONTE #ARLO SIMULATIONS TO EVALUATE THE PERFORMANCE OF THE ESTIMATORS IN WHICH WE CONSIDER AN /&$- SYSTEM WITH  SUBCARRIERS )N EACH SIMULATION  SYMBOLS ARE USED 7E EVALUATE THE PERFORMANCE OF THE ESTIMATORS BY MEANS OF THE ESTIMATOR MEAN SQUARED ERROR 0ERFORMANCE RESULTS FOR THE !7'. CHANNEL ARE SHOWN IN &IGURES  AND  &IRST

&IGURE  0ERFORMANCE OF THE TIMING TOP AND FREQUENCY BOTTOM ESTIMATORS FOR THE !7'. CHANNEL  D"  D" AND  D"  4HE DIMENSIONLESS PERFORMANCE MEASURE IS EXPRESSED IN SQUARED UNITS RELATIVE TO THE SAMPLE INTERVAL TOP AND THE INTER TONE SPACING BOTTOM  4HE NUMBER OF SUBCARRIERS IS -   THE ESTIMATOR MEAN SQUARED ERROR AS A FUNCTION OF + IS ESTIMATED &IGURE  SHOWS THE ESTIMATOR PERFORMANCE FOR 2-1 VALUES OF  D"  D" AND  D" .OTICE THAT THE PERFORMANCE OF THE TIMING ESTIMATOR IS ASYMPTOTICALLY INDEPENDENT OF + PROVIDED THAT THE CYCLIC PREX IS LONGER THAN A CERTAIN THRESHOLD VALUE 4HIS THRESHOLD VALUE DECREASES WITH THE 2-1 "OTH THE TIMING ESTIMATOR AND THE FREQUENCY ESTIMATOR EXHIBIT SUCH A



0ART  -, %STIMATION OF 4IMING AND &REQUENCY /dSET IN /&$- 3YSTEMS

&IGURE  0ERFORMANCE OF THE TIMING TOP AND FREQUENCY BOTTOM ESTIMATORS FOR THE !7'. CHANNEL +   +   AND +   AND THE DISPERSIVE CHANNEL +    4HE DIMENSIONLESS PERFORMANCE MEASURE IS EXPRESSED IN SQUARED UNITS RELATIVE TO THE SAMPLE INTERVAL TOP AND THE INTER TONE SPACING BOTTOM  4HE NUMBER OF SUBCARRIERS IS -   PERFORMANCE THRESHOLD BASED ON + "UT NOTICE THAT AS + INCREASES BEYOND THESE RESPECTIVE THRESHOLDS ONLY THE FREQUENCY ESTIMATOR WILL SHOW CONTINUED IMPROVEMENT 4HUS FOR THE !7'. CHANNEL AND FROM A TIME SYNCHRONIZATION VIEWPOINT THERE IS VERY LITTLE ADVANTAGE IN INCREASING THE LENGTH OF THE CYCLIC PREX BEYOND THE TIMING ESTIMATORS THRESHOLD 3ECOND THE ESTIMATOR VARIANCES AS A FUNCTION OF 2-1 FOR +   +   AND +   ARE SHOWN IN &IGURE  .OTICE THAT EVEN IN THESE PLOTS A THRESHOLD PHENOMENON AS IN &IGURE  OCCURS 4HIS PHENOMENON IS A PROPERTY OF TIME DELAY ESTIMATION AND IS DOCUMENTED IN EG ;= 4HE ABOVE RESULTS DO NOT DIRECTLY APPLY TO A TIME DISPERSIVE CHANNEL ENVIRONMENT 4HEREFORE WE ALSO CONSIDER THE PERFORMANCE OF OUR ESTIMATORS IN A WIRELESS SYSTEM OPER

 ATING AT  '(Z WITH A BANDWIDTH OF  -(Z !N OUTDOOR DISPERSIVE FADING ENVIRONMENT WITH MICRO CELL CHARACTERISTICS IS CHOSEN THE CHANNEL HAS AN EXPONENTIALLY DECAYING POWER DELAY PROLE WITH ROOT MEAN SQUARED WIDTH EQUAL TO  xS CORRESPONDING TO  SAMPLES AND A MAXIMUM DELAY SPREAD OF  xS CORRESPONDING TO  SAMPLES  )T IS MODELLED TO CONSIST OF  INDEPENDENT 2AYLEIGH FADING TAPS ;= AND ADDITIVE NOISE 7E CHOOSE A CYCLIC PREX CONSISTING OF  SAMPLES 4HIS CHOICE AVOIDS )3) WHILE THE LOSS OF POWER AND BANDWIDTH DUE TO THE CYCLIC PREX + -
+ IS ABOUT   4HIS SYSTEM TRANSMITS ABOUT  /&$- SYMBOLS PER SECOND EACH CONTAINING  COMPLEX INFORMATION SYMBOLS )N THIS DISPERSIVE ENVIRONMENT THE DENITION OF t IS AMBIGIOUS 7E DENE THE TRUE DELAY AS THE CENTER OF GRAVITY OF THE CHANNEL IMPULSE RESPONSE -OREOVER WE DENE THE 2-1   AS 2-1  } /q }w WHERE /q IS THE SUM OF THE AVERAGE POWER IN ALL CHANNEL TAPS )N THE SIMULATIONS WE HAVE NORMALIZED /q   4HE ERROR OOR IN &IGURE  CLEARLY SHOWS THE PERFORMANCE DEGRADATION CAUSED BY THE DISPERSIVE CHANNEL AS COMPARED TO THE CORRESPONDING CURVES FOR THE !7'. CHANNEL &IGURE   )N THE DISPERSIVE CASE THE ESTIMATORS OPERATE IN AN ENVIRONMENT FOR WHICH THEY ARE NOT DESIGNED THEY ARE NOT OPTIMAL  3IGNALS PASSED THROUGH THE !7'. CHANNEL WILL HAVE THE SIMPLE PAIRWISE CORRELATION STRUCTURE   BUT SIGNALS PASSED THROUGH A DISPERSIVE CHANNEL GENERALLY HAVE A MORE COMPLEX CORRELATION STRUCTURE $EPENDING ON THE APPLICATION AND THE PRESENCE OF A HIGH PERFORMANCE CHANNEL ESTIMATOREQUALIZER THE PERFORMANCE OF THE TIMING ESTIMATE IN &IGURE  STANDARD DEVI ATION OF  SAMPLES MAY BE GOOD ENOUGH TO GENERATE A STABLE CLOCK )N MOST SITUATIONS THIS PERFORMANCE WILL SUbCE AT LEAST IN AN ACQUISITION MODE 4HE FREQUENCY OdSET ESTI MATOR SHOWS AN ERROR STANDARD DEVIATION OF LESS THAN  OF THE INTER TONE SPACING SEE &IGURE   WHICH SATISES THE REQUIREMENTS DISCUSSED IN 3ECTION  &INALLY THE PERFORMANCE OF THE FREQUENCY ESTIMATOR IS PLOTTED IN &IGURE  BY MEANS OF THE 3.2 LOSS IN D" 2-1h 2-1 4HE 3.2 LOSS IS CF  A FUNCTION OF  7E ASSUME THAT A FREQUENCY OdSET CAN BE CORRECTED USING THE ESTIMATE B- , AND WE THUS USE THE STANDARD DEVIATION OF THE ESTIMATE AS THE ARGUMENT IN   4HE 3.2 LOSS IS PLOTTED FOR THE !7'. CHANNEL AND THE DISPERSIVE CHANNEL AND FOR +   .OTICE THAT EVEN FOR THE DISPERSIVE CHANNEL THIS LOSS DOES NOT EXCEED  D"



$ISCUSSION

7E HAVE PRESENTED THE JOINT -, ESTIMATOR OF TIMING AND FREQUENCY OdSET IN /&$SYSTEMS )T USES THE REDUNDANT INFORMATION CONTAINED WITHIN THE CYCLIC PREX )T IS DERIVED UNDER THE ASSUMPTION THAT THE CHANNEL DISTORTION ONLY CONSISTS OF ADDITIVE NOISE BUT SIMULATIONS SHOW THAT IT CAN PERFORM WELL EVEN WITH A DISPERSIVE CHANNEL 4HE -, ESTIMATOR FOR THE LATTER CASE MAY BE DERIVED BUT IT WILL NOT HAVE THE SAME SIMPLE STRUCTURE AS OUR PROPOSED ESTIMATOR 4HE FREQUENCY ESTIMATOR PERFORMS BETTER THAN THE TIMING ESTIMATOR BECAUSE OF ITS IMPLICIT AVERAGING &ROM  AND   THE ESTIMATE IS THE ARGUMENT OF A SUM OF COMPLEX NUMBERS 7ITHOUT ADDITIVE NOISE M J  EACH TERM Q J Q J
- HAS THE SAME ARGUMENT `{  (ENCE THEY CONTRIBUTE COHERENTLY TO THE SUM WHILE THE ADDITIVE NOISE CONTRIBUTES INCOHERENTLY 4HIS EXPLAINS WHY THE PERFORMANCE WILL IMPROVE AS THE SIZE OF



0ART  -, %STIMATION OF 4IMING AND &REQUENCY /dSET IN /&$- 3YSTEMS

&IGURE  0ERFORMANCE OF THE FREQUENCY ESTIMATOR FOR THE !7'. CHANNEL AND THE DISPERSIVE CHANNEL +    4HE NUMBER OF SUBCARRIERS IS -   THE CYCLIC PREX INCREASES )N WIRELESS SYSTEMS PILOTS ARE NEEDED FOR CHANNEL ESTIMATION 4HESE PILOTS CAN BE USED BY THE SYNCHRONIZER IN ORDER TO FURTHER INCREASE PERFORMANCE 2ESULTING SYNCHRONIZERS MAY BE HYBRID STRUCTURES USING BOTH PILOTS AND THE REDUNDANCY OF THE CYCLIC PREX (OW TO INCORPORATE PILOT SYMBOLS IN SUCH TIMING AND FREQUENCY ESTIMATORS IS NOT STRAIGHTFOR WARD AND NEEDS FURTHER RESEARCH

!

4HE LOG LIKELIHOOD FUNCTION

t C  8 t (t

4HE LOG LIKELIHOOD FUNCTION  CAN BE WRITTEN AS c t 

t KNF

E Q J Q J
E Q J E Q J
-

u  

4HE NOMINATOR IS A TWO DIMENSIONAL COMPLEX VALUED 'AUSSIAN DISTRIBUTION WHICH USING THE CORRELATION PROPERTIES   BECOMES t u s ts | IhFhs{ t tE GstE s DWO `  }  | } w E Q J Q J
-       {  }
}w  ` |
c

WHERE | IS THE MAGNITUDE OF THE CORRELATION COEbCIENT BETWEEN Q J AND Q J

- AS DENED IN   4HE DENOMINATOR OF  CONSISTS OF TWO ONE DIMENSIONAL COMPLEX

 'AUSSIAN DISTRIBUTIONS s r sts DWO ` }  }w E Q J     { }

}w c t  B
B Jo t J BNR {
;o t ` |h t 



AND THE LOG LIKELIHOOD FUNCTION   AFTER SOME ALGEBRAIC MANIPULATIONS BECOMES 

WHERE o L AND h L ARE DENED IN  AND  AND B AND B ARE CONSTANTS IN DEPENDENT OF t AND  3INCE THE MAXIMIZING ARGUMENT OF c t IS INDEPENDENT OF THE r s B CONSTANTS B AND B  THE -, ESTIMATE t- ,  B- , ALSO MAXIMIZES  



0ART  -, %STIMATION OF 4IMING AND &REQUENCY /dSET IN /&$- 3YSTEMS

0ART  /&$- #HANNEL %STIMATION BY 3INGULAR 6ALUE $ECOMPOSITION

4HIS PART HAS BEEN PUBLISHED AS / %DFORS - 3ANDELL ** VAN DE "EEK 3+ 7ILSON 0/ "RJESSON /&$- #HANNEL %STI MATION BY 3INGULAR 6ALUE $ECOMPOSITION 2ESEARCH 2EPORT 45,%!  $IVISION OF 3IGNAL 0ROCESSING ,ULE 5NIVERSITY OF 4ECHNOLOGY 4HIS PAPER HAS BEEN PRESENTED IN PART AT THE  6EHICULAR 4ECHNOLOGY #ONFERENCE !TLANTA 'EORGIA 53! !PR  -AY  PP  



/&$- #HANNEL %STIMATION BY 3INGULAR 6ALUE $ECOMPOSITION

!BSTRACT )N THIS PAPER WE PRESENT AND ANALYSE LOW RANK CHANNEL ESTIMATORS FOR OR THOGONAL FREQUENCY DIVISION MULTIPLEXING /&$- USING THE FREQUENCY CORRELATION OF THE CHANNEL ,OW RANK APPROXIMATIONS BASED ON THE DISCRETE &OURIER TRANSFORM $&4 HAVE BEEN PROPOSED BUT THEY SUdER FROM POOR PERFORMANCE WHEN THE CHAN NEL IS NOT SAMPLE SPACED 7E APPLY THE THEORY OF OPTIMAL RANK REDUCTION TO LINEAR MINIMUM MEAN SQUARED ERROR ,--3% ESTIMATORS AND SHOW THAT THESE ESTIMA TORS WHEN USING A XED DESIGN ARE ROBUST TO CHANGES IN CHANNEL CORRELATION AND SIGNAL TO NOISE RATIO 3.2  4HE PERFORMANCE IS PRESENTED IN TERMS OF UNCODED SYMBOL ERROR RATE 3%2 FOR A SYSTEM USING  1!-



)NTRODUCTION

7IRELESS DIGITAL COMMUNICATION SYSTEMS USING MULTI AMPLITUDE MODULATION SCHEMES SUCH AS QUADRATURE AMPLITUDE MODULATION 1!-  REQUIRE ESTIMATION AND TRACKING OF THE FADING CHANNEL )N GENERAL THIS MEANS A MORE COMPLEX RECEIVER THAN FOR DIdERENTIAL MODULATION SCHEMES SUCH AS DIdERENTIAL PHASE SHIFT KEYING $03+  WHERE THE RECEIVERS OPERATE WITHOUT A CHANNEL ESTIMATE ;= )N ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING /&$- SYSTEMS $03+ IS APPROPRIATE FOR RELATIVELY LOW DATA RATES SUCH AS IN THE %UROPEAN DIGITAL AUDIO BROADCAST $!" SYSTEM ;= (OWEVER FOR MORE SPECTRALLY EbCIENT /&$- SYSTEMS COHERENT MODULATION IS MORE APPROPRIATE 4HE STRUCTURE OF /&$- SIGNALLING ALLOWS A CHANNEL ESTIMATOR TO USE BOTH TIME AND FREQUENCY CORRELATION 3UCH A TWO DIMENSIONAL ESTIMATOR STRUCTURE IS GENERALLY TOO COM PLEX FOR A PRACTICAL IMPLEMENTATION 4O REDUCE THE COMPLEXITY SEPARATING THE USE OF TIME AND FREQUENCY CORRELATION HAS BEEN PROPOSED ;= 4HIS COMBINED SCHEME USES TWO SEPARATE &)2 7IENER LTERS ONE IN THE FREQUENCY DIRECTION AND THE OTHER IN THE TIME DIRECTION )N THIS PAPER WE PRESENT AND ANALYSE A CLASS OF BLOCK ORIENTED CHANNEL ESTIMATORS FOR /&$- WHERE ONLY THE FREQUENCY CORRELATION OF THE CHANNEL IS USED IN THE ESTIMATION 7HATEVER THEIR LEVEL OF PERFORMANCE IT MAY BE IMPROVED WITH THE ADDITION OF A SECOND LTER USING THE TIME CORRELATION ; = 4HOUGH A LINEAR MINIMUM MEAN SQUARED ERROR ,--3% ESTIMATOR USING ONLY FRE QUENCY CORRELATION HAS LOWER COMPLEXITY THAN ONE USING BOTH TIME AND FREQUENCY COR RELATION IT STILL REQUIRES A LARGE NUMBER OF OPERATIONS 7E INTRODUCE A LOW COMPLEXITY APPROXIMATION TO A FREQUENCY BASED ,--3% ESTIMATOR THAT USES THE THEORY OF OPTIMAL RANK REDUCTION /THER TYPES OF LOW RANK APPROXIMATIONS BASED ON THE DISCRETE TIME &OURIER TRANSFORM $&4  HAVE BEEN PROPOSED FOR /&$- SYSTEMS BEFORE ;  = 4HE WORK PRESENTED IN THIS PAPER WAS INSPIRED BY THE OBSERVATIONS IN ;= WHERE IT IS SHOWN THAT $&4 BASED LOW RANK CHANNEL ESTIMATORS HAVE LIMITED PERFORMANCE FOR NON



0ART  /&$- #HANNEL %STIMATION BY 3INGULAR 6ALUE $ECOMPOSITION

SAMPLE SPACED CHANNELS AND HIGH 3.2S !FTER PRESENTING THE /&$- SYSTEM MODEL AND OUR SCENARIO IN 3ECTION  WE IN TRODUCE THE ESTIMATORS AND DERIVE THEIR COMPLEXITIES IN 3ECTION  7E ANALYSE THE SYMBOL ERROR RATE 3%2 PERFORMANCE IN 3ECTION  WHERE WE ALSO DISCUSS DESIGN CONSID ERATIONS 4HE PROPOSED LOW RANK ESTIMATOR IS COMPARED TO OTHER ESTIMATORS IN 3ECTION  AND A SUMMARY AND CONCLUDING REMARKS APPEAR IN 3ECTION 




3YSTEM DESCRIPTION
3YSTEM MODEL

&IGURE  DISPLAYS THE /&$- BASE BAND MODEL USED IN THIS PAPER 7E ASSUME THAT THE USE OF A CYCLIC PREX #0 ;= BOTH PRESERVES THE ORTHOGONALITY OF THE TONES AND ELIMINATES INTER SYMBOL INTERFERENCE )3) BETWEEN CONSECUTIVE /&$- SYMBOLS &URTHER THE CHANNEL F S ~ IS ASSUMED TO BE SLOWLY FADING SO IT IS CONSIDERED TO BE CONSTANT DURING ONE /&$- SYMBOL 4HE NUMBER OF TONES IN THE SYSTEM IS -  AND THE LENGTH OF THE CYCLIC PREX IS + SAMPLES

&IGURE  "ASE BAND MODEL OF AN /&$- SYSTEM #0 DENOTES THE CYCLIC PREX 5NDER THESE ASSUMPTIONS WE CAN DESCRIBE THE SYSTEM AS A SET OF PARALLEL 'AUSSIAN CHANNELS SHOWN IN &IGURE  WITH CORRELATED ATTENUATIONS Gt  4HE ATTENUATIONS ON

&IGURE  4HE /&$- SYSTEM DESCRIBED AS A SET OF PARALLEL 'AUSSIAN CHANNELS WITH CORRELATED ATTENUATIONS

 EACH TONE ARE GIVEN BY t Gt  & J -3 u  J      - `  

WHERE & a IS THE FREQUENCY RESPONSE OF THE CHANNEL F S ~ DURING THE /&$- SYMBOL AND 3 IS THE SAMPLING PERIOD OF THE SYSTEM )N MATRIX NOTATION WE DESCRIBE THE /&$SYSTEM AS X  7G
M 

WHERE X IS THE RECEIVED VECTOR 7 IS A MATRIX CONTAINING THE TRANSMITTED SIGNALLING POINTS ON ITS DIAGONAL G IS A CHANNEL ATTENUATION VECTOR AND M IS A VECTOR OF IID COMPLEX   ZERO MEAN 'AUSSIAN NOISE WITH VARIANCE }w



#HANNEL MODEL
D  8 t(

7E ARE USING A FADING MULTI PATH CHANNEL MODEL ;= CONSISTING OF , IMPULSES F ~ 

mt p ~ ` ~t 3 



WHERE mt ARE ZERO MEAN COMPLEX 'AUSSIAN RANDOM VARIABLES WITH A POWER DELAY PROLE t ~t  )N THIS PAPER WE HAVE USED ,   IMPULSES AND TWO VERSIONS OF THIS CHANNEL MODEL q 3YNCHRONIZED CHANNEL 4HIS IS A MODEL OF A PERFECTLY TIME SYNCHRONIZED /&$SYSTEM WHERE THE RST FADING IMPULSE ALWAYS HAS A ZERO DELAY ~   AND OTHER FADING IMPULSES HAVE DELAYS THAT ARE UNIFORMLY AND INDEPENDENTLY DISTRIBUTED OVER THE LENGTH OF THE CYCLIC PREX 4HE IMPULSE POWER DELAY PROLE t ~t  "D~t (~v  DECAYS EXPONENTIALLY ;= q 5NIFORM CHANNEL !LL IMPULSES HAVE THE SAME AVERAGE POWER AND THEIR DELAYS ARE UNIFORMLY AND INDEPENDENTLY DISTRIBUTED OVER THE LENGTH OF THE CYCLIC PREX



3CENARIO

/UR SCENARIO CONSISTS OF A WIRELESS  1!- /&$- SYSTEM DESIGNED FOR AN OUTDOOR ENVIRONMENT THAT IS CAPABLE OF CARRYING DIGITAL VIDEO 4HE SYSTEM OPERATES AT  K(Z BANDWIDTH AND IS DIVIDED INTO  TONES WITH A TOTAL SYMBOL PERIOD OF  xS OF WHICH  xS IS THE CYCLIC PREX /NE /&$- SYMBOL THUS CONSISTS OF  SAMPLES -
+    FOUR OF WHICH ARE CONTAINED IN THE CYCLIC PREX +    4HE UNCODED DATA RATE OF THE SYSTEM IS  -"ITSEC 7E ASSUME THAT ~v   SAMPLE FOR THE SYNCHRONIZED CHANNEL



0ART  /&$- #HANNEL %STIMATION BY 3INGULAR 6ALUE $ECOMPOSITION



,INEAR CHANNEL ESTIMATION ACROSS TONES

)N THE FOLLOWING WE PRESENT THE ,--3% ESTIMATE OF THE CHANNEL ATTENUATIONS G FROM THE RECEIVED VECTOR X AND THE TRANSMITTED DATA 7 7E ASSUME THAT THE RECEIVED /&$SYMBOL CONTAINS DATA KNOWN TO THE ESTIMATOR EITHER TRAINING DATA OR RECEIVER DECISIONS 4HE COMPLEXITY REDUCTION OF THE ,--3% ESTIMATOR CONSISTS OF TWO SEPARATE STEPS )N THE RST STEP WE MODIFY THE ,--3% BY AVERAGING OVER THE TRANSMITTED DATA OBTAINING A SIMPLIED ESTIMATOR )N THE SECOND STEP WE REDUCE THE NUMBER OF MULTIPLICATIONS REQUIRED BY APPLYING THE THEORY OF OPTIMAL RANK REDUCTION ;=



,--3% ESTIMATION

4HE ,--3% ESTIMATE OF THE CHANNEL ATTENUATIONS G IN   GIVEN THE RECEIVED DATA X AND THE TRANSMITTED SYMBOLS 7 IS ;= r ` a s 9  B u B G Guvvh  1qq 1qq
}w 77  WHERE B u  7 X  G


X X XE  aaa W W WE 

wQ



 IS THE LEAST SQUARES CHANNEL NOISE AND h 9 i ,3 ESTIMATE OF G }w IS THE VARIANCE OF THE ADDITIVE 9 1qq  $ GG IS THE CHANNEL AUTOCORRELATION 4HE SUPERSCRIPT a DENOTES (ERMITIAN TRANSPOSE )N THE FOLLOWING WE ASSUME WITHOUT LOSS OF GENERALITY THAT THE VARIANCES OF h i  THE CHANNEL ATTENUATIONS IN G ARE NORMALIZED TO UNITY IE $ JGt J   4HE ,--3% ESTIMATOR  IS OF CONSIDERABLE COMPLEXITY SINCE A MATRIX INVERSION IS NEEDED EVERY TIME THE TRAINING DATA IN 7 CHANGES 7E REDUCE THE COMPLEXITY OF THIS ESTIMATOR BY AVERAGING OVER THE h TRANSMITTED DATA ;= IE WE REPLACE THE TERM 779  i IN  WITH ITS EXPECTATION $ 779   !SSUMING THE SAME SIGNAL CONSTELLATION i ON h 9  ALL TONES AND EQUAL PROBABILITY ON ALL CONSTELLATION POINTS WE HAVE $ 77   $ FJWt J G ( WHERE ( IS THE IDENTITY MATRIX $ENING THE AVERAGE SIGNAL TO NOISE RATIO  AS 2-1  $ FJWt J G }w  WE OBTAIN A SIMPLIED ESTIMATOR t u n B  1qq 1qq
Bu  G ( G  2-1

WHERE

h i h i n  $ JWt J $ JWt J



IS A CONSTANT DEPENDING ON THE SIGNAL CONSTELLATION )N THE CASE OF  1!- TRANSMISSION n   "ECAUSE 7 IS NO LONGER A FACTOR IN THE MATRIX CALCULATION THE INVERSION OF n 1qq
PEI ( DOES NOT NEED TO BE CALCULATED EACH TIME THE TRANSMITTED DATA IN 7 CHANGES &URTHERMORE IF 1qq AND 2-1 ARE KNOWN BEFOREHAND OR ARE SET TO XED NOMINAL VALUES n THE MATRIX 1qq 1qq
PEI (  NEEDS TO BE CALCULATED ONLY ONCE 5NDER THESE CONDITIONS THE ESTIMATION REQUIRES - MULTIPLICATIONS PER TONE 4O FURTHER REDUCE THE COMPLEXITY OF THE ESTIMATOR WE PROCEED WITH LOW RANK APPROXIMATIONS BELOW





/PTIMAL LOW RANK APPROXIMATIONS

/PTIMAL RANK REDUCTION IS ACHIEVED BY USING THE SINGULAR VALUE DECOMPOSITION 36$ ;= 4HE 36$ OF THE CHANNEL AUTOCOVARIANCE MATRIX IS 1qq  4c49  

WHERE 4 IS A UNITARY MATRIX CONTAINING THE EIGENVECTORS AND c IS A DIAGONAL MATRIX CONTAINING THE SINGULAR VALUES w w w w    w wE  ON ITS DIAGONAL  )N !PPENDIX ! IT IS SHOWN THAT THE OPTIMAL RANK O ESTIMATOR IS B y  4aO 49 G Bu G WHERE ay IS A DIAGONAL MATRIX WITH THE VALUES pt 
wt n wt  2-1



J        O `  J  O     - ` 



6IEWING THE ORTHONORMAL MATRIX 49 AS A TRANSFORM  THE SINGULAR VALUE wt OF 1qq IS THE CHANNEL POWER VARIANCE CONTAINED IN THE Jq TRANSFORM COEbCIENT AFTER TRANSFORMING B u  3INCE 4 IS UNITARY THIS TRANSFORMATION CAN BE VIEWED AS ROTATING THE THE ,3 ESTIMATE G Bu SO THAT ALL ITS COMPONENTS ARE UNCORRELATED ;= 4HE DIMENSION OF THE SPACE VECTOR G OF ESSENTIALLY TIME AND BAND LIMITED SIGNALS LEADS US TO THE RANK NEEDED IN THE LOW RANK ESTIMATOR )N ;= IT IS SHOWN THAT THIS DIMENSION IS ABOUT !3
 WHERE ! IS THE ONE SIDED BANDWIDTH AND 3 IS THE TIME INTERVAL OF THE SIGNAL !CCORDINGLY THE MAGNITUDE OF THE SINGULAR VALUES OF 1qq SHOULD DROP RAPIDLY AFTER ABOUT +
 LARGE VALUES WHERE + IS THE LENGTH OF THE CYCLIC PREX !  3  3  +3 AND !3
  +
  7E PRESENT THE CHANNEL POWER CONTAINED IN THE RST  COEbCIENTS IN &IGURE  4HE CALCULATIONS ARE BASED ON OUR SCENARIO AND THE TWO CHANNEL MODELS THE SYNCHRONIZED AND THE UNIFORM 4HE MAGNITUDE OF THE CHANNEL POWER DROPS RAPIDLY AFTER ABOUT J   IE  COEbCIENTS WHICH IS CONSISTENT WITH THE OBSERVATION THAT THE DIMENSION OF THE SPACE SPANNED BY 1qq IS APPROXIMATELY +
 THAT IS 
   IN THIS CASE ! BLOCK DIAGRAM OF THE RANK`O ESTIMATOR IN  IS SHOWN IN &IGURE  WHERE THE ,3 ESTIMATE IS CALCULATED FROM X BY MULTIPLYING BY 7 



%STIMATOR COMPLEXITY

4HE LIMITING FACTOR OF THE RANK`O ESTIMATORS IS AN ERROR OOR SEE 3ECTION  4O ELIMINATE THIS ERROR OOR UP TO A GIVEN 3.2 WE NEED TO MAKE SURE OUR ESTIMATOR RANK IS LARGE ENOUGH 4HIS PROMPTS AN ANALYSIS OF THE COMPUTATIONAL COMPLEXITY OF THE RANK`O ESTIMATOR 4HE
WE ARE DEALING WITH (ERMITIAN MATRICES THE wK S ARE ALSO EIGENVALUES (OWEVER WE USE THE TERMINOLOGY OF THE 36$ SINCE IT IS MORE GENERAL AND CAN BE USED IN OPTIMAL RANK REDUCTION OF NON (ERMITIAN MATRICES  4HE TRANSFORM IN THIS SPECIAL CASE OF LOW RANK APPROXIMATION IS THE +ARHUNEN ,OEVE AKA (OTELLING TRANSFORM OF H
 3INCE



0ART  /&$- #HANNEL %STIMATION BY 3INGULAR 6ALUE $ECOMPOSITION

&IGURE  2ELATIVE CHANNEL POWER wt %JGt J  IN THE TRANSFORM COEbCIENTS FOR THE TWO EXAMPLE CHANNELS

&IGURE  "LOCK DIAGRAM OF THE RANK`O CHANNEL ESTIMATOR IMPLEMENTATION WE HAVE CHOSEN IS BASED ON WRITING  AS A SUM OF RANK  MATRICES WHICH GIVES US THE EXPRESSION By  G y 8
t(

pt Tt T9 t $

B u  G %

y  8 t(

$ % Bu Pt T t  G



Bu  T9 G B WHERE Pt  pt Tt AND Tt  G t u IS THE %UCLIDIAN INNER PRODUCT 4HE LINEAR COM BINATION OF O VECTORS OF LENGTH - ALSO REQUIRES O- MULTIPLICATIONS 4HE ESTIMATION THUS REQUIRES O- MULTIPLICATIONS AND THE TOTAL NUMBER OF MULTIPLICATIONS PER TONE BECOMES O )N COMPARISON WITH THE FULL ESTIMATOR   WE HAVE MANAGED TO REDUCE THE NUMBER OF MULTIPLICATIONS FROM - TO O PER TONE 4HE SMALLER O IS THE LOWER THE COMPUTATIONAL COMPLEXITY BUT THE LARGER THE APPROXIMATION ERROR BECOMES &OLLOWING THE ANALYSIS IN 3ECTION  WE CAN EXPECT A GOOD APPROXIMATION WHEN O IS IN THE RANGE OF SAMPLES IN THE CYCLIC PREX WHICH IS USUALLY MUCH SMALLER THAN THE NUMBER OF TONES - 

 ! LEGITIMATE QUESTION AT THIS POINT IS WHAT HAPPENS FOR A SYSTEM WITH MANY TONES AND MANY SAMPLES IN THE CYCLIC PREX 4HE NUMBER OF CALCULATIONS PER TONE CAN BE CONSIDERABLE IF A RANK`O ESTIMATOR IS USED DIRECTLY ON ALL TONES IN THE SYSTEM /NE SOLUTION TO THIS PROBLEM IS A PARTITIONING OF THE TONES INTO REASONABLE SIZED BLOCKS AND AT A CERTAIN PERFORMANCE LOSS PERFORM THE ESTIMATION INDEPENDENTLY IN THESE BLOCKS "Y DIVIDING THE CHANNEL ATTENUATIONS INTO * EQUALLY SIZED BLOCKS THE BANDWIDTH IN EACH BLOCK IS REDUCED BY A FACTOR *  2EFERRING AGAIN TO THE DIMENSION OF THE SPACE OF ESSENTIALLY TIME AND BANDLIMITED SIGNALS ;= THE EXPECTED NUMBER OF ESSENTIAL BASE VECTORS IS REDUCED FROM +
 TO +*
 (ENCE THE COMPLEXITY OF THE ESTIMATOR DECREASES ACCORDINGLY 4O ILLUSTRATE THE IDEA LET US ASSUME WE HAVE A SYSTEM WITH -   TONES AND A +   SAMPLE CYCLIC PREX 4HE UNIFORM CHANNEL CORRELATION BETWEEN THE ATTENUATIONS Gv AND Gw IN THIS SYSTEM IS SEE !PPENDIX " Qv&w  
`w s {C vE  h `w s {C vE `

IF L  M IF L  M



4HIS ONLY DEPENDS ON THE DISTANCE BETWEEN THE TONES L ` M AND THE RATIO BETWEEN THE LENGTH OF THE CYCLIC PREX AND THE NUMBER OF TONES +-  4HE  TONE SYSTEM CAN BE DESCRIBED BY        X 7 G M                   
    ! ! ! ! X 7 G M THAT IS AS  PARALLEL  TONE SYSTEMS Xt  7t Gt
Mt  J        



7E HAVE THE SAME CHANNEL CORRELATION IN EACH SUBSYSTEM AS WE HAVE IN THE  TONE SCENARIO IN THIS PAPER +-      "Y ESTIMATING THE CHANNEL ATTENUATIONS Gt IN EACH SUB SYSTEM INDEPENDENTLY WE NEGLECT THE CORRELATION BETWEEN TONES IN DIdERENT SUB SYSTEMS BUT OBTAIN THE SAME -3% PERFORMANCE AS IN OUR  TONE SCENARIO



%STIMATOR PERFORMANCE AND DESIGN

7E PROPOSE A GENERIC LOW RANK FREQUENCY BASED CHANNEL ESTIMATOR IE THE ESTIMATOR IS DESIGNED FOR XED NOMINAL VALUES OF 3.2 AND CHANNEL CORRELATION (ENCE WE NEED TO ANALYSE HOW THE RANK CHANNEL CORRELATION AND 3.2 SHOULD BE CHOSEN FOR THIS ESTIMATOR SO THAT IT IS ROBUST TO VARIATIONS IN THE CHANNEL STATISTICS IE MISMATCH !S A PERFORMANCE MEASURE WE USE UNCODED SYMBOL ERROR RATE 3%2 FOR  1!- SIGNALLING 4HE 3%2 IN THIS CASE CAN BE CALCULATED FROM THE MEAN SQUARED ERROR -3% WITH THE FORMULAE IN ;=



0ART  /&$- #HANNEL %STIMATION BY 3INGULAR 6ALUE $ECOMPOSITION



h i 4HE MEAN SQUARED ERROR RELATIVE TO THE CHANNEL POWER $ JGt J  OF THE RANK`O ESTIMATOR IS MAINLY DETERMINED BY THE CHANNEL POWER CONTAINED IN THE TRANSFORM COEbCIENTS AND CAN BE EXPRESSED SEE !PPENDIX # u y t E   8  8 n   wt  ` pt
wt  LRD O  p
- t( 2-1 t - t (y WHERE wt AND pt ARE GIVEN BY  AND  RESPECTIVELY 4HE -3%  IS A MONOTON ICALLY DECREASING FUNCTION OF 2-1 AND CAN BE BOUNDED FROM BELOW BY THE LAST TERM
E   8 LRD O  wt v LRD O  - t (y

2ANK REDUCTION



WHICH IS THE SUM OF THE CHANNEL POWER IN THE TRANSFORM COEbCIENTS NOT USED IN THE ESTIMATE 4HIS -3% OOR LRD O  WILL GIVE RISE TO A ERROR OOR IN THE SYMBOL ERROR RATES 4HE ERROR OOR IS THE MAIN LIMITATION ON THE COMPLEXITY REDUCTION ACHIEVED BY OPTIMAL RANK REDUCTION !S AN ILLUSTRATION &IGURE  DISPLAYS THE 3%2 RELATIVE TO THE CHANNEL VARIANCE FOR THREE DIdERENT RANKS AS A FUNCTION OF THE 3.2 4HE RANKS CHOSEN ARE O    AND  AND THE CHANNEL USED IN THE EXAMPLE IS THE SYNCHRONIZED CHANNEL 4HE CORRESPONDING 3%2 OORS ARE SHOWN AS HORIZONTAL LINES &OR O   THE 3%2 OOR IS RELATIVELY SMALL AND THE 3%2 OF THE RANK` ESTIMATOR IS COMPARABLE TO THE ORIGINAL FULL RANK ESTIMATOR  IN THE RANGE  TO  D" IN 3.2 "Y CHOOSING THE APPROPRIATE RANK ON THE ESTIMATOR WE CAN ESSENTIALLY AVOID THE IMPACT FROM THE 3%2 OOR UP TO A GIVEN 3.2 7HEN WE HAVE FULL RANK O  -  NO 3%2 OOR EXISTS "ASED ON THE CHANNEL POWERS PRESENTED IN &IGURE  WE SHOW THE CORRESPONDING 3%2 OORS RELATIVE TO THE CHANNEL VARIANCE IN &IGURE  !FTER ABOUT RANK` THE 3%2 OOR DECREASES RAPIDLY 7E ARE THEREFORE ABLE TO OBTAIN A GOOD ESTIMATOR APPROXIMATION WITH A RELATIVELY LOW RANK



3%2 PERFORMANCE UNDER MISMATCH

)N PRACTICE THE TRUE CHANNEL CORRELATION AND 3.2 ARE NOT KNOWN 4O GET A GENERAL EXPRESSION FOR THE ESTIMATOR 3%2 WE DERIVE IT UNDER THE ASSUMPTION THAT THE ESTIMATOR IS DESIGNED FOR CORRELATION 1qq AND SIGNAL TO NOISE RATIO 2-1 BUT THE TRUE VALUES ARE E = 1E qE q AND 2-1 RESPECTIVELY WHERE G DENOTES A CHANNEL WITH DIdERENT STATISTICS THAN G 4HIS ALLOWS US TO ANALYSE THIS ESTIMATORS SENSITIVITY TO DESIGN ERRORS 5NDER THESE ASSUMPTIONS THE RELATIVE -3% OF THE RANK`O ESTIMATE  BECOMES SEE !PPENDIX # w y v E   8 n   8  LRD O  xt  ` pt
pt
xt  - t( - t (y = 2-1 WHERE xt IS THE Jq DIAGONAL ELEMENT OF 49 1E qE q 4 CF   )T CAN BE INTERPRETED AS THE 9 E VARIANCE OF THE TRANSFORMED CHANNEL 4 G UNDER CORRELATION MISMATCH SINCE |r sr s9 } 9 9 E 4 G E $ 4 G  49 1E  qE q4



&IGURE  ,OW RANK ESTIMATOR SYMBOL ERROR RATE AS A FUNCTION OF 3.2 WITH RANKS O    AND  #ORRESPONDING 3%2 OORS SHOWN AS HORIZONTAL LINES 3YNCHRONIZED CHANNEL

&IGURE  %STIMATOR 3%2 OOR AS A FUNCTION OF ESTIMATOR RANK #IRCLES SHOW THE 3%2 OORS APPEARING IN &IGURE  E ARE NO LONGER UNCORRELATED (OWEVER DUE TO )T SHOULD BE NOTED THAT THE ELEMENTS OF 49 G THE FACT THAT THE POWER DELAY PROLE IS SHORT COMPARED TO THE /&$- SYMBOL THE RST O ELEMENTS CAN BE EXPECTED TO CONTAIN MOST OF THE POWER 4HIS PROPERTY WILL ENSURE ONLY A SMALL PERFORMANCE LOSS WHEN THE ESTIMATOR IS DESIGNED FOR WRONG CHANNEL STATISTICS )F RANK`O ESTIMATORS ARE USED IN A REAL SYSTEM THE SENSITIVITY TO MISMATCH IN BOTH CHANNEL CORRELATION AND 3.2 ARE IMPORTANT 7E WILL SHOW THAT A RANK`O ESTIMATOR BASED



0ART  /&$- #HANNEL %STIMATION BY 3INGULAR 6ALUE $ECOMPOSITION

ON THE UNIFORM CHANNEL MODEL AND A NOMINAL 3.2 CAN BE USED AS XED GENERIC ESTIMATOR WITH ONLY A SMALL LOSS IN AVERAGE PERFORMANCE 7E DIVIDE THE MISMATCH ANALYSIS INTO TWO PARTS RST WE ANALYSE THE 3%2 WHEN WE HAVE A MISMATCH IN CHANNEL CORRELATION AND LATER WE ANALYSE THE 3%2 WHEN WE HAVE A MISMATCH IN 3.2  )NCORRECT CHANNEL CORRELATION

=  BUT INCORRECT CHANNEL CORRELATION &ROM   WITH NO 3.2 MISMATCH 2-1  2-1 1qq  1E qE q  WE OBTAIN THE PERFORMANCE FOR THE CORRELATION MISMATCH CASES 7E COM PARED THE PERFORMANCE OF OUR CHANNEL ESTIMATOR IN TWO MISMATCH SITUATIONS I USING THE A UNIFORM CHANNEL WHEN THE TRUE CHANNEL MODEL WAS THE SYNCHRONIZED CHANNEL AND II USING THE SYNCHRONIZED CHANNEL WHEN THE TRUE CHANNEL MODEL WAS THE UNIFORM CHANNEL 4HE RESULTING CHANNEL ESTIMATES THAT WERE USED IN THE DETECTION OF THE DATA PRODUCED NO NOTICABLE DIdERENCE IN SYMBOL ERROR RATES LESS THAN  D" CHANGE IN EdECTIVE 3.2 FOR AN AVERAGE 3.2 UP TO  D" 4O SHOW THE DIdERENCES MORE CLEARLY THE -3%S ARE DISPLAYED IN &IGURE  &OR THE MISMATCHED CASES MARKED WITH q THE UNIFORM DESIGN IS MORE ROBUST IE THE ERROR IN CASE OF MISMATCH IS LOWER 7ITH THE RESTRICTION THAT THE TRUE CHANNEL HAS A POWER DELAY PROLE SHORTER THAN THE CYCLIC PREX DESIGNING FOR A UNIFORM POWER DELAY PROLE CAN BE SEEN AS A MINIMAX DESIGN

&IGURE  -3% FOR CORRECT AND MISMATCHED DESIGN 4HE LATTER IS MARKED WITH CIRCLES q  )NCORRECT 3.2

&INALLY WE EVALUATE THE SENSITIVITY TO MISMATCH IN DESIGN 3.2 FOR A RANK  ESTIMATOR 7HEN THERE IS NO MISMATCH IN CHANNEL CORRELATION AND NOMINAL 3.2S OF   AND  D" ARE USED IN THE DESIGN THE SENSITIVITY TO 3.2 MISMATCH IS NOT THAT LARGE (OWEVER IN &IGURE  WE PRESENT THE 3%2 FOR THE SAME RANK  ESTIMATORS BUT WITH THE DIdERENCE

 THAT THE TRUE CHANNEL CORRELATION IS MISMATCHED WITH THE DESIGN CORRELATION )N THIS SECOND CASE THERE IS A CLEAR DIdERENCE BETWEEN THE TWO DESIGNS THE HIGHER THE NOMINAL DESIGN 3.2 THE BETTER THE OVERALL PERFORMANCE OF THE ESTIMATOR IN THE RANGE  TO  D" IN 3.2 )T SHOULD BE NOTED THAT A ,--3% ESTIMATOR DESIGNED FOR A LARGE 3.2 APPROACHES THE ,3 ESTIMATOR

&IGURE  2ANK  ESTIMATOR 3%2 WHEN 3.2S OF   AND  D" ARE USED IN THE DESIGN 4HE ESTIMATORS ARE DESIGNED FOR INCORRECT CHANNEL CORRELATION



'ENERIC LOW RANK ESTIMATOR

)F WE WANT A ROBUST GENERIC CHANNEL ESTIMATOR DESIGN FOR /&$- SYSTEMS OF THE LOW RANK TYPE THE ANALYSIS IN THE PREVIOUS SECTION SUGGESTS THE USE OF THE UNIFORM CHANNEL CORRELATION AND A RELATIVELY HIGH 3.2 AS NOMINAL DESIGN PARAMETERS 4HE DESIGN OF SUCH AN ESTIMATOR ONLY REQUIRES KNOWLEDGE ABOUT THE LENGTH OF THE CYCLIC PREX THE NUMBER OF TONES IN THE SYSTEM AND THE TARGET RANGE OF 3.2S FOR THE APPLICATION )F THE RECEIVER CANNOT AdORD AN ESTIMATOR THAT INCLUDES TRACKING OF CHANNEL CORRELATION AND 3.2 THIS CHANNEL ESTIMATOR WORKS REASONABLY WELL FOR XED 3.2 AND CHANNEL CORRELATION



0ERFORMANCE GAIN

&OR THE SCENARIO USED IN THIS PAPER 3EC  WE CHOOSE A RANK  ESTIMATOR WITH UNIFORM DESIGN AND 2-1   D" 4HE PERFORMANCE OF THIS ESTIMATOR IS PRESENTED IN &IG  WHERE THE 3%2 FOR THE ,3 ESTIMATE  AND KNOWN CHANNEL ARE ALSO SHOWN !S CAN BE SEEN THE LOW RANK ESTIMATOR IS  D" BETTER THAN THE ,3 ESTIMATOR AND LESS THAN  D" FROM THE KNOWN CHANNEL



0ART  /&$- #HANNEL %STIMATION BY 3INGULAR 6ALUE $ECOMPOSITION

&IGURE  3%2 FOR  1!- TRAINING DATA AND A SYNCHRONIZED CHANNEL 4HE GENERIC RANK` ESTIMATOR DESIGNED FOR A UNIFORM CHANNEL AND  D" IN 3.2 IS COMPARED TO THE ,3 ESTIMATOR AND KNOWN CHANNEL AT THE RECEIVER



#OMPARISON TO &)2 LTERS

!N ALTERNATIVE TO USING LOW RANK ESTIMATORS TO SMOOTH THE CHANNEL ESTIMATES IS TO USE A &)2 LTER INSTEAD (ENCE WE WILL COMPARE OUR PROPOSED LOW RANK ESTIMATORS TO &)2 LTERS OF THE SAME COMPLEXITY 4HE &)2 LTERS ARE O TAPS 7IENER LTERS ;= IE O MULTIPLICATIONS PER TONE THAT ARE DESIGNED FOR THE SAME CHANNEL CORRELATION AND 3.2 AS THE LOW RANK ESTIMATORS &IGURE  SHOWS THE 3%2 FOR RANK`O ESTIMATORS IN COMPARI SON WITH &)2 LTERS OF THE SAME COMPUTATIONAL COMPLEXITY 7HEN THE COMPLEXITY IS  MULTIPLICATIONS PER TONE ! THE RANK`O ESTIMATOR HAS ABOUT  D" ADVANTAGE IN 3.2 OVER THE &)2 LTER IN THE RANGE OF 3.2S SHOWN 7HEN THE NUMBER OF CALCULATIONS GOES DOWN TO  MULTIPLICATIONS PER TONE " THE 3%2 OOR OF THE RANK`O ESTIMATOR BECOMES VISIBLE AND THE &)2 LTER PERFORMS BETTER AT 3.2S ABOVE  D" (OWEVER IT SHOULD BE NOTED THAT THE PERFORMANCE OF THE LOW RANK ESTIMATORS DEPEND HEAVILY OF THE SIZE OF THE CYCLIC PREX )F THE CYCLIC PREX WERE TO BE DECREASED RELATIVE TO THE /&$- SYMBOL  THE LOW RANK ESTIMATOR WOULD INCREASE ITS PERFORMANCE 4HIS IS DUE TO THE FACT THAT THE DIMENSION OF THE CHANNEL WHOSE DURATION IS ASSUMED TO BE SHORTER THAN THE CYCLIC PREX DECREASES AND CAN THUS BE REPRESENTED WITH FEWER COEbCIENTS /N THE OTHER HAND IF THE CYCLIC PREX INCREASES IN SIZE MORE COEbCIENTS ARE NEEDED TO AVOID LARGE APPROXIMATION ERRORS (ENCE WHETHER OR NOT THE LOW RANK ESTIMATOR IS BETTER THAN THE &)2 LTER DEPENDS ON THE RELATIVE SIZE OF THE CYCLIC PREX AND THE ALLOWED COMPLEXITY



4HE USE OF TIME CORRELATION

4HE LOW RANK ESTIMATOR PRESENTED IN THIS PAPER IS BASED ON FREQUENCY CORRELATION ONLY BUT THE TIME CORRELATION OF THE CHANNEL CAN ALSO BE USED 4HE TWO DIMENSIONAL ,--3%



&IGURE  3%2 COMPARISON BETWEEN THE RANK`O ESTIMATORS AND &)2 7IENER LTERS OF THE SAME COMPLEXITY "OTH ESTIMATORS ARE DESIGNED FOR THE UNIFORM CHANNEL AND  D" 3.2 !  MULTIPLICATIONS PER TONE AND "  MULTIPLICATIONS PER TONE ESTIMATOR CAN BE SIMPLIED USING THE SAME TECHNIQUE WITH RANK REDUCTION AS DESCRIBED HERE (OWEVER IN ;= IT IS SHOWN THAT SUCH AN ESTIMATOR GIVES AN INFERIOR PERFORMANCE FOR A XED COMPLEXITY (ENCE IT SEEMS THAT SEPARATING THE USE OF FREQUENCY AND TIME CORRELATION IS THE MOST EbCIENT WAY OF ESTIMATING THE CHANNEL /THER APPROACHES TO USE THE TIME CORRELATION IS EG TO USE A DECISION DIRECTED SCHEME ;= OR &)2 LTERS ; = 4HE FORMER CAN BE USED IN A SLOW FADING ENVIRONMENT WHERE IT OdERS GOOD PERFORMANCE FOR A MINIMAL COMPLEXITY AND THE LATTER IS PREFERRED IN CASE OF FAST FADING )T IS POSSIBLE TO USE A BANK OF &)2 LTERS AND CHOOSE THE MOST APPROPRIATE ACCORDING THE ESTIMATED $OPPLER FREQUENCY ;=



#ONCLUSIONS

7E HAVE INVESTIGATED LOW COMPLEXITY LOW RANK APPROXIMATIONS OF THE ,--3% CHANNEL ESTIMATOR FOR NON SAMPLE SPACED CHANNELS 4HE INVESTIGATION SHOWS THAT AN ESTIMATOR ERROR OOR INHERENT IN THE LOW RANK APPROXIMATION IS THE SIGNICANT LIMITATION TO THE ACHIEVED COMPLEXITY REDUCTION 7E SHOWED THAT A GENERIC LOW RANK ESTIMATOR DESIGN BASED ON THE UNIFORM CHANNEL CORRELATION AND A NOMINAL 3.2 CAN BE USED IN OUR  TONE SCENARIO #OMPARED WITH THE FULL ,--3%   THERE IS ONLY A SMALL LOSS IN PERFORMANCE UP TO A 3.2 OF  D" BUT A REDUCTION IN COMPLEXITY WITH A FACTOR -O   &OR SYSTEMS WITH MORE SUBCHANNELS THIS GAIN IS EVEN LARGER 4HE GENERIC ESTIMATOR DESIGN ONLY REQUIRES KNOWLEDGE ABOUT THE LENGTH OF THE CYCLIC PREX THE NUMBER OF TONES IN THE SYSTEM AND THE TARGET RANGE OF 3.2S FOR THE APPLICATION 7E ALSO COMPARED LOW RANK ESTIMATORS TO &)2 LTERS ACROSS THE TONES 4HE COMPARISON SHOWED THAT AT LOW COMPLEXITIES AND HIGH 3.2S THE &)2 LTERS IS THE PREFERABLE CHOICE



0ART  /&$- #HANNEL %STIMATION BY 3INGULAR 6ALUE $ECOMPOSITION

DUE TO THE ERROR OOR IN THE LOW RANK APPROXIMATION (OWEVER IF WE CAN ALLOW UP TO  MULTIPLICATIONS PER TONE IN OUR SCENARIO THE LOW RANK ESTIMATOR IS MORE ADVANTAGEOUS !LSO THE LOW RANK ESTIMATORS IMPROVE THEIR PERFORMANCE AS THE CYCLIC PREX DECREASES IN SIZE

!

/PTIMAL RANK REDUCTION

 

4HE OPTIMAL RANK REDUCTION IS FOUND FROM THE CORRELATION MATRICES N O 9 B 1qB qu  $ GGu  1qq N O n 9 B B 1B qu B qu  $ Gu Gu  1qq
2-1 ( AND THE 36$  0 #09 1qB  qu 1B qu B qu
 



WHERE 0 AND 0 ARE UNITARY MATRICES AND # IS A DIAGONAL MATRIX WITH THE SINGULAR VALUES C w C w a a a w CE  ON ITS DIAGONAL 4HE BEST LOW RANK ESTIMATOR ;= IS THEN v w #y  ( B B  Gy  0 09  1qLS qLS Gu    WHERE #y IS THE O b O UPPER LEFT CORNER OF # IE WE EXCLUDE ALL BUT THE O LARGEST SINGULAR n VECTORS )N THIS PAPER WE HAVE 1qB qu  1qq AND 1B qu B qu  1qq
PEI ( AND WE NOTE THAT THEY SHARE THE SAME SINGULAR VECTORS IE THE ONES OF 1qq  4c49  4HUS WE MAY EXPRESS  AS t t u u( n 9 9 4c4 ( 4  4 c
2-1 t u( n 4c c
49  0 #09 (  2-1 t u( n 0  0  4 AND #  c c
(  2-1 4HE RANK O ESTIMATOR v B y  4 #y G  v #y  4   NOW BECOMES t t u( w u n  9 9 Bu  4 c
4 ( 4 G  2-1 u( v wt w n  a  y 9 Bu  4 B u c
( 4 G 49 G    2-1



WHERE ay IS THE O b O UPPER LEFT CORNER OF t u n w wE  ac c

(  CH@F  a a a  n n  2-1 w
PEI wE 
PEI



 .OTE THAT 0  0 SINCE WE ARE ESTIMATING THE SAME TONES AS WE ARE OBSERVING IE SMOOTHING AND AN EIGENVALUE DECOMPOSITION COULD BE USED TO ACHIEVE OPTIMAL RANK RE DUCTION )N THE GENERAL CASE WHEN EG PILOT SYMBOL ASSISTED MODULATION ;= IS USED AND THERE ARE KNOWN SYMBOLS PILOTS ON ONLY A PART OF THE SUBCHANNELS WE HAVE 0  0 SINCE 1qB qu AND 1B qu B qu DONT SHARE THE SAME SINGULAR VECTORS THE MATRICES ARE NOT EVEN OF THE SAME SIZE  (ENCE THE MORE GENERAL 36$ MUST BE USED WHICH MOTIVATES THE NOMENCLATURE IN THIS ARTICLE

"

#HANNEL CORRELATION MATRICES

D  8 r(

5SING THE CHANNEL MODEL IN   THE ATTENUATION ON TONE J BECOMES Gt 

mr Ds { E ~r 
t



AND THE CORRELATION MATRIX FOR THE ATTENUATION VECTOR G h i 1GG  $ GG9  :Qv&w < CAN BE EXPRESSED AS ~t S INDEPENDENT : Qv&w   aaa
D : 8 r(



: D  9
t(

E~t ~t

D  8 r(

t ~r Ds {~r
v`w E

v`w E

C~    C~D  

E~r ~r t ~r Ds {~r

C~r 

WHERE t ~ IS THE MULTI PATH INTENSITY PROLE AND E~t ~t IS THE PROBABILITY DENSITY FUNC TION OF ~t  4HE CORRELATION MATRICES OF THE THREE CHANNELS USED IN THIS PAPER ARE CALCULATED BELOW



0ART  /&$- #HANNEL %STIMATION BY 3INGULAR 6ALUE $ECOMPOSITION q 3YNCHRONIZED CHANNEL 4HE PROBABILITY DISTRIBUTIONS FOR THE DELAYS ARE E~ ~ E~r ~r  p ~  | + IF ~r  : +<   H        ,   OTHERWISE  

AND THE POWER DELAY PROLE IS t ~  " D~(~v  3UBSTITUTING IN   AND NOR MALIZING Qt&t TO UNITY GIVES US r s  {s v w C ~v D  E +
  s { v w  ` D ~v E r s Qv&w    C +
, `  ~v  ` D ~v
` `

q 5NIFORM CHANNEL 4HE PROBABILITY DISTRIBUTIONS FOR THE DELAYS ARE | + IF ~r  : +< E~r ~r   H        ,   OTHERWISE



AND THE POWER DELAY PROLE IS CONSTANT t ~  "  3UBSTITUTED IN   AND NORMALIZING Qt&t TO UNITY GIVES US Qv&w  


`w s {C vE v `w s {C E `

IF L  M IF L  M



#

%STIMATOR MEAN SQUARED ERROR

)N THIS APPENDIX WE DERIVE THE -3% OF THE RANK`O ESTIMATOR IN   7E ALSO PRESENT THE -3% OOR WHICH BOUNDS THE ACHIEVABLE -3% FROM BELOW IN LOW RANK APPROXIMATIONS OF THE ,--3% ESTIMATOR 4O GET A GENERAL EXPRESSION FOR THE MEAN SQUARED ERROR FOR THE RANK`O APPROXIMATION OF THE ,--3% ESTIMATOR WE ASSUME THAT THE ESTIMATOR HAS BEEN DESIGNED FOR CHANNEL CORRELATION 1qq AND SIGNAL TO NOISE RATIO 2-1 BUT THE REAL CHANNEL E HAS THE CORRELATION 1 E qq AND THE REAL SIGNAL TO NOISE RATIO IS 2-1 =  &ROM  AND   G  E B E  WHERE THE NOISE TERM M E  7 M HAS THE AUTOCOVARIANCE MATRIX WE HAVE Gu  G
M n E B 1w Ew E  PEI ( 4HE ESTIMATION ERROR Dy  G ` Gy OF THE RANK`O ESTIMATOR  IS t v wu v w ay  ay  'E E Dy  4 ( ` 4 G`4 4' M      AND THE MEAN SQUARED ERROR IS h i  LRD O  3Q@BD $ Dy D9 y  4O SIMPLIFY THE EXPRESSION WE USE THAT



 E AND M E ARE UNCORRELATED HENCE THE CROSS TERMS ARE CANCELLED IN THE EXPECTATION q G ` a q 3Q@BD 4 49  3Q@BD IF 4 IS A UNITARY MATRIX AND 3Q@BD
!  3Q@BD
3Q@BD ! ;= 0 q 3Q@BD # #  t @t&t C t WHEN # IS A DIAGONAL MATRIX WITH THE ELEMENTS Ct ON ITS DIAGONAL AND NOT NECESSARILY A DIAGONAL MATRIX HAS DIAGONAL ELEMENTS @t&t  5SING  IN   THE MEAN SQUARED ERROR BECOMES t v wu t v wu9  ay  ay  ' 49
LRD O  3Q@BD 4 ( ` 4 1G EG E4 ( `     v w w9 v ay  ay  ' 4 4 1M 4' EM E4     y  y E  8  8  8 n   xt  ` pt
xt
 p - t( - t( 2-1 = t t (y u y  t E   8 n   8   xt   xt  ` pt
pt
- t( - t (y = 2-1 WHERE xt IS THE CHANNEL POWER IN THE J TH TRANSFORM COEbCIENT IE THE JTH DIAGONAL E qq 4 4HE -3% CAN BE LOWER BOUNDED LRD O w LRD O  BY ELEMENT OF THE MATRIX 49 1 WHAT WE CALL THE -3% OOR
E   8 LRD O  xt  - t (y



` a )F THERE IS NO MISMATCH IN 2-1 OR CHANNEL CORRELATION WE HAVE xt  CH@F 49 1qq 4  =  2-1 AND THE -3% BECOMES wt AND 2-1
y t  8 LRD O  wt  ` pt - t(


u E  n   8
p
wt  2-1 t - t (y





0ART  /&$- #HANNEL %STIMATION BY 3INGULAR 6ALUE $ECOMPOSITION

0ART  ! #OMPARATIVE 3TUDY OF 0ILOT BASED #HANNEL %STIMATORS FOR 7IRELESS /&$-

4HIS PART HAS BEEN PUBLISHED AS - 3ANDELL / %DFORS ! #OMPARATIVE 3TUDY OF 0ILOT BASED #HANNEL %STIMATORS FOR 7IRELESS /&$-  2ESEARCH 2EPORT 45,%!  $IVISION OF 3IGNAL 0ROCESSING ,ULE 5NIVERSITY OF 4ECHNOLOGY



! #OMPARATIVE 3TUDY OF 0ILOT BASED #HANNEL %STIMATORS FOR 7IRELESS /&$!BSTRACT 4HIS REPORT DEALS WITH PILOT BASED CHANNEL ESTIMATION IN WIRELESS /&$SYSTEMS 7E ASSUME THAT THE RECEIVER IS ABLE TO USE ALL TRANSMITTED PILOTS WHICH IS THE CASE IN BROADCASTING AND IN THE DOWNLINK OF A MULTIUSER SYSTEM &OUR ESTI MATORS OF WHICH TWO HAVE BEEN PROPOSED IN THE LITERATURE ARE COMPARED BOTH IN TERMS OF MEAN SQUARED ERROR AND BIT ERROR RATES )N THE LATTER CASE WE SIMULATE A MULTIUSER SYSTEM WHICH INCORPORATES CHANNEL CODING 4HE CHANNEL ESTIMATION IS A TWO DIMENSIONAL PROBLEM TIME AND FREQUENCY  AND BOTH SEPARABLE AND NON SEPARABLE ESTIMATORS ARE INVESTIGATED 7E DESIGN LOW RANK APPROXIMATIONS OF THESE ESTIMATORS AND COMPARE THE PERFORMANCE AT GIVEN COMPLEXITIES 4HE COMPARISON SHOWS THAT THE USE OF SEPARABLE ESTIMATORS INCREASES THE PERFORMANCE SUBSTANTIALLY COMPARED TO NON SEPARABLE ESTIMATORS WITH THE SAME COMPLEXITY &OR THE SCENARIO INVESTIGATED IN THIS REPORT THE PERFORMANCE IS FURTHER IMPROVED BY APPLYING LOW RANK APPROXIMATIONS TO SEPARABLE ESTIMATORS



)NTRODUCTION

7IRELESS ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING /&$- IS CURRENTLY USED AND PRO POSED FOR SEVERAL BROADCASTING APPLICATIONS 4HE %UROPEAN STANDARD FOR DIGITAL AUDIO BROADCAST $!" ;= USES /&$- WITH DIdERENTIAL PHASE SHIFT KEYING $03+  4HIS IS SUITABLE FOR LOW BIT RATE SYSTEMS BUT WHEN HIGHER BIT RATES ARE REQUIRED MULTIAMPLI TUDE MODULATION IS MORE APPROPRIATE 0ROPOSALS FOR DIGITAL VIDEO BROADCASTING ; = HAVE INCLUDED MULTIAMPLITUDE MODULATION /&$- 4HESE SCHEMES CAN BE MADE DIdEREN TIAL WHICH OdERS THE ADVANTAGE OF AVOIDING CHANNEL ESTIMATION $IdERENTIAL AMPLITUDE AND PHASE SHIFT KEYING $!03+ ;= IS AN EXAMPLE OF THIS APPROACH )N $!03+ HOW EVER THE CONSTELLATION POINTS ARE NON UNIFORMLY DISTRIBUTED IN THE SIGNAL SPACE WHICH REDUCES PERFORMANCE 4HERE MAY ALSO BE METRIC DIbCULTIES CONCERNING DECODING #O HERENT MODULATION ON THE OTHER HAND GIVES BETTER PERFORMANCE BUT BECAUSE OF THE NECESSARY CHANNEL ESTIMATION IT REQUIRES MORE COMPLEXITY AT THE RECEIVER )T IS OF INTER EST THEREFORE TO INVESTIGATE THE PERFORMANCE OF COHERENT /&$- SYSTEMS USING CHANNEL ESTIMATORS WITH DIdERENT LEVELS OF COMPLEXITY )N THIS PAPER WE ANALYZE LOW COMPLEXITY COHERENT DEMODULATION RECEIVER SCHEMES SUITABLE FOR HIGH BIT RATE /&$- /NE WAY OF ESTIMATING THE CHANNEL IN A AT FADING ENVIRONMENT IS TO MULTIPLEX PILOTS KNOWN SYMBOLS INTO THE TRANSMITTED SIGNAL &ROM THESE SYMBOLS ALL CHANNEL ATTENU ATIONS ARE ESTIMATED WITH AN INTERPOLATION LTER 4HIS TECHNIQUE IS CALLED PILOT SYMBOL ASSISTED MODULATION 03!- AND WAS INTRODUCED FOR SINGLE CARRIER SYSTEMS BY -OHER AND ,ODGE ;= AND ANALYZED BY #AVERS ;= 3INCE EACH SUBCHANNEL IN /&$- IS AT FADING 03!- CAN BE GENERALIZED TO TWO DIMENSIONS WHERE PILOTS ARE TRANSMITTED IN CER TAIN POSITIONS IN THE TIME AND FREQUENCY GRID OF /&$- 4HE CHANNEL ESTIMATION IS THEN PERFORMED BY A TWO DIMENSIONAL INTERPOLATION (HER ;= PROPOSES TO USE NITE IMPULSE



0ART  ! #OMPARATIVE 3TUDY OF 0ILOT BASED #HANNEL %STIMATORS

RESPONSE &)2 LTERS FOR THIS AND TO SEPARATE THE USE OF TIME AND FREQUENCY CORRELATION (E ARGUES THAT THIS IS A GOOD TRADE Od BETWEEN COMPLEXITY AND PERFORMANCE 4HE SPACING OF PILOT SYMBOLS IN 03!- FOR SINGLE CARRIER SYSTEMS WAS INVESTIGATED IN ;= )T WAS FOUND THAT THE OPTIMUM SPACING WAS SOMEWHAT CLOSER THAN THE .YQUIST RATE IE THE INVERSE OF THE BANDWIDTH OF THE CHANNEL COVARIANCE FUNCTION 7E GENERALIZE THIS RESULT TO TWO DIMENSIONS FOR THE /&$- TIME FREQUENCY GRID 5SING A DENSE PILOT PATTERN MEANS THAT THE CHANNEL IS OVERSAMPLED IMPLYING THAT LOW RANK ESTIMATION METHODS ;= CAN WORK WELL 4HIS TYPE OF LOW COMPLEXITY ESTIMATION PROJECTS THE OBSERVATIONS ONTO A SUBSPACE OF SMALLER DIMENSION AND PERFORMS THE ESTIMATION IN THAT SUBSPACE "Y OVER SAMPLING THE CHANNEL IE PLACING THE PILOT SYMBOLS CLOSE TO EACH OTHER THE OBSERVATIONS ESSENTIALLY LIE IN A SUBSPACE AND LOW RANK ESTIMATION IS VERY EdECTIVE )N THIS REPORT WE PRESENT AND ANALYZE PILOT BASED /&$- CHANNEL ESTIMATORS THAT RELY ON BOTH TIME AND FREQUENCY CORRELATION OF THE FADING CHANNEL 4HE ESTIMATORS ARE LINEAR AND FEEDFORWARD IE NO DECISION DIRECTION OR FEEDBACK IS USED 7E DIVIDE THEM INTO TWO CLASSES  DIMENSIONAL  $ AND SEPARABLE ESTIMATORS 4HE LATTER USES  DIMENSIONAL  $ INTERPOLATION LTERS IN THE TIME AND FREQUENCY DIRECTIONS SEPARATELY )N EACH CLASS WE COMPARE A &)2 7IENER LTER ;= WITH A LOW RANK APPROXIMATION OF THE LINEAR MINIMUM MEAN SQUARED ERROR ,--3% ESTIMATOR ;= 4HE ESTIMATORS ARE COMPARED BOTH IN TERMS OF MEAN SQUARED ERROR -3% AND CODED BIT ERROR RATE "%2  4HE SYSTEM AND THE SCENARIO ARE INTRODUCED IN 3ECTION  4HE ESTIMATORS ARE DESCRIBED IN 3ECTION  AND THEIR PERFORMANCE IS PRESENTED IN 3ECTION  BOTH IN TERMS OF MEAN SQUARED ERROR AND CODED BIT ERROR RATE &INALLY IN 3ECTION  WE PRESENT CONCLUSIONS




3YSTEM DESCRIPTION
/&$- SYSTEM

)N THIS REPORT WE CONSIDER AN /&$- SYSTEM OPERATING IN A 2AYLEIGH FADING CHANNEL ENVIRONMENT 4HIS SYSTEM USES A CYCLIC PREX ;= WHICH IS A COPY OF THE LAST PART OF THE /&$- SYMBOL AND ACTS AS A GUARDSPACE BETWEEN CONSECUTIVE /&$- SYMBOLS (ENCE IF THE IMPULSE RESPONSE OF THE CHANNEL IS SHORTER THAN THE CYCLIC PREX INTER SYMBOL INTERFERENCE )3) IS AVOIDED &URTHERMORE IF THE CHANNEL IS ASSUMED CONSTANT DURING ONE /&$- SYMBOL INTER CARRIER INTERFERENCE )#) IS ALSO AVOIDED ;= )N &IGURE A A SCHEMATIC VIEW OF THE BASE BAND /&$- SYSTEM IS SHOWN 4HE MOD ULATION ON - SUBCARRIERS IS PERFORMED BY AN INVERSE DISCRETE &OURIER TRANSFORM )$&4 IN THE TRANSMITTER ;= 3IMILARLY DEMODULATION IS DONE WITH A DISCRETE &OURIER TRANSFORM $&4 IN THE RECEIVER 4HE EdECTIVE SYMBOL LENGTH IS 3  -3  WHERE 3 IS THE SAMPLING PERIOD OF THE SYSTEM !DDING A CYCLIC PREX #0 WITH A LENGTH OF 38  +3 MAKES THE TOTAL SYMBOL LENGTH 3
38  )F )3) AND )#) ARE ELIMINATED WE CAN DESCRIBE THE SYSTEM AS A SET OF PARALLEL 'AUSSIAN CHANNELS ;= SHOWN IN &IGURE B WITH CORRELATED CHANNEL ATTENUATIONS t u J Gt  &  J      - `   - 3



&IGURE  /&$- SYSTEM A "ASE BAND MODEL B PARALLEL SUBCHANNELS MODEL #0 AND #0 DENOTE THE INSERTION AND DELETION OF THE CYCLIC PREX RESPECTIVELY WHERE & a IS THE FREQUENCY RESPONSE OF THE CHANNEL F ~ DURING THE /&$- SYMBOL 4HE RECEIVED SIGNAL Xt ON SUBCHANNEL J CAN THUS BE DESCRIBED AS Xt  Gt Wt
Mt  

WHERE Wt IS THE TRANSMITTED DATA SYMBOL AND Mt THE CHANNEL NOISE AT SUBCARRIER J 4HE RELATION  HOLDS FOR EVERY /&$- SYMBOL THUS CREATING A TWO DIMENSIONAL GRID WITH FREQUENCY SUBCARRIERS ON ONE AXIS AND TIME /&$- SYMBOLS ON THE OTHER



3CENARIO

4HE CHANNEL ESTIMATION IS BASED ON PILOTS TRANSMITTED AT CERTAIN POSITIONS IN THE TIME FREQUENCY GRID OF THE /&$- SYSTEM 4HE CHANNEL ATTENUATIONS ARE ESTIMATED BY MEANS OF INTERPOLATION BETWEEN THESE PILOTS WHERE WE ASSUME THAT THE CHANNEL ESTIMATORS CAN USE ALL TRANSMITTED PILOTS 4HIS IS THE CASE IN EG BROADCASTING OR IN THE DOWNLINK OF A MULTIUSER SYSTEM )N BOTH THESE CASES THERE IS ONLY ONE PHYSICAL CHANNEL BETWEEN THE TRANSMITTER AND THE RECEIVER 4HUS CHANNEL ATTENUATIONS IN NEIGHBORING TIME FREQUENCY GRIDPOINTS ARE HIGHLY CORRELATED A FEATURE THAT CAN BE USED FOR CHANNEL ESTIMATION )N THE UPLINK OF A MULTIUSER SYSTEM ON THE OTHER HAND EACH USER HAS THEIR OWN PHYSICAL CHANNEL SO CHANNEL ATTENUATIONS STEMMING FROM DIdERENT MOBILE TRANSMITTERS MUST BE ASSUMED TO BE UNCORRELATED 4O ESTIMATE THE ATTENUATIONS FOR ONE USER ONLY PILOTS TRANSMITTED BY THAT USER CAN BE USED 4HUS THE UPLINK IS QUITE DIdERENT FROM OUR SCENARIO AND WILL NOT BE CONSIDERED HERE



0ART  ! #OMPARATIVE 3TUDY OF 0ILOT BASED #HANNEL %STIMATORS

4HE INVESTIGATED /&$- SYSTEM HAS A BANDWIDTH OF  -(Z AND IS OPERATING IN THE  '(Z FREQUENCY BAND 4HE NUMBER OF SUBCARRIERS IS -   WHICH MAKES THE EdECTIVE SYMBOL LENGTH  xS 4HE ENVIRONMENT IS A MACROCELL WHICH IS ASSUMED TO HAVE A MAXIMUM DELAY SPREAD OF  xS AND A MAXIMUM $OPPLER FREQUENCY OF  (Z 4HUS THE MAXIMUM $OPPLER FREQUENCY RELATIVE TO THE INTER CARRIER SPACING IS E5&MAX    WHICH CORRESPONDS TO A VEHICLE SPEED OF  KMH 4HE POWER DELAY PROLE IS EXPONENTIALLY DECAYING WITH ROOT MEAN SQUARE 2-3 WIDTH ~RMS   xS 4O ELIMINATE )3) WE USE A GUARD SPACE OF  xS WHICH CORRESPONDS TO +   SAMPLES 4HE LENGTH OF THE /&$SYMBOL IS 
   xS WHICH MAKES THE RELATIVE SIZE OF THE GUARD SPACE   4HE CORRESPONDING 2-1 LOSS IS  D" /UR SYSTEM MODEL ASSUMES THAT THE CHANNEL IS CONSTANT DURING AN /&$- SYMBOL )N REALITY )#) OCCURS DUE TO CHANNEL FADING DURING THE TRANSMISSION OF AN /&$- SYMBOL ; = (OWEVER WITH A MAXIMUM RELATIVE $OPPLER FREQUENCY OF   THE SIGNAL TO )#) RATIO IS  D" ;= 4HIS IS NEGLIGIBLE IN THE 2-1 RANGES WE ARE LOOKING AT AND CONSEQUENTLY WE IGNORE THE )#) AND USE THE MODEL IN   )NTERLEAVING IS PERFORMED SEPARATELY IN FREQUENCY AND TIME OVER A FRAME CONSISTING OF  /&$- SYMBOLS 4HIS CORRESPONDS TO A MAXIMUM DELAY OF  MS &IRST INTER LEAVING IN FREQUENCY IS DONE BY PLACING CONSECUTIVE DATA SYMBOLS  SUBCARRIERS APART )NTERLEAVING IS THEN PERFORMED IN TIME WITH A UNIQUE PATTERN FOR EACH SUBCARRIER 4HESE PATTERNS ARE RANDOM PERMUTATIONS AND CHANGED EVERY FRAME "Y HAVING DIdERENT INTER LEAVING PATTERNS ON ALL SUBCARRIERS CHANNEL ATTENUATIONS ARE INTERLEAVED IN BOTH TIME AND FREQUENCY 4HIS PRODUCES AN ALMOST PERFECT INTERLEAVING WITH NO SIGNICANT PERFORMANCE LOSS &OR ERROR CORRECTION A RATE  CONVOLUTIONAL CODE WITH THE OCTAL POLYNOMIALS   IS USED IE THE CODE POLYNOMIALS ARE ;= F F
 

# #

 
#
# 
#
# !  
#
# 
#
# ! 

 

4HE CONSTRAINT LENGTH IS M6   AND A TAIL OF M6

   ZEROS IS APPENDED TO CLEAR THE ENCODERS MEMORY 4HE RECEIVER USES A SOFT DECISION 6ITERBI DECODER WITH A TRUNCATED MEMORY LENGTH OF M6   BITS 4HE BITS ARE MODULATED USING "03+ IN EACH DIMENSION AND THE INPHASE AND QUADRATURE PARTS ARE CONCATENATED TO FORM 103+ SYMBOLS 4HIS MAKES THE DATA RATE OF THE SYSTEM  -BITS



#HANNEL MODEL

)N OUR ANALYSIS WE USE THE WIDE SENSE STATIONARY UNCORRELATED SCATTERING 73353 CHANNEL MODEL INTRODUCED IN ;= "Y CONSIDERING THE CHANNEL TO BE CONSTANT OVER ONE /&$SYMBOL THE INSTANTANEOUS FREQUENCY RESPONSE OF THE , `PATH CHANNEL AT TIME S IS
D  8 s tw {75w {i~w & E S  P D  , w(



WHERE tw IS THE PHASE %5w THE $OPPLER FREQUENCY AND ~w THE DELAY OF THE Mq PATH !LL THESE PARAMETERS ARE INDEPENDENT RANDOM VARIABLES 4O OBTAIN 2AYLEIGH FADING WITH THE

 *AKES SPECTRUM ;= AND AN EXPONENTIALLY DECAYING POWER DELAY PROLE WITH 2-3 VALUE ~RMS  WE CHOOSE THE PROBABILITY DENSITY FUNCTIONS AS ;= Ot t  { O75 %5  O~ ~ 
{75&MAX  75 (75&M AX h`~ (~RMS  ~RM S  h`Qfy (~RM S
Q

 v t  {   J%5 J  %5&MAX  v ~ v 3fy



4HE RANDOM VARIABLES %5 AND ~ CAN EASILY BE OBTAINED FROM A UNIFORMLY DISTRIBUTED RANDOM GENERATOR WITH OUTPUTS  : < BY USING THE INVERSES OF THE DESIRED CUMULATIVE DISTRIBUTION FUNCTIONS ;=



0ILOT PATTERN

"Y USING A TWO DIMENSIONAL GENERALIZATION ;= OF PILOT SYMBOL ASSISTED MODULATION ;= KNOWN SYMBOLS PILOTS ARE TRANSMITTED ON CERTAIN POSITIONS IN THE TIME FREQUENCY GRID 4HE NUMBER OF PILOTS TO USE IS A TRADE Od BETWEEN DATA RATE AND CHANNEL ESTIMATION PERFORMANCE (OWEVER BY VIEWING THE CHANNEL ESTIMATION IN THE TIME FREQUENCY GRID AS A TWO DIMENSIONAL INTERPOLATION FUNDAMENTAL LIMITS ON THE DENSITY OF PILOTS CAN BE DE RIVED 4HE SCATTERED PILOT SYMBOLS CAN BE SEEN AS NOISY SAMPLES OF THE TWO DIMENSIONAL STOCHASTIC SIGNAL & E  S  4HESE SAMPLES HAVE TO BE PLACED CLOSE ENOUGH TO FULL THE SAM PLING THEOREM AND AVOID ALIASING .OTE THAT THE EdECTIVE 3.2 IS LOWERED BY USING MANY PILOTS SINCE A SMALLER PART OF THE TRANSMITTED POWER IS USED FOR DATA SYMBOLS 3INCE & E  S IN  IS THE &OURIER TRANSFORM OF THE CHANNEL IMPULSE RESPONSE AT TIME S WHICH IS ASSUMED TO BE CONSTANT FOR ONE /&$- SYMBOL  THE AUTO COVARIANCE FUNCTION OF & E  S IS THE SPACED FREQUENCY SPACED TIME CORRELATION FUNCTION 4 OF THE CHANNEL ;= 188 aE aS  $ F& E  S & E ` aE  S ` aS G  4 aE aS   4HE BANDWIDTH OF THIS FUNCTION IS !g THE $OPPLER SPREAD IN THE aE DIRECTION AND ~MAX THE MULTIPATH SPREAD IN THE aS DIRECTION ;= &OR THE ANALYZED /&$- SYSTEM WE HAVE E5&MAX  !g  %5&MAX  - 3 ~MAX  +3   WHERE E5&MAX IS THE MAXIMUM $OPPLER FREQUENCY RELATIVE TO THE INTER CARRIER SPACING )F WE ASSUME THAT PILOTS ARE PLACED -i SUBCARRIERS APART IN EVERY - /&$- SYMBOLS WE HAVE t u -i Gt&u  & J a  K a - -
+ 3  - 3 SINCE THE INTER CARRIER SPACING IS - 3 AND THE DURATION OF AN /&$- SYMBOL IS -
+ 3  4O FULLL THE SAMPLING THEOREM ;= WE NEED -i  -  +  

 ` a  C  
E E5&MAX



0ART  ! #OMPARATIVE 3TUDY OF 0ILOT BASED #HANNEL %STIMATORS

&IGURE  4HE PILOT PATTERN USED IN THE SYSTEM 0ILOT SYMBOLS ARE MARKED WITH GREY SQUARES )N THE ANALYZED SYSTEM WE HAVE -   +   AND E5&MAX   WHICH GIVES -i       ` a -       
   

)N ;= WHERE 03!- FOR SINGLE CARRIER SYSTEMS IS ANALYZED IT IS SHOWN THAT THE "%2 CAN BE LOWERED BY PLACING THE PILOT SYMBOLS CLOSER THAN THAT SPECIED BY THE SAMPLING THEOREM .OTE THAT THERE EXISTS A PILOT SPACING WHICH OPTIMIZES THE TRADE Od BETWEEN IMPROVED CHANNEL ESTIMATION AND REDUCED 3.2 ON THE DATA SYMBOLS "Y VARYING THE PILOT SPACINGS -i AND -  IT WAS FOUND THAT -i   AND -   WAS CLOSE TO OPTIMAL IN TERMS OF "%2 4HE USED PILOT PATTERN IS SHOWN IN &IGURE  4HIS MEANS THAT  { OF THE BANDWIDTH AND THE TRANSMITTED POWER IS USED FOR PILOTS .OTE THAT THE CHANNEL IS OVERSAMPLED WHICH MEANS THAT LOW RANK ESTIMATORS CAN BE VERY EdECTIVE ;= )N OUR STUDY THE PILOT SYMBOLS HAVE THE SAME AVERAGE POWER AS THE DATA SYMBOLS (OWEVER A TECHNIQUE CALLED BOOSTED PILOTS CAN ALSO BE USED WHICH IS PROPOSED FOR $6" ;= 4HIS MEANS THAT THE PILOT SYMBOLS ARE TRANSMITTED WITH A HIGHER AVERAGE POWER THAN THE DATA SYMBOLS 4HE AVERAGE 3.2 ON THE DATA SYMBOLS IS REDUCED BUT THE CHANNEL ESTIMATES ARE BETTER SINCE THE 3.2 AT THE PILOT SYMBOLS IS INCREASED 4HUS BY CHOOSING A SUITABLE POWER LEVEL FOR THE PILOT SYMBOLS THE BIT ERROR RATE CAN BE DECREASED



%STIMATORS

)N /&$- SYSTEMS THE OPTIMAL LINEAR ESTIMATOR IN THE MEAN SQUARED ERROR SENSE IS A  $ BOTH TIME AND FREQUENCY LTER (OWEVER THE COMPLEXITY OF THIS ESTIMATOR IS USUALLY TOO LARGE FOR IT TO BE OF PRACTICAL USE ! NUMBER OF SUBOPTIMAL LOW COMPLEXITY CHANNEL ESTIMATORS HAVE BEEN SUGGESTED IN THE LITERATURE SEE EG ; = 7E WILL INVESTIGATE

 TWO CLASSES OF ESTIMATORS  DIMENSIONAL AND SEPARABLE 4HE USE OF SEPARABLE LTERS IS A COMMON METHOD TO REDUCE COMPLEXITY IN MULTIDIMENSIONAL SIGNAL PROCESSING ;= &OR BOTH SEPARABLE AND NON SEPARABLE ESTIMATORS WE LOOK AT &)2 7IENER LTERS AND LOW RANK APPROXIMATIONS OF ,--3% ESTIMATORS 7E COMPARE ALL ESTIMATORS PERFORMANCES FOR TWO LEVELS OF COMPLEXITY 3INCE THEY ARE ALL LINEAR ESTIMATORS A REASONABLE MEASURE OF COMPLEXITY IS THE AVERAGE NUMBER OF MULTIPLICATIONS PER ESTIMATED ATTENUATION )N THE SEQUEL WE USE THE FOLLOWING NOTATION 4HE BACKROTATED OR LEASTSQUARES ESTIMATED CHANNEL ATTENUATIONS AT PILOT POSITIONS ARE DENOTED BY Ot&u  Xt&u  Wt&u 

WHERE Xt&u IS THE RECEIVED SIGNAL AT SUBCARRIER J IN /&$- SYMBOL K AND Wt&u IS THE CORRE SPONDING TRANSMITTED PILOT SYMBOL 4HE NAL ESTIMATE OF THE CHANNEL ATTENUATIONS Gt&u ARE LINEAR COMBINATIONS OF THE Ot&u S WHERE THE COEbCIENTS ARE CHOSEN ACCORDING TO EACH ESTIMATORS STRUCTURE "Y ARRANGING THE AVAILABLE ,3 ESTIMATES AT PILOT POSITIONS IN A VECTOR O AND THE CHANNEL ATTENUATIONS TO BE ESTIMATED IN A VECTOR G THE LINEAR MINIMUM MEAN SQUARED ERROR ESTIMATOR OF G IS ;= B  1GO 1 O G OO 

WHERE 1GO IS THE CROSS COVARIANCE MATRIX BETWEEN G AND O AND 1OO IS THE AUTO COVARIANCE MATRIX OF O $EPENDING ON THE NUMBER OF PILOTS USED AND THEIR RELATIVE LOCATIONS THE SIZE OF O AND THE CORRESPONDING AUTO COVARIANCE MATRIX 1OO WILL CHANGE !LSO DEPENDING ON THE NUMBER OF ESTIMATED ATTENUATIONS THE SIZE OF G WILL CHANGE &URTHERMORE 1GO DEPENDS ON THE RELATIVE POSITIONS BETWEEN ESTIMATED ATTENUATIONS AND THE USED PILOT POSITIONS "ELOW WE ADDRESS SEVERAL CHOICES ON USED PILOTS AND ESTIMATED ATTENUATIONS



 $ LTERS

4HE  $ 7IENER LTER IS OPTIMAL IN TERMS OF -3% IF COMPLEXITY IS NOT CONSIDERED (OW EVER FOR A XED COMPLEXITY THE NUMBER OF LTER TAPS THAT CAN BE USED IS QUITE SMALL 7E USE THIS ESTIMATOR AS A REFERENCE AND INVESTIGATE A REDUCED COMPLEXITY ESTIMATOR WHICH IS DERIVED USING THE THEORY OF OPTIMAL RANK REDUCTION ;=   $ ESTIMATOR

)F THE ALLOWED COMPLEXITY IS * MULTIPLICATIONS PER ATTENUATION THE TWO DIMENSIONAL LTER USES THE * PILOTS CLOSEST TO THE ESTIMATED ATTENUATION )N &IGURE  WE DISPLAY AN EXAMPLE OF THE SEVEN PILOT POSITIONS USED *   TO ESTIMATE ONE CHANNEL ATTENUATION &OR EVERY ESTIMATED CHANNEL ATTENUATION THERE IS A SET OF * ASSOCIATED PILOTS /PTIMAL WEIGHTS ARE CALCULATED ACCORDING TO   &OR THE ESTIMATOR WITH THE LOWER COMPLEXITY WE WILL USE THE  CLOSEST PILOTS AND FOR THE HIGHER COMPLEXITY THE  CLOSEST



0ART  ! #OMPARATIVE 3TUDY OF 0ILOT BASED #HANNEL %STIMATORS

&IGURE  4WO DIMENSIONAL &)2 7IENER LTER 4HE ESTIMATED TONE b IS A LINEAR COMBINATION OF THE  PILOT TONES d 

&IGURE  3TRUCTURE OF THE LOWRANK  DIMENSIONAL ESTIMATOR 4HE TONES TO ESTIMATE ARE MARKED WITH b AND THE PILOTS USED ARE MARKED WITH d   ,OWRANK  $ ESTIMATOR

4HE LOWRANK  $ ESTIMATOR IS IN A SENSE AN APPROXIMATION OF THE OPTIMAL  $ ESTIMATOR IN THE PREVIOUS SECTION 4HE LOW COMPLEXITY IS ACHIEVED BY A GENERALIZATION OF THE IDEAS IN ;= 4O ALLOW A LOW RANK APPROXIMATION *q ATTENUATIONS G ARE ESTIMATED SIMULTANEOUSLY USING THE *y CLOSEST PILOTS O )F THE ATTENUATIONS TO BE ESTIMATED AND THE PILOTS USED ARE CHOSEN PROPERLY THE ESTIMATOR CAN BE WELL APPROXIMATED BY A LOW RANK ESTIMATOR THEREBY REDUCING THE COMPLEXITY CONSIDERABLY WHILE MAINTAINING MOST OF THE PERFORMANCE .OTE THAT THE ESTIMATED ATTENUATIONS CAN BE CHOSEN ARBITRARILY IN THE TIME FREQUENCY GRID )N &IGURE  AN EXAMPLE IS GIVEN FOR THE LOCATION OF ESTIMATED ATTENUATIONS *q   AND THE USED PILOT SYMBOLS *y   

 4HE ESTIMATOR BECOMES B  & O  G 

WHERE & IS A LOW RANK 7IENER LTER ;= &ROM THE SINGULAR VALUE DECOMPOSITION 36$

(  4f59  1GO 1OO ( &  4f 59 1OO



WHERE 4 AND 5 ARE UNITARY MATRICES AND f IS A DIAGONAL MATRIX ;= THE LOW RANK 7IENER LTER IS DETERMINED BY ;= 

WHERE f IS A *q b *y DIAGONAL MATRIX CONTAINING THE Q LARGEST SINGULAR VALUES 4HE COMPLEXITY OF THIS ESTIMATOR IS FOUND IN !PPENDIX ! TO BE u t *y  Q 
*q MULTIPLICATIONS PER ESTIMATED ATTENUATION WHERE Q IS THE RANK OF THE ESTIMATOR NUMBER OF SINGULAR VALUES USED  *y THE NUMBER OF PILOTS USED AND *q THE NUMBER OF ATTENUATIONS TO BE ESTIMATED &OR THE LOWER COMPLEXITY ESTIMATOR WE CHOSE *y   PILOTS  IN THE TIME DIRECTION AND  IN THE FREQUENCY DIRECTION AND *q   ATTENUATIONS TO ESTIMATE  IN THE TIME DIRECTION AND  IN THE FREQUENCY DIRECTION  4HE LATTER WERE PLACED IN THE MIDDLE OF THE FORMER IN ORDER TO EXPLOIT AS MUCH CORRELATION AS POSSIBLE 7ITH A RANK OF Q   THE NUMBER OF MULTIPLICATIONS PER ATTENUATION IS ACCORDING TO %Q   t u  "LOW   
    &OR THE HIGHER COMPLEXITY WE CHOSE  PILOTS  IN THE TIME DIRECTION AND  IN THE FREQUENCY DIRECTION AND  ATTENUATIONS TO ESTIMATE  IN THE TIME DIRECTION AND  IN THE FREQUENCY DIRECTION  4HE RANK USED WAS  GIVING t u  "HIGH   
    MULTIPLICATIONS PER ATTENUATION



3EPARABLE LTERS

3INCE  $ LTERS IN GENERAL TEND TO HAVE A LARGE COMPUTATIONAL COMPLEXITY THE OUTER PRODUCT OF TWO  $ LTERS CAN GIVE A GOOD TRADE Od BETWEEN PERFORMANCE AND COMPLEX ITY 4HIS IS A STANDARD TECHNIQUE IN MULTIDIMENSIONAL SIGNAL PROCESSING ;= AND IT HAS ALSO BEEN PROPOSED FOR PILOT BASED CHANNEL ESTIMATION IN /&$- SYSTEMS ;= )N PILOT BASED ESTIMATION SCHEMES THE MAJOR ADVANTAGE IS THE LOW NUMBER OF MULTIPLICATIONS PER USED PILOT 4HIS ALLOWS THE ESTIMATOR TO BE BASED ON MORE PILOTS THUS IMPROVING THE PERFORMANCE



0ART  ! #OMPARATIVE 3TUDY OF 0ILOT BASED #HANNEL %STIMATORS

&IGURE  3EPARABLE LTER BASED ON ONE DIMENSIONAL LTERS IN FREQUENCY  AND TIME  DIRECTIONS &ILTERING IN THE FREQUENCY DIRECTION  IS PERFORMED RST "ASED ON THE PILOT PATTERN CHOSEN THE GENERAL CONCEPT USED IN THIS REPORT IS SHOWN IN &IGURE  WHERE A  $ LTER IS APPLIED IN THE FREQUENCY DIRECTION 4HEREAFTER A  $ LTER IS APPLIED IN THE TIME DIRECTION TO COMPLETE THE INTERPOLATION TO ALL POINTS IN THE GRID 7E INVESTIGATE BOTH AN ESTIMATOR BASED ON THE PROPOSAL IN ;= AND A VARIANT THEREOF WHICH ALLOWS THE USE OF MORE PILOTS BY A LOW RANK APPROXIMATION IN THE FREQUENCY DIREC TION  3EPARABLE &)2 LTERS

4HE USE OF SEPARATE  $ &)2 LTERS IN THE TIME AND FREQUENCY DIRECTIONS HAS BEEN PRO POSED BY (HER IN ;= &IRST ALL ATTENUATIONS IN /&$- SYMBOLS CONTAINING PILOTS ARE ESTIMATED 4HIS IS DONE WITH A &)2 7IENER LTER OF LENGTH *i  4WO ATTENUATIONS ON DIdERENT POSITIONS RELATIVE TO THE PILOTS WILL NEED DIdERENT LTERS SO THERE WILL BE -i DIdERENT LTERS FOR THIS ESTIMATION .OTE THAT THESE LTERS ARE NON CAUSAL IN THE SENSE THAT THEY WILL USE PILOTS ON BOTH SIDES OF THE ESTIMATED ATTENUATION IN ORDER TO EXPLOIT THE CLOSEST PILOTS !FTER THIS PROCEDURE THERE WILL BE ESTIMATES OF ALL ATTENUATIONS IN EVERY -q /&$- SYMBOL &)2 7IENER LTERS OF LENGTH * ARE NOW USED IN THE TIME DIRECTION TO OBTAIN ESTIMATES OF ALL ATTENUATIONS (ERE THERE WILL BE - `  DIdERENT LTERS DEPENDING ON WHICH ATTENUATION IS ESTIMATED 4HESE LTERS CAN BE NONCAUSAL WHICH WILL INTRODUCE A DELAY IN THE SYSTEM )F THIS DELAY CANNOT BE ACCEPTED CAUSAL LTERS MUST BE USED 4HE TOTAL NUMBER OF MULTIPLICATIONS PER ESTIMATED ATTENUATION IS *i
* - 

SINCE THE FREQUENCY DIRECTION LTER HAS TO BE APPLIED IN ONLY ONE OUT OF EVERY - /&$SYMBOLS 4HIS FACT CAN BE USED IN THE DESIGN OF THE LTERS SINCE THE FREQUENCY DIRECTION

 LTER TAPS ARE CHEAPER IN TERMS OF COMPLEXITY "Y REDUCING THE NUMBER OF TAPS IN THE TIME DIRECTION LTER BY ONE -i TAPS CAN BE ADDED TO THE FREQUENCY DIRECTION LTER WITHOUT CHANGING THE TOTAL COMPLEXITY &OR THE LOWER COMPLEXITY WE USED  TAPS IN THE FREQUENCY LTER AND  TAPS IN THE TIME LTER 4HIS RESULTS IN "LOW  
    

MULTIPLICATIONS PER ATTENUATION &OR THE HIGHER COMPLEXITY  AND  TAPS WERE USED FOR THE FREQUENCY AND TIME LTERS RESPECTIVELY 4HIS MEANS A COMPLEXITY OF "HIGH  
    

MULTIPLICATIONS PER ATTENUATION  ,OW RANK SEPARABLE LTER

"Y USING OBSERVATIONS FROM ;= WHERE LOW RANK APPROXIMATIONS OF CHANNEL ESTIMATORS ARE PRESENTED WE REPLACE ONE OF THE &)2 LTERS IN THE ESTIMATOR PROPOSED BY (HER )NSTEAD OF &)2 7IENER LTERS IN BOTH DIRECTIONS A LOW RANK APPROXIMATION OF THE FREQUENCY DIRECTION ,--3% ESTIMATOR IS USED IN COMBINATION WITH THE TIME DIRECTION &)2 LTER (ENCE A FREQUENCY DIRECTION LTERING IS PERFORMED FOR EACH /&$- SYMBOL CONTAINING PILOTS !N OBVIOUS WAY OF DOING THIS LTERING IS TO ESTIMATE ALL ATTENUATIONS IN AN /&$- SYMBOL USING ALL PILOTS (OWEVER WHEN USING ALL PILOTS THE COMPLEXITY REDUCTION IS NOT SO LARGE THAT IT CAN COMPETE WITH A SHORT &)2 7IENER LTER 4HE NUMBER OF MULTIPLICATIONS PER ATTENUATION IS SEE !PPENDIX ! u t *y Q 
  *q WHERE Q IS RANK USED *q AND *y ARE THE NUMBER OF ATTENUATIONS TO ESTIMATE AND NUMBER OF PILOTS USED RESPECTIVELY 3INCE PILOTS FAR AWAY FROM THE ESTIMATED ATTENUATIONS ARE WEAKLY CORRELATED THEY DO NOT CONTRIBUTE MUCH TO THE ESTIMATE "Y EXCLUDING THEM THE COMPLEXITY GOES DOWN WHILE THE PERFORMANCE IS ALMOST THE SAME (ENCE THE /&$SYMBOL IS PARTITIONED INTO A NUMBER OF SUB SYMBOLS WHERE THE ATTENUATIONS ARE ESTIMATED USING ONLY THE *y PILOTS CLOSEST TO THE SUB SYMBOL CONSISTING OF *q SUBCARRIERS ;= )N &IGURE  AN EXAMPLE IS SHOWN FOR *q   CHANNEL ATTENUATIONS AND *y   PILOTS ,OW RANK APPROXIMATIONS CAN BE DONE FOR THE TIME DIRECTION LTERING AS WELL BUT IN THIS REPORT WE HAVE CHOSEN TO USE AN &)2 LTER INSTEAD 7E CHOSE FOR THE LOWER COMPLEXITY TO ESTIMATE  ATTENUATIONS IN THE FREQUENCY DIRECTION USING  PILOTS AND A RANK OF  4HIS RESULTS IN  
   MULTIPLICATIONS PER ATTENUATION IN THE FREQUENCY DIRECTION #OMBINED WITH A  TAP TIME LTER THIS GIVES A TOTAL OF "LOW  
    



0ART  ! #OMPARATIVE 3TUDY OF 0ILOT BASED #HANNEL %STIMATORS

&IGURE  3TRUCTURE OF THE LOWRANK ESTIMATOR IN THE FREQUENCY DIRECTION 4HE *q   ATTENUATIONS TO ESTIMATE ARE MARKED WITH b  AND THE *y   PILOT SYMBOLS USED ARE MARKED WITH d  %STIMATOR  $ 3TRUCTURE 5SES THE *y CLOSEST PILOTS MULTATT r *y s
Bi E
* E

,OW RANK  $ %STIMATES *q ATTENUATIONS USING *y PILOTS AND A RANK OF Q 3EPARABLE 3EPARABLE &)2 LTER WITH *i FREQUENCY AND * TIME TAPS ,OW RANK SEPARABLE %STIMATES *q ATTENUATIONS USING *y PILOTS AND A RANK OF Q FREQUENCY AND * TAPS &)2 LTER TIME

By Q 
B q

PILOTS *y *y *i * *y *

s By 
B
* q

4ABLE  4HE FOUR CHANNEL ESTIMATORS INVESTIGATED IN THIS REPORT MULTIPLICATIONS PER ATTENUATION &OR THE HIGHER COMPLEXITY WE USED  PILOTS TO ESTIMATE  ATTENUATIONS WITH A RANK OF  4OGETHER WITH A TIME LTER WITH  TAPS THIS GIVES ` a  
! #
     "HIGH   MULTIPLICATIONS PER ATTENUATION 4HE FOUR INVESTIGATED ESTIMATORS ARE SUMMARIZED IN 4ABLE 



#OMPLEXITY AND USED PILOTS

3INCE ALL ESTIMATORS ARE LINEAR THE CHANNEL ESTIMATES ARE LINEAR COMBINATIONS OF A NUMBER OF PILOTS 4HE NUMBER OF PILOTS USED DEPENDS ON THE COMPLEXITY AND THE TYPE OF ESTIMATOR 4HE AVERAGE NUMBER OF MULTIPLICATIONS PER ATTENUATION AND THE NUMBER OF PILOTS THE ESTIMATORS ARE BASED ON ARE SHOWN IN 4ABLE  4HE ESTIMATORS USE THE PARAMETERS DESCRIBED IN THE PREVIOUS SECTION !S CAN BE SEEN IN THE TABLE FOR A XED COMPLEXITY THE NUMBER OF PILOTS USED IN THE ESTIMATES CAN BE INCREASED BY POSING RESTRICTIONS ON THE ESTIMATORS SUCH AS SEPARABILITY AND LOW RANK (OWEVER BECAUSE A LARGE NUMBER OF PILOTS ARE USED MANY WILL BE ONLY WEAKLY CORRELATED WITH THE ESTIMATED CHANNEL ATTENUATIONS AND WILL NOT THEREFORE CONTRIBUTE MUCH 4HUS CONSIDERING CORRELATION MISMATCH IN THE DESIGN THE ESTIMATION MAY ACTUALLY BE DEGRADED BY INCREASING THE NUMBER OF USED PILOTS 4HE MAIN ADVANTAGE OF USING MANY PILOTS IS THAT THE IMPACT OF THE CHANNEL NOISE IS REDUCED BY A LARGE AVERAGING





$ESIGN ASPECTS

)N ORDER TO MAKE THE CHANNEL ESTIMATORS ATTRACTIVE TO IMPLEMENT WE ASSUME THAT THEY ARE XED IE DESIGNED FOR BOTH A XED CHANNEL CORRELATION AND A XED 3.2 4HE FREQUENCY CORRELATION IS DETERMINED BY THE POWER DELAY PROLE OF THE CHANNEL ;= AND THE TIME CORRELATION IS DETERMINED BY THE $OPPLER FREQUENCY ;= .EITHER THE POWER DELAY PROLE NOR THE $OPPLER FREQUENCY ARE KNOWN BY THE RECEIVER 5SING OUR CHANNEL MODEL WE DESIGN THE ESTIMATORS FOR A MAXIMUM RELATIVE $OPPLER FREQUENCY OF  AND A UNIFORM POWER DELAY PROLE OVER THE LENGTH OF THE CYCLIC PREX 4HE USE OF THESE WORST CASE PARAMETERS FOLLOWS THE RECOMMENDATIONS IN ;= WHERE PILOT SYMBOL ASSISTED MODULATION IS ANALYZED &OR THE DETERMINATION OF THE CORRELATION MATRICES SEE !PPENDIX " #ONTRARY TO THE WORST CASE RECOMMENDATIONS FOR XED DESIGN CORRELATIONS THE XED DESIGN 3.2 SHOULD BE CHOSEN TO A BEST CASE 4HIS IMPLIES A CLOSE TO OPTIMAL PERFORMANCE FOR 3.2S BELOW THE DESIGN 3.2 WHERE THE EdECTS OF 3.2 MISMATCH ARE SMALL COMPARED TO THE OVERALL NOISE LEVEL 7E HAVE CHOSEN THE DESIGN 3.2 TO  D" 7E EVALUATE THE ESTIMATORS UNDER MISMATCH IE THEY ARE DESIGNED FOR THE WRONG CHANNEL CORRELATION AND 3.2 0ARAMETERS FOR THE DESIGN AND THE TRUE VALUES OF THE CHANNEL STATISTICS ARE SHOWN BELOW IN 4ABLE  .OTE THAT THERE IS NO MISMATCH IN $OPPLER FREQUENCY 'IVEN THE DESIGN FOR  RELATIVE $OPPLER FREQUENCY THE PERFORMANCE IS APPROXIMATELY THE SAME FOR E5&MAX v  ;= 4HE 3.2 IS DENED AS THE TRANSMITTED ENERGY PER DATA BIT OVER THE CHANNEL NOISE VARIANCE h i h i $ JGt J $ JWt J  h i 2-1   a   A $ JMt J WHERE A DENOTES THE NUMBER OF INFORMATION BITSSYMBOL )N OUR CASE WE HAVE A  



0ERFORMANCE EVALUATION

7E EVALUATE THE PERFORMANCE OF THE FOUR INVESTIGATED ESTIMATORS BOTH IN TERMS OF MEAN SQUARED ERROR -3% AND CODED BIT ERROR RATE "%2  4HE -3% IS THEORETICALLY CALCULATED WHILE THE "%2 IS SIMULATED 4O SIMPLIFY THE -3% CALCULATIONS WE IGNORE EDGE EdECTS IN THE TIME FREQUENCY GRID AND ASSUME THAT IT IS OF INNITE EXTENT &OR A SYSTEM WHERE THE NUMBER OF SUBCARRIERS IS MUCH LARGER THAN THE LENGTH OF THE ESTIMATOR THESE EdECTS CAN BE IGNORED %STIMATOR  $ ,OW RANK  $ 3EPARABLE &)2 ,OW RANK SEPARABLE ,OW COMPLEXITY MULTATT PILOTS         (IGH COMPLEXITY MULTATT PILOTS        

4ABLE  !VERAGE NUMBER OF MULTIPLICATIONS PER TONE AND THE NUMBER OF PILOTS THE ESTIMATORS ARE BASED ON

 0ARAMETER .O OF SUBCARRIERS 4IME DISPERSION

0ART  ! #OMPARATIVE 3TUDY OF 0ILOT BASED #HANNEL %STIMATORS 4RUE   | xS ~(%x "D   ~  3fy 0OWER DELAY PROLE  OTHERWISE -AX REL $OPPLER FREQUENCY  3.2 6ARYING $ESIGN   | xS 3fy   ~  3fy  OTHERWISE   D"

4ABLE  $ESIGN AND TRUE VALUES OF SYSTEM PARAMETERS 4HE CONSTANT " IS A NORMALIZA TION FACTOR



-EAN SQUARED ERROR



3INCE ALL ESTIMATORS ARE LINEAR THEY CAN BE EXPRESSED AS B  &O G

BY COLLECTING ALL USED PILOTS IN A VECTOR O AND CALCULATING THE CORRESPONDING ESTIMATOR B CAN BE EXPRESSED AS ;= MATRIX & 4HE COVARIANCE MATRIX OF THE ERROR D  G ` G
9 1DD  1GG ` 1GO &9 ` &19 GO
&1OO & 



&OR ALL ESTIMATORS THERE WILL BE DIdERENT MEAN SQUARED ERRORS DEPENDING ON THE ESTIMATED ATTENUATION )N ORDER TO COMPARE THE ESTIMATORS WE ONLY LOOK AT THE AVERAGE ERROR OVER ALL ATTENUATIONS IE
Bq  8  -3%  1DD M M  TR 1DD  *q w( *q



WHERE *q IS THE NUMBER OF ESTIMATED ATTENUATIONS AND TR DENOTES THE TRACE OF A MATRIX ;= )N &IGURE  THE -3% FOR THE ESTIMATORS WITH THE LOWER COMPLEXITY LEVEL ARE SHOWN AS A FUNCTION OF 3.2 )T SHOULD BE NOTED THAT THE NON SEPARABLE ESTIMATORS HAVE AN ERROR OOR THAT IS ALREADY VISIBLE AT LOW 3.2 &OR THE SEPARABLE ESTIMATORS THE ERROR CURVES LEVEL OUT FOR VERY HIGH 3.2S BUT THIS ERROR OOR IS HARDLY NOTICEABLE IN THE GURE "ECAUSE THERE WILL ALWAYS BE AN INTERPOLATION ERROR EVEN IN THE NOISELESS CASE ALL THE ESTIMATORS HAVE AN ERROR OOR 4HIS IS DUE TO NITE LTER LENGTHS &OR 2-1   D" THE LOW RANK SEPARABLE ESTIMATOR IS  D" BETTER THAN SEPARABLE &)2 LTERS 4HE -3% CURVES FOR THE HIGHER COMPLEXITY ARE SHOWN IN &IGURE  4HE ERROR OORS HAVE NOW BEEN LOWERED AND ARE ONLY NOTICEABLE FOR THE  $ ESTIMATOR .OTE ALSO THAT THE MUTUAL ORDERING IS THE SAME AS FOR THE LOWER COMPLEXITY IE THE LOW RANK SEPARABLE ESTIMATOR IS THE BEST AND THE  $ ESTIMATOR IS THE WORST 4HE DIdERENCE BETWEEN THE LOW RANK SEPARABLE ESTIMATOR AND SEPARABLE &)2 LTERS IS NOW ABOUT  D" )N BOTH &IGURES  AND  IT CAN BE SEEN THAT SEPARABLE ESTIMATORS PERFORM BETTER THAN NON SEPARABLE 4HIS WAS NOTED IN ;= WHERE IT WAS ARGUED THAT SEPARABLE ESTIMATORS PROVIDE A GOOD TRADE Od BETWEEN COMPLEXITY AND PERFORMANCE



&IGURE  -EAN SQUARED ERROR RELATIVE TO CHANNEL POWER FOR THE FOUR ANALYSED ESTIMA TORS WITH THE LOWER COMPLEXITY  MULTIPLICATIONS PER TONE 

&IGURE  -EAN SQUARED ERROR RELATIVE TO THE CHANNEL POWER FOR THE FOUR ANALYSED ESTIMATORS WITH THE HIGHER COMPLEXITY  MULTIPLICATIONS PER TONE 



0ART  ! #OMPARATIVE 3TUDY OF 0ILOT BASED #HANNEL %STIMATORS



"IT ERROR RATE

4HE ESTIMATORS HAVE BEEN SIMULATED IN THE CODED SYSTEM )N &IGURE  THE CODED "%2 IS SHOWN FOR THE ESTIMATORS WITH THE LOWER COMPLEXITY { MULTIPLICATIONS PER ATTENUATION  !S A REFERENCE A SYSTEM WITH PERFECT KNOWLEDGE OF THE CHANNEL AT THE RECEIVER IS INCLUDED !S OBSERVED ABOVE ALL ESTIMATORS SUdER FROM AN ERROR OOR WHICH APPEARS DUE TO THE INTERPOLATION )N &IGURE  THIS IS ONLY NOTICEABLE FOR THE  $ ESTIMATOR 4HE OTHER ESTIMATORS DO NOT HAVE THIS DRAWBACK FOR 2-1   D" AND THEY PERFORM BETTER )T IS ALSO NOTICEABLE THAT ESTIMATORS THAT PERFORM WELL IN TERMS OF -3% ALSO HAVE A LOW "%2 WHICH IS EXPECTED 4HE BEST ESTIMATOR IS THE LOW RANK SEPARABLE ESTIMATOR WHICH IS ONLY ABOUT  D" WORSE THAN KNOWN CHANNEL AND  D" BETTER THAN SEPARABLE &)2 LTERS .OTE THAT FOR -3% THIS LATTER DIdERENCE WAS  D" )N &IGURE  THE "%2 CURVES FOR THE ESTIMATORS WITH THE HIGHER COMPLEXITY ARE SHOWN !GAIN THE "%2 WITH KNOWN CHANNEL IS INCLUDED AS A REFERENCE 4HE LOW RANK SEPARABLE ESTIMATOR IS NOW ONLY  D" AWAY FROM KNOWN CHANNEL AND STILL ABOUT  D" BETTER THAN SEPARABLE &)2 LTERS )N THIS GURE WE NOTE THAT THE PER FORMANCE OF ALL ESTIMATORS HAS INCREASED BUT THE ORDERING BETWEEN THEM IS NOT CHANGED IE THE LOW RANK SEPARABLE ESTIMATOR IS THE BEST AND THE  $ IS THE WORST 4HE INVESTIGA TION HERE SUGGESTS THAT THIS HOLDS FOR MOST COMPLEXITY LEVELS (OWEVER FOR OTHER TYPES OF CHANNELS AND SCENARIOS SUCH AS THE UPLINK  ANOTHER ESTIMATOR MIGHT BE BETTER 3EPARATE STUDIES ARE REQUIRED FOR THESE CIRCUMSTANCES &INALLY THE TWO LEVELS OF COMPLEXITY FOR THE LOW RANK SEPARABLE ESTIMATOR WHICH WAS SHOWN TO BE THE BEST ARE COMPARED TO KNOWN CHANNEL )N &IGURE  IT CAN BE SEEN THAT THE LOW AND HIGH COMPLEXITY ESTIMATORS ARE ABOUT  D" AND  D" AWAY FROM KNOWN CHANNEL RESPECTIVELY 4HE "%2 WILL DECREASE WITH INCREASING COMPLEXITY BUT TO GET REALLY CLOSE TO THE PERFORMANCE OF KNOWN CHANNEL A VERY HIGH COMPLEXITY IS NEEDED 4HIS PROMPTS AN ANALYSIS OF THE TRADE Od BETWEEN COMPLEXITY AND PERFORMANCE BUT THIS IS BEYOND THE SCOPE OF THIS REPORT



#ONCLUSIONS

)N THIS REPORT WE HAVE INVESTIGATED FOUR /&$- CHANNEL ESTIMATORS SUITABLE FOR BROAD CASTING OR FOR THE DOWNLINK IN A MULTIUSER SYSTEM 4HE ESTIMATORS USE PILOTS IE KNOWN SYMBOLS TRANSMITTED IN CERTAIN POSITIONS IN THE TIME FREQUENCY GRID OF /&$- 4WO CLASSES OF ESTIMATORS  DIMENSIONAL AND SEPARABLE WERE INVESTIGATED 7ITHIN EACH CLASS WE COM PARED AN &)2 7IENER LTER WITH A LOW RANK ,--3% ESTIMATOR 4HROUGH ANALYTICAL CAL CULATIONS OF THE -3% AND SIMULATION OF THE CODED "%2 IT WAS FOUND THAT THE SEPARABLE ESTIMATORS WERE THE BEST FOR A XED COMPLEXITY 7ITHIN THE CLASS OF SEPARABLE ESTIMATORS THE LOW RANK ESTIMATOR WAS SHOWN TO BE ABOUT  D" BETTER THAN THE &)2 ESTIMATOR FOR THE CODED "%2 4WO LEVELS OF COMPLEXITIES WERE INVESTIGATED  AND  MULTIPLICATIONS PER ESTIMATED ATTENUATION AND IT WAS FOUND THAT FOR THE CODED "%2 THE FORMER IS  D" FROM KNOWN CHANNEL AND THE LATTER  D" ! NATURAL CONTINUATION OF THIS INVESTIGATION OF CHANNEL ESTIMATION IN /&$- SYSTEMS IS A MORE COMPREHENSIVE STUDY OF THE TRADE Od BETWEEN COMPLEXITY AND PERFORMANCE



&IGURE  #ODED BIT ERROR RATE WITH THE LOW COMPLEXITY ESTIMATORS

&IGURE  #ODED BIT ERROR RATE WITH THE HIGH COMPLEXITY ESTIMATORS



0ART  ! #OMPARATIVE 3TUDY OF 0ILOT BASED #HANNEL %STIMATORS

&IGURE  "%2 OF KNOWN CHANNEL AND THE SEPARABLE LOW RANK ESTIMATOR OF LOW COM PLEXITY  MULTATT AND HIGH COMPLEXITY  MULTATT  7E HAVE USED A PILOT PATTERN WHERE  OF THE TRANSMITTED SYMBOLS ARE KNOWN 4HIS PILOT PATTERN IS SUbCIENT TO OBTAIN GOOD ESTIMATIONS OF THE CHANNEL ATTENUATIONS WHILE INTRODUCING ONLY A SMALL OVERHEAD 7E HAVE ASSUMED THAT THE RECEIVER CAN USE ALL PILOT SYMBOLS THAT ARE TRANSMITTED 'ENERALLY THIS IS NOT THE CASE IN THE UPLINK IN A MULTIUSER SYSTEM WHERE THE CHANNEL ESTIMATION CAN BE BASED ONLY ON PILOTS TRANSMITTED BY A SINGLE USER (ENCE FOR THE UPLINK A SEPARATE STUDY MUST BE MADE TO INVESTIGATE THE PERFORMANCE OF THE CHANNEL ESTIMATORS

!

#OMPLEXITY OF LOW RANK ESTIMATORS

8 t( 8 t(

4HE LOW RANK  $ ESTIMATOR CAN BE FORMULATED AS B  & O  G

9O  Ft Ft

HFt  OI Ft 



SINCE & IS A RANK Q MATRIX 4HE INNER PRODUCTS HFt  OI REQUIRE *y MULTIPLICATIONS EACH IE A TOTAL OF Q*y MULTIPLICATIONS 4HE LINEAR COMBINATION IS OVER Q VECTORS OF LENGTH *q  IE REQUIRES Q*q MULTIPLICATIONS 3INCE *q ATTENUATIONS ARE SIMULTANEOUSLY ESTIMATED THE NUMBER OF MULTIPLICATIONS PER ATTENUATION BECOMES t u Q*y
Q*q *y Q 
 *q *q 



"

#ORRELATION MATRICES
188 aE aS $  $ F& E  S & E ` aE  S ` aS G D  8 s tw tw s { 75&w 75&w s {i~w i i ~w D D D , w&w (
   

4HE AUTO CORRELATION OF THE CHANNEL MODEL  IS



3INCE ALL RANDOM VARIABLES ARE INDEPENDENT WE HAVE 188 aE aS

D  8 h s {75&w i h s {i~w i  $ D $ D , w( h i h i  $ Ds {75&w $ Ds {i~w  1i aE 1 aS 



IE THE CHANNEL CORRELATION IS SEPARABLE 4HE EXPECTATIONS CAN BE FOUND FROM STANDARD &OURIER TRANSFORMS ;= h i 1 aS  $ Ds {75&w  ) {%5&MAX aS  a ` h i  ` DQfy (~RMS s {i   1i aE  $ Ds {i~w   ` DQfy (~RM S 
I { aE~RMS WHERE ) a IS THE ZEROTH ORDER "ESSEL FUNCTION OF THE RST KIND .OTE THAT THE CORRELATION FUNCTION FOR THE UNIFORM POWER DELAY PROLE CAN BE OBTAINED BY LETTING ~RMS 

s{iQfy UNIFORM aE   ` D  1i I {aE 3fy



WHERE

4HE CORRELATION BETWEEN CHANNEL ATTENUATIONS SEPARATED BY J SUBCARRIERS AND K /&$SYMBOLS IS h i $ Gt &u G  t t&u u  Qi J Qu K 
   

Qi J Q K

a ERM S s {t(E  ` DC(~  1i  ERM S 

I {J ~  ` DC(~ ERMS t t u u +  1 K -
+ 3  ) {E5&MAX 
K  t J - 3 u

 

AND ~ ERMS  ~RMS 3 IS THE 2-3 SPREAD RELATIVE TO THE SAMPLING INTERVAL 3INCE THE ,3 ESTIMATES AT PILOT POSITIONS ARE Xt&u Mt&u Ot&u   Gt&u
  Wt&u Wt&u THE CROSS CORRELATION AND THE AUTO CORRELATION ARE i h $ Gt&u O  Qi J ` J  Q K ` K   t &u | } h i   $ Ot&u O  Qi J ` J  Q K ` K 
}w $ p J ` J  K ` K   t &u  JWt J 5SING THESE FUNCTIONS THE AUTO CORRELATION 1yy AND THE CROSS CORRELATION 1qy CAN BE CALCULATED
   



0ART  ! #OMPARATIVE 3TUDY OF 0ILOT BASED #HANNEL %STIMATORS

0ART  0ERFORMANCE !NALYSIS OF #ODED /&$- ON &ADING #HANNELS WITH .ON IDEAL )NTERLEAVING AND #HANNEL +NOWLEDGE

4HIS PART HAS BEEN PUBLISHED AS - 3ANDELL 3+ 7ILSON 0/ "RJESSON 0ERFORMANCE !NALYSIS OF #ODED /&$- ON &AD ING #HANNELS WITH .ON IDEAL )NTERLEAVING AND #HANNEL +NOWLEDGE 2ESEARCH 2EPORT 45,%!  $IVISION OF 3IGNAL 0ROCESSING ,ULE 5NIVERSITY OF 4ECHNOLOGY



0ERFORMANCE !NALYSIS OF #ODED /&$- ON &ADING #HANNELS WITH .ON IDEAL )NTERLEAVING AND #HANNEL +NOWLEDGE
!BSTRACT )N THIS PAPER WE INVESTIGATE THE CODED BIT ERROR RATE IN AN ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING /&$- SYSTEM 4HE SYSTEM USES A PILOT BASED CHANNEL ESTIMATOR AND THE EdECTS OF NON IDEAL CHANNEL KNOWLEDGE AND NON IDEAL IN TERLEAVING ARE ANALYSED 4HE RESULTING CODED BIT ERROR RATE IS CALCULATED WITH AN AN ALYTICAL METHOD PROPOSED BY #AVERS AND (O 4HIS METHOD AVOIDS TIME CONSUMING SIMULATIONS WHICH IS IMPORTANT WHEN DESIGNING A COMMUNICATION SYSTEM 4HE THE ORETICAL RESULTS ARE COMPARED WITH SIMULATIONS AND A GOOD AGREEMENT IS ACHIEVED



)NTRODUCTION

/RTHOGONAL FREQUENCY DIVISION MULTIPLEXING /&$- IS AN EMERGING TECHNIQUE FOR WIRE LESS COMMUNICATION )T IS USED IN %UROPE IN DIGITAL AUDIO BROADCASTING $!" ;= AND IS PROPOSED FOR DIGITAL VIDEO BROADCASTING ;= )TS RESISTANCE TO MULTIPATH FADING HAS SHOWN IT TO BE USEFUL IN BROADCASTING APPLICATIONS AND IT IS CURRENTLY ALSO CONSIDERED FOR MULTI USER SYSTEMS ; = )N THOSE SYSTEMS A HIGH SPECTRAL EbCIENCY IS NEEDED TO USE THE AVAILABLE BANDWIDTH (ENCE MULTIAMPLITUDE MODULATION SCHEMES MAY BECOME NECESSARY TO USE 4HESE CAN BE MADE DIdERENTIAL ;= WHICH SINCE EXPLICIT CHANNEL ESTIMATION IS NOT NECESSARY SIMPLIES THE RECEIVER (OWEVER A PENALTY IN THE FORM OF INCREASED NOISE POWER IS INTRODUCED WHICH IS AVOIDED BY COHERENT DEMODULATION )N THAT CASE CHANNEL ESTIMATION BECOMES AN IMPORTANT PART OF A COMMUNICATION SYSTEM #HANNEL ESTIMATORS ARE USUALLY EVALUATED BY THEIR MEAN SQUARED ERROR PERFORMANCE (OWEVER IN A COMMUNICATION SYSTEM THE AVERAGE BIT ERROR RATE "%2 IS A MORE RELEVANT MEASURE 4HE "%2 CAN BE OBTAINED BY SIMULATIONS BUT THAT IS TIME CONSUMING AND OdERS LITTLE OR NO INSIGHT TO THE DESIGN PROBLEM OF A CHANNEL ESTIMATOR !NALYSIS TOOLS FOR CODED SYSTEMS HAVE EXISTED FOR SOME TIME ; = BUT THESE USUALLY ASSUME PERFECT CHANNEL KNOWLEDGE AND ARE NOT APPLICABLE IN THIS CASE )N  #AVERS AND (O ;= DERIVED A NEW METHOD FOR CALCULATING THE BIT ERROR RATE OF A CODED SYSTEM THAT ASSUMES IDEAL INTERLEAVING BUT ALLOWS FOR NON IDEAL CHANNEL KNOWLEDGE )N  AND  NON IDEAL INTERLEAVING WAS ANALYZED IN ; = 4HE PURPOSE OF THIS REPORT IS TO APPLY THE GENERAL TECHNIQUES MENTIONED ABOVE TO A CODED /&$- SYSTEM WHERE WE INVESTIGATE THE IMPACT OF CHANNEL ESTIMATION IN THOSE SYSTEMS 4HE EdECTS OF INTERLEAVING AND CHANNEL ESTIMATOR COMPLEXITY ARE ANALYZED AND VERIED WITH SIMULATIONS 4HIS ANALYSIS WILL BE IMPORTANT IN THE DESIGN OF A CHANNEL ESTIMATOR IN A /&$- SYSTEM SINCE THE "%2 CURVES CAN BE OBTAINED VERY QUICKLY (ENCE IT IS EASY TO VARY CERTAIN PARAMETERS AND ALMOST IMMEDIATELY SEE THE RESULT WITHOUT LENGTHY SIMULATIONS 4HE SYSTEM MODEL OF THE /&$- SYSTEM WE ARE CONSIDERING IS DESCRIBED IN 3ECTION



0ART  0ERFORMANCE !NALYSIS OF #ODED /&$- ON &ADING #HANNELS

 4HE CHANNEL ESTIMATOR STRUCTURE AND THE CODE USED ARE ALSO EXPLAINED THERE )N 3ECTION  THE ERROR ANALYSIS METHOD IS DESCRIBED FOR BOTH THE CHANNEL ESTIMATOR AND THE DECODER )N 3ECTION  THE RESULTS FROM THE ANALYTICAL METHOD ARE COMPARED WITH SIMULATIONS AND IN 3ECTION  THE RESULTS ARE DISCUSSED




3YSTEM DESCRIPTION
/&$- MODEL

&IGURE A DEPICTS THE /&$- BASE BAND MODEL USED IN THIS PAPER 7E ASSUME THAT

&IGURE  /&$- SYSTEM A "ASE BAND MODEL B EQUIVALENT MODEL THE USE OF A CYCLIC PREX #0 ;= BOTH PRESERVES THE ORTHOGONALITY OF THE TONES AND ELIMINATES INTERSYMBOL INTERFERENCE )3) BETWEEN CONSECUTIVE /&$- SYMBOLS &URTHER THE CHANNEL F ~  S IS ASSUMED TO BE SLOWLY FADING SO IT IS CONSIDERED TO BE CONSTANT DURING ONE /&$- SYMBOL 4HE NUMBER OF TONES IN THE SYSTEM IS -  WHICH MAKES THE EdECTIVE SYMBOL LENGTH 3  - 3  WHERE 3 IS THE SAMPLING PERIOD OF THE SYSTEM 4HE LENGTH OF THE CYCLIC PREX IS 38  +3 AND THE TOTAL SYMBOL LENGTH IS 3
38  )F THE DURATION OF THE IMPULSE RESPONSE OF THE CHANNEL IS SHORTER THAN THE CYCLIC PREX WE CAN DESCRIBE THE SYSTEM AS A SET OF PARALLEL 'AUSSIAN CHANNELS ;= SHOWN IN &IGURE B WITH CORRELATED CHANNEL ATTENUATIONS Gt  4HE RECEIVED SIGNAL ON SUBCHANNEL J CAN THUS BE DESCRIBED AS Xt  Gt Wt
Mt  

 WHERE Gt  & t u  J      - `  

J -3

IS THE ATTENUATION AT SUBCARRIER J AND & a IS THE FREQUENCY RESPONSE OF THE CHANNEL F S ~ DURING THE /&$- SYMBOL (OWEVER IT SHOULD BE NOTED THAT INTERCARRIER INTER FERENCE )#) OCCURS IF THE CHANNEL FADES DURING AN /&$- SYMBOL ; = 4HIS )#) INCREASES WITH THE FADING RATE AND FOR FAST FADING ENVIRONMENTS THIS SHOULD BE TAKEN INTO CONSIDERATION )N ;= IT IS SHOWN THAT )#) CAN BE MODELLED AS 'AUSSIAN ADDITIVE INTERFERENCE IF THE NUMBER OF SUBCARRIERS IS LARGE 4HIS INTERFERENCE IS UNCORRELATED WITH THE FREQUENCY RESPONSE Gt  &OR A 2AYLEIGH FADING CHANNEL THE SIGNAL TO INTERFERENCE RATIO 3)2 IS ;= u  t E  8  {E5 J - ` J )   3)2)#)   `  -
 t( WHERE E5 IS THE MAXIMUM $OPPLER FREQUENCY RELATIVE TO THE INTERTONE SPACING AND ) a IS THE ZEROTH ORDER "ESSEL FUNCTION OF THE RST KIND ;= )N THIS REPORT THE 2(1 IS LARGE ENOUGH TO IGNORE INTERSYMBOL FADING SEE NEXT SECTION AND WE USE THE SIMPLE MODEL  WITHOUT CONSIDERING THE )#)



3CENARIO

4HE SYSTEM WE ARE CONSIDERING IS A WIRELESS MULTIUSER SYSTEM EG A THIRD GENERATION MOBILE TELEPHONE SYSTEM )T IS OPERATING AT THE  '(Z BAND WITH A BANDWIDTH OF  -(Z AND  SUBCARRIERS 4HIS MEANS THAT THE INTERTONE SPACING IS  a !    K(Z AND THE SYMBOL DURATION EXCLUDING THE CYCLIC PREX IS  a    xS 4HE LENGTH OF THE CYCLIC PREX IS CHOSEN TO BE  SAMPLES WHICH MAKES THE OVERHEAD EQUAL TO    AND THE EdECTIVE /&$- SYMBOL LENGTH  a    xS )N THIS REPORT WE ASSUME PERFECT SYNCHRONIZATION BETWEEN TRANSMITTER AND RECEIVER 7E WILL ONLY DEAL WITH THE DOWNLINK WHERE SYNCHRONIZATION IS EASIER ALTHOUGH THE METHODS ARE GENERAL AND ARE APPLICABLE TO THE UPLINK AS WELL 4HE ENVIRONMENT THE SYSTEM IS WORKING IN IS ASSUMED TO BE A MACROCELL WHICH CAN BE CHARACTERIZED BY A 2AYLEIGH FADING CHANNEL WITH A $OPPLER FREQUENCY OF  (Z CORRESPONDING TO  KMH AND A MAXIMUM TIME DISPERSION OF  xS .ORMALIZED TO THE INTERTONE SPACING AND SAMPLING INTERVAL RESPECTIVELY THIS MEANS     SAMPLES 4HIS $OPPLER RATE GIVES US A 2(1 E5   AND ~MAX    ( {  (  OF  D" SEE   3INCE WE EVALUATE THE SYSTEM FOR 2-1   D" THIS INTERFERENCE BECOMES NEGLIGIBLE AND WE IGNORE THE )#) CAUSED BY CHANNEL FADING WITHIN AN /&$SYMBOL 4HE IMPULSE RESPONSE OF THE CHANNEL AT TIME S IS MODELLED AS ;=
`

F ~ S 

D 8 v(

mv S p ~ ` ~v 



WHERE THE FADING AMPLITUDES mv S ARE INDEPENDENT COMPLEX 'AUSSIAN RANDOM VARIABLES AND THE DELAYS ~v ARE UNIFORMLY DISTRIBUTED BETWEEN  AND ~MAX  4HIS DISPERSION IS SHORTER



0ART  0ERFORMANCE !NALYSIS OF #ODED /&$- ON &ADING #HANNELS

THAN THE CYCLIC PREX AND HENCE )3) IS AVOIDED 4HE POWER SPREAD PROLE IS EXPONENTIALLY DECAYING WITH A ROOT MEAN SQUARE 2-3 WIDTH OF  xS   SAMPLES IE | h i "D~(~RM S   ~  ~MAX    $ JF ~  S J   OTHERWISE WHERE $ IS THE EXPECTED VALUE OVER S AND " IS A NORMALIZATION CONSTANT



#HANNEL ESTIMATION

0ILOT SYMBOL ASSISTED MODULATION ;= HAS BEEN PROPOSED AS AN EbCIENT WAY OF COMBATING FADING 0ILOT SYMBOLS KNOWN TO THE RECEIVER ARE MULTIPLEXED INTO THE TRANSMITTED SYMBOL STREAM AND WITH THE AID OF THESE THE FADING CHANNEL IS ESTIMATED BY INTERPOLATING BETWEEN THE PILOT SYMBOLS &OR /&$- THIS INTERPOLATION CAN BE DONE IN TWO DIMENSIONS TIME AND FREQUENCY  BUT THIS USUALLY LEADS TO ESTIMATORS WITH A HIGH COMPLEXITY 4HEREFORE SEPARATING THE CHANNEL ESTIMATOR INTO TWO ONE DIMENSIONAL ESTIMATORS HAS BEEN PROPOSED ;= &IRST THE CHANNEL ATTENUATIONS ON ALL SUBCARRIERS ARE ESTIMATED IN ONLY THOSE /&$SYMBOLS THAT CONTAIN PILOT SYMBOLS SEE &IGURE  4HEN ALL CHANNEL ATTENUATIONS ARE

&IGURE  0ILOT SYMBOLS MARKED IN GRAY USED FOR CHANNEL ESTIMATION 4HE CHANNEL ATTENUATIONS ARE RST ESTIMATED IN THE FREQUENCY DIRECTION  AND THEN IN THE TIME DIRECTION   ESTIMATED USING THESE ESTIMATES 4HE SEPARATE CHANNEL ESTIMATORS CAN BE &)2 LTERS ;= OR LOW RANK ESTIMATORS ;= THAT ARE DESIGNED TO MINIMIZE THE MEAN SQUARED ERROR -3%  )N THIS REPORT WE WILL CONSIDER ONLY THE &)2 LTERS .OMINAL VALUES OF THE CHANNEL CORRELATION AND SIGNAL TO NOISE RATIO 3.2 ARE USED IN ORDER TO KEEP THE ESTIMATOR XED WHICH SIMPLIES THE IMPLEMENTATION $ESIGN FOR THE WORST CASE HAS SHOWN TO BE ROBUST AND BENECIAL WITH REGARD TO THE MAXIMUM ERROR LEVEL ;  = )N THIS REPORT WE WILL

 DESIGN THE ESTIMATORS FOR A UNIFORM POWER DELAY PROLE FOR   ~  ~MAX  2-1   D" AND E5    )F -i AND - DENOTE THE NUMBER OF TAPS IN THE FREQUENCY AND TIME LTERS THE AVERAGE NUMBER OF MULTIPLICATIONS PER ATTENUATION IS -i 
-  SINCE ONLY EVERY FOURTH /&$- SYMBOL NEEDS TO BE ESTIMATED IN THE FREQUENCY DIRECTION ;= 7E ANALYZE TWO ESTIMATORS WITH DIdERENT COMPLEXITIES ONE WITH  TAPS IN THE FREQUENCY DIRECTION AND  TAPS IN THE TIME DIRECTION AND ONE WITH  AND  TAPS RESPECTIVELY 4HESE TWO ESTIMATORS HAVE AN AVERAGE COMPLEXITY OF  AND  MULTIPLICATIONS PER ESTIMATED ATTENUATION RESPECTIVELY



#ODING

&OR ERROR PROTECTION A CONVOLUTIONAL CODE IS USED 4O MAKE THE ANALYSIS TRACTABLE WE USE A RATHER SMALL CODE WITH CONSTRAINT LENGTH  WHICH HAS BEEN FOUND TO BE VERY EdECTIVE ON FADING CHANNELS ;= 4HE IN PHASE AND QUADRATURE SIGNALS ARE ENCODED SEPARATELY WITH A RATE  CODE WITH GENERATOR POLYNOMIALS ;= F F


# #

 
#
#  
# 

  4HE ENCODED OUTPUTS ARE MODULATED USING 

 

4HE ENCODER IS DEPICTED IN &IGURE 

&IGURE  4HE CONVOLUTIONAL ENCODER 0!- AND CONCATENATED TO A  1!- SYMBOL &OR A FADING CHANNEL THE DIVERSITY ORDER OF A CODE IS NOT THE FREE DISTANCE BUT THE LENGTH OF THE SHORTEST ERROR EVENT ;= 4HE LENGTH OF AN ERROR EVENT + IS THE NUMBER OF SYMBOLS ON AN ERRONEOUS PATH THAT DIdER FROM THE CORRECT PATH &ROM THE TRELLIS OF THE CODE &IGURE  IT CAN BE SEEN THAT THE SHORTEST ERROR EVENT IS +MIN   4HE RECEIVER RST EQUALIZES THE RECEIVED SIGNAL Xt USING THE CHANNEL ESTIMATE B Gt AND THEN SEPARATES THE SIGNAL INTO ITS IN PHASE AND QUADRATURE PARTS Yt  Xt @ H  Yt
IYt  B Gt n 8n nB n G Et ` W Et J  n t n JY
t



4HE DECODING IS PERFORMED SEPARATELY WITH THE DECODING METRIC E  , W 



0ART  0ERFORMANCE !NALYSIS OF #ODED /&$- ON &ADING #HANNELS

&IGURE  4RELLIS FOR THE RATE  CONVOLUTIONAL CODE USED 4HE BOLD PATH IS THE SHORTEST ERROR EVENT Et IS EITHER Yt OR Yt  WHERE W Et IS THE SEQUENCE FOR WHICH THE METRIC IS COMPUTED AND Y 5SING THE 6ITERBI ALGORITHM THE SEQUENCE W Bt WHICH MINIMIZES THIS METRIC IS FOUND IE B  @QF LHM , W E  W
E W
@ H



)T SHOULD BE NOTED THAT THE METRIC IN  IS NOT THE MAXIMUM LIKELIHOOD METRIC WHEN MULTIAMPLITUDE MODULATION IS USED ;= BUT WE WILL NEVERTHELESS USE IT FOR SIMPLICITY )N ORDER TO OBTAIN DIVERSITY IN THE SYSTEM FROM THE CHANNEL CODE INTERLEAVING MUST BE USED )NTERLEAVING BREAKS UP THE CHANNEL CORRELATION AND IN THE IDEAL CASE PROVIDES THE DECODER WITH UNCORRELATED SYMBOLS )N THIS REPORT WE WILL USE FREQUENCY INTERLEAVING ON A SYMBOL LEVEL IE INTERLEAVING IS PERFORMED AFTER THE BITS HAVE BEEN ASSIGNED TO CONSTELLATION POINTS 4HE SYMBOLS ARE BLOCK INTERLEAVED WITH DEPTH #i SUBCARRIERS ACROSS THE TONES 4HE INTERLEAVING IS ILLUSTRATED IN &IGURE  WITH #i  

&IGURE  )NTERLEAVING IN FREQUENCY ACROSS THE TONES )N THIS REPORT WE ASSUME A ,   PATH CHANNEL WHICH ALLOWS THE CODE TO USE ITS FULL DIVERSITY OF ORDER  ;= "Y CHOOSING #i   WE HAVE    SYMBOLS WHOSE DISTANCE FROM EACH OTHER IS LARGE ENOUGH TO MINIMIZE THE EdECTS OF CORRELATION )N &IGURE  THE FREQUENCY CORRELATION OF THE CHANNEL IS DEPICTED FOR THE SCENARIO IN THIS REPORT SEE 3ECTION  FOR PARAMETERS 4WO CONSECUTIVE CHANNEL SYMBOLS INTO THE DECODER HAVE



&IGURE  &REQUENCY CORRELATION FOR THE MULTI PATH CHANNEL 4HE CORRELATIONS FOR THE INTERLEAVED SYSTEM #i   ARE MARKED WITH O
 THE CORRELATION COEbCIENT J$ FG L G L `  GJ }q   WHICH IS SMALL ENOUGH NOT TO DEGRADE THE PERFORMANCE SIGNICANTLY




0ERFORMANCE ANALYSIS
#HANNEL ESTIMATION
 $ FG L G L ` J G N O  $ GLB G L ` J N O  $ B GLB G L ` J N O B  $ M L G L`J 

)N ORDER TO EVALUATE A CODED /&$- SYSTEM WE WILL NEED FOUR COVARIANCES FROM THE SYSTEM SEE 3ECTION  Qqq J QqB q J QB qB q J QwB q J    

.OTICE THAT WE ASSUME THE CHANNEL AND THE ADDITIVE NOISE TO BE UNCORRELATED $ FM L G J G   L J  (OW TO DERIVE EXPLICIT EXPRESSIONS FOR THE ABOVE CORRELATIONS IS DESCRIBED IN !PPENDIX "



#ODING

4O EVALUATE THE PROBABILITY OF ERROR FOR A CODE ONE USUALLY STARTS WITH THE ERROR EVENT PROBABILITY !N ERROR EVENT IS DENED AS A PATH IN THE TRELLIS WHICH STARTS AT THE SAME NODE AS THE CORRECT PATH DIVERGES AND RE EMERGES SOME STAGES LATER WITH THE CORRECT PATH 4HE PROBABILITY OF SUCH AN ERROR EVENT CAN BE UPPER BOUNDED WITH EG THE #HERNOd BOUND ; = 4O OBTAIN THE AVERAGE BIT ERROR RATE A UNION BOUND OVER ALL ERROR EVENTS MAY BE APPLIED 4HIS ENUMERATION OF ALL ERROR EVENTS CAN BE CALCULATED WITH THE TRANSFER



0ART  0ERFORMANCE !NALYSIS OF #ODED /&$- ON &ADING #HANNELS

FUNCTION OF THE ERROR STATE DIAGRAM ; = OF THE CODE 4HIS TECHNIQUE HOWEVER ASSUMES PERFECT KNOWLEDGE OF THE CHANNEL SO IT IS NOT SUITABLE FOR THIS REPORT 7E ARE INVESTIGATING THE IMPACT OF CHANNEL ESTIMATION IE WHEN WE DO NOT HAVE PERFECT CHANNEL KNOWLEDGE )NSTEAD WE WILL USE THE ANALYSIS INTRODUCED BY #AVERS AND (O ; = 4HIS TECHNIQUE IS APPLICABLE TO A MORE GENERAL DESCRIPTION OF THE SYSTEM WHICH INCLUDES NON IDEAL CHANNEL KNOWLEDGE  0AIRWISE ERROR PROBABILITY

4HE EXACT PROBABILITY OF AN ERROR EVENT WAS FOUND IN  BY #AVERS AND (O ;= !SSUME THAT THE DECODER CHOOSES THE SEQUENCE WHICH MINIMIZES THE METRIC n 8n n n B , Br  nXt ` Gt Brt n 
t



WHERE Br IS A POSSIBLE SEQUENCE OF CODEWORDS AND J LABELS THE SYMBOLS AFTER INTERLEAVING IE THE ORDER IN WHICH THEY INPUT TO THE DECODER 4HE SEQUENCE WHICH MINIMIZES  CAN BE FOUND WITH THE 6ITERBI ALGORITHM 4HE PROBABILITY OF AN ERROR EVENT IE THAT THE TRANSMITTED SEQUENCE Br IS DECODED AS Bs CAN BE WRITTEN AS /Q Br Bs  /Q , Br  , Bs  /Q #    

WHERE #  , Bs ` , Br IS THE METRIC DIdERENCE FOR THE TWO SEQUENCES 4HE PROBABILITY IN  CAN BE WRITTEN AS ;= /Q #    ` 8 h5 R 2ESIDUE R v w  

2(0 POLES

WHERE THE SUM IS OVER THE POLES OF h5 R R WHICH LIES IN THE RIGHT HALFPLANE OF THE COMPLEX PLANE 1D   AND h5 R IS THE TWO SIDED ,APLACE TRANSFORM OF THE PROBABILITY DISTRIBUTION FUNCTION OF # "Y REWRITING # AS #  , Bs ` , Br n n 8 nB n  B  JXt J ` Xt B G B ` X G B
t st t t st nGt n JBst J ` 8 8
t t t

n n nB n  B JXt J ` Xt B B ` X B
G G t rt t t rt nGt n JBrt J n n ` ` a a n n B B Xt Gt Brt ` Bst
Xt Gt Brt ` Bst
nB Gt n JBst J ` JBrt J  

r s WE NOTE THAT IT IS A QUADRATIC FORM IN THE COMPLEX 'AUSSIAN VARIABLES Xt  B Gt  5SING MATRIX NOTATION AND DENOTING THE LENGTH OF THE SEQUENCES Br AND Bs BY +  CAN BE

 WRITTEN AS 9  X  " # "   "  # "  " # " " XC # " # "  #  " "B # " " G # " Br ` Bs  " # "      ! B  GC \  Y9 ,Y   X #"  # #"   # #" # " XC # BrC ` BsC # #" # #" B #  G  #" #   # " #  !    ! B JBsC J ` JBrC J GC ]\ [Z ]    
Y

aaa  aaa aaa 

 aaa  JBs  J ` JBr J a a a     a a a B  aaa rC ` BsC [Z



   

    

Br ` Bs    

aaa 





"Y USING A LINEAR TRANSFORMATION AND AN EIGENVALUE DECOMPOSITION SEE !PPENDIX $ # CAN BE WRITTEN AS ;= #
C 8

t(

wt JPt J 



WHERE Pt ARE INDEPENDENT 'AUSSIAN RANDOM VARIABLES wt ARE THE EIGENVALUES h COMPLEX i 9 OF 1 , AND 1  $ YY IS THE AUTOCOVARIANCE MATRIX OF Y 3INCE # IS A SUM OF INDEPENDENT VARIABLES THE ,APLACE TRANSFORM h5 R IS FOUND AS ; = h5 R    
wt R t(
C 9



.OTE THAT THE POLES OF h5 R ARE `wt  (ENCE TO ND THE PROBABILITY OF A GIVEN ERROR EVENT IN   WE ND THE EIGENVALUES wt OF THE MATRIX 1 , AND CALCULATE THE RESIDUES OF h5 R R AT THE POLES `wt  WHO LIE IN THE RIGHT HALFPLANE )N THE CASE OF NON IDEAL INTERLEAVING THE CORRELATION MATRIX 1 IS NON TRIVIAL BUT CAN BE CALCULATED AS ;=   9  X X      #"  #   "       "  #"  #     #" #   " " XC # " XC #  " # " # 1  $ " B # " B #   " G # " G #     "  #"  #         !  !       B  B GC GC N O N O   G a a a $ FXXC G $ X B B $ FX X a a a $ X G G  C  " #       " #           " O N O # N " # B " $ FXC X G a a a $ FXC X G $ XCB # G a a a $ X G C  C  C " O N O N O N O #  " N  # B B B B B " $ B # G X a a a $ G X $ G G a a a $ G G      C  C # " " #       " #           N O N O N O N O ! aaa $ B $ B $ B GC X GC XC GCB G aaa $ B GCB G  C



0ART  0ERFORMANCE !NALYSIS OF #ODED /&$- ON &ADING #HANNELS

4HE ELEMENTS OF THIS MATRIX ARE $ FXt Xu G O N $ Xt B G u N O B $ Gt Xu O N G $ B Gt B u

  $ F Gt Wt
Mt Gt Wu
Mu G  1qq J ` K Wt W u
}w p J ` K N O  $ Gt Wt
Mt B G u  1qB q J ` K Wt
1wB q J`K N O B K ` J W K`J  $ Xu B G  1q t u
1wB q q

   

 1B qB q J`K

AND DEPEND ON THE CHANNEL ESTIMATOR USED  "IT ERROR PROBABILITY

4O ND THE AVERAGE BIT ERROR PROBABILITY WE CAN SUM UP ALL THE ERROR EVENTS EACH ASSOCI ATED WITH A NUMBER OF INFORMATION BITS TO CONSTRUCT A UNION BOUND 4HIS CAN BE DONE WITH THE TRANSFER FUNCTION OF THE CODE ;= (OWEVER THE ERROR EVENT PROBABILITIES  ARE NOT IN A FORM SUITABLE TO USE THE TRANSFER FUNCTION 7E WILL TRUNCATE THE SUM AND ONLY CONSIDER ERROR EVENTS OF SMALL LENGTHS ;= 4HIS WILL NOT BE AN UPPER BOUND TO THE BIT ERROR RATE ANYMORE (OWEVER TRUNCATING THE UNION BOUND CAN BE A GOOD APPROXIMATION AS ARGUED IN ;= 4HE AVERAGE BIT ERROR PROBABILITY IS APPROXIMATED BY /e { 8 Lrs /Q Br M s Bs  

WHERE M IS NUMBER OF INPUT BITS PER ENCODING INTERVAL AND Lrs IS THE AVERAGE NUMBER OF BIT ERRORS ASSOCIATED WITH EACH ERROR EVENT ;= 4HE SUM IS TRUNCATED TO CONTAIN ONLY ERROR EVENTS OF LENGTH +  +MAX  4HE MAXIMUM LENGTH +MAX IS CHOSEN AS IN ;= $ENE /e J AS THE ESTIMATED "%2 WHEN ERROR EVENTS OF LENGTH + v J ARE CONSIDERED 7HEN THE INCREMENT /e J
 ` /e J IS MUCH SMALLER THAN THE INCREMENTS FOR SMALLER J  WE CHOOSE +MAX  J )N OUR CASE WE FOUND +MAX   TO BE A SUITABLE CHOICE SINCE CONSIDERING LONGER ERROR EVENTS DID NOT CHANGE THE ESTIMATED "%2 SIGNICANTLY 4HE ERROR EVENTS CAN BE ENUMERATED USING THE TRANSFER FUNCTION OF THE ERROR STATE DIAGRAM ;= 4HE OUTPUT BITS OF THE 
#
#   
# CODE ARE MAPPED TO A  0!- CONSTELLATION 4HE TRELLIS FOR THIS CODE ALONG WITH THE BIT MAPPING FOR THE  0!- CONSTELLATION ARE SHOWN IN &IGURE  .OTE THAT WE ENCODE THE IN PHASE AND QUADRATURE DIMENSIONS SEPARATELY SO FOR A CODED  1!- SYMBOL WE HAVE TWO INPUT BITS AND TWO OUTPUT BITS "ECAUSE WE ENCODE THE IN PHASE AND QUADRATURE SEPARTELY WE CAN ALSO ANALYZE THE TWO SYSTEMS SEPARATELY 4O ENUMERATE ALL ERROR EVENTS WE USE THE TECHNIQUE DESCRIBED IN ; #HAP = &ROM THE  0!- CONSTELLATION WE CAN ASSOCIATE EACH ERROR WITH A DISTANCE 6  6  6  6     p   p
p   p     



&IGURE  4RELLIS OF THE CONVOLUTIONAL CODE AND THE 0!- CONSTELLATION USED IN THE /&$- SYSTEM 4HE DISTANCE 6  MAY DERIVED AS FOLLOWS WHEN THE ERROR  OCCURS THE ASSOCIATED DISTANCE WILL BE p WHEN  OR  IS TRANSMITTED AND p WHEN  OR  IS TRANSMITTED (ENCE WE CAN DENOTE 6  BY  p
p  WHICH CAN BE INTERPRETED AS p HALF OF THE TIME AND  p HALF OF THE TIME &IGURE  DISPLAYS THE ERROR STATE DIAGRAM FOR THIS IMPLEMENTATION OF THE CODE %ACH BRANCH IN THE ERROR STATE DIAGRAM IS LABELED BY 6 a ( + WHERE 6 a IS THE %UCLIDEAN DISTANCE ASSOCIATED WITH THE ERROR Q DENOTES THE NUMBER OF INFORMATION BIT ERRORS AND + ENUMERATES THE BRANCHES 5SING A STATE SPACE DESCRIPTION OF THE ERROR STATE

&IGURE  %RROR STATE DIAGRAM FOR THE CONVOLUTIONAL CODE 

#
#   
# CODE WITH 0!- MODULATION DIAGRAM THE ERROR EVENTS SHOWN IN 4ABLE  CAN BE DERIVED SEE !PPENDIX # FOR THE DETAILS  4HE EdECTIVE LENGTH OF A CODE AS DENED IN SECTION  DENOTES THE NUMBER OF CHANNEL SYMBOLS THAT DIdER IN THE TRANSMITTED AND DECODED SEQUENCE 4HE SMALLEST EdECTIVE



0ART  0ERFORMANCE !NALYSIS OF #ODED /&$- ON &ADING #HANNELS %RROR EVENT         !CTUAL LENGTH %dECTIVE LENGTH Lrs                    p         p         p        

4ABLE  $ESCRIPTION OF THE DOMINATING ERROR EVENTS 4HE EdECTIVE LENGTH IS THE NUMBER OF SYMBOLS THAT DIdERS FROM THE TRANSMITTED CODEWORD Lrs IS THE AVERAGE NUMBER BIT ERRORS ASSOCIATED WITH THE ERROR EVENT AND pr ARE THE %UCLIDEAN DISTANCES OF THE CODEWORD SYMBOLS SEE &IG  LENGTH DETERMINES THE DIVERSITY ORDER OF SYSTEM ;= (ENCE FOR THE CODE CONSIDERED HERE WE HAVE A DIVERSITY ORDER OF 




3IMULATION
0ERFECT CHANNEL ESTIMATION

4O EVALUTE THE EdECT OF NON IDEAL INTERLEAVING ON OUR /&$- SYSTEM WE APPLY THE TECH NIQUES DESCRIBED IN SECTION  FOR THE CASE OF IDEAL AND NON IDEAL INTERLEAVING WITH PERFECT CHANNEL ESTIMATION 4HE IDEAL INTERLEAVING ASSUMES THAT ALL CHANNEL ATTENUATIONS ARE UNCORRELATED AND THE NON IDEAL INTERLEAVING IS DONE AS DESCRIBED IN 3ECTION  4HE CODING AND MODULATION IS DONE WITH THE 
#
#  
# CODE AND  0!- RESPEC TIVELY SEPARATELY ON THE INPHASE AND QUADRATURE SIGNALS 4HEY ARE THEN COMBINED TO FORM A  1!- SYMBOL )N THE ANALYSIS ALL ERROR EVENTS OF LENGTH  AND SHORTER ARE CONSIDERED )N &IGURE  THE ANALYTICAL BIT ERROR RATE CURVE IS SHOWN TOGETHER WITH SIMULATIONS !S CAN BE NOTED FROM THE GURE THE DIdERENCE BETWEEN IDEAL AND NON IDEAL INTERLEAVING IS VERY SMALL 4HUS THE RESULTING CHANNEL CORRELATION AFTER INTERLEAVING DOES NOT DEGRADE THE PERFORMANCE IN ANY SIGNICANT WAY



#HANNEL ESTIMATION

4HE CHANNEL ESTIMATION IN THESE SIMULATIONS IS AS DESCRIBED IN 3ECTION  7E RST USE AN &)2 LTER TO ESTIMATE THE CHANNEL ATTENUATIONS ON ALL TONES IN A GIVEN /&$- SYMBOL THAT CONTAIN PILOT SYMBOLS 7E THEN USE AN &)2 LTER IN THE TIME DOMAIN TO OBTAIN ESTIMATES OF ALL TONES 4HUS THE CHANNEL ESTIMATOR IS A SEPARABLE TWO DIMENSIONAL LINEAR LTER THAT INTERPOLATES BETWEEN THE INITIAL ESTIMATES OBTAINED AT THE PILOT POSITIONS 4HE RST ESTIMATOR USES A  TAP LTER IN THE FREQUENCY DIRECTION AND A  TAP LTER IN THE TIME DIRECTION WHICH GIVES AN AVERAGE OF  MULTIPLICATIONS PER ATTENUATION 4HE SECOND ESTIMATOR USES  AND  TAPS RESPECTIVELY WHICH RESULTS IN  MULTIPLICATIONS PER



&IGURE  "IT ERROR RATE WITH KNOWN MULTI PATH FADING CHANNEL 3IMULATIONS FOR IDEAL AND NON IDEAL INTERLEAVING ARE MARKED WITH O AND X RESPECTIVELY 4HE SOLID LINES ARE THE ANALYTICAL RESULTS INDISTINGUISABLE FOR THE TWO CASES  ATTENUATION 4HE THEORETICAL BIT ERROR RATE FOR THESE TWO ESTIMATORS IS SHOWN IN &IGURE  TOGETHER WITH THE SIMULATIONS !S A REFERENCE THE BIT ERROR RATE WITH PERFECT CHANNEL KNOWLEDGE IS INCLUDED 4HE DEGRADATION COMPARED TO PERFECT CHANNEL KNOWLEDGE IS APPROXIMATELY  D" FOR THE LOW COMPLEXITY ESTIMATOR AND  D" FOR THE HIGH COMPLEXITY ESTIMATOR 4HE THEORETICAL RESULTS SHOW GOOD AGREEMENT WITH THE SIMULATIONS EXCEPT FOR LOW 3.2 &OR THOSE LOW 3.2 VALUES THE ANALYTICAL METHOD OVER ESTIMATES THE BIT ERROR RATE



#ONCLUSIONS

)N THIS REPORT WE HAVE APPLIED A GENERAL ANALYSIS METHOD TO A CODED /&$- SYSTEM 4HIS METHOD IS MOST USEFUL IN THE DESIGN OF A SYSTEM WITH RESPECT TO CHANNEL ESTIMA TOR COMPLEXITY INTERLEAVING SCHEMES ETC SINCE IT DOES NOT REQUIRE ANY TIME CONSUMING SIMULATIONS &URTHERMORE IT IS GENERAL AND ALLOWS THE ANALYSIS TO CONSIDER SYSTEMS WITH NON IDEAL CHANNEL ESTIMATION AND INTERLEAVING 7E HAVE INVESTIGATED TWO VERSIONS OF A CHANNEL ESTIMATOR WITH DIdERENT COMPLEXITIES 4HE LOWER COMPLEXITY ESTIMATOR IS ABOUT  D" FROM THE KNOWN CHANNEL WHILE THE DEGRADATION FOR THE HIGHER COMPLEXITY ESTIMATOR IS ABOUT  D" 4HE THEORETICAL RESULTS AGREE WELL WITH SIMULATIONS EXCEPT AT LOW 3.2S WHERE IT OVER ESTIMATES THE BIT ERROR RATE (OWEVER FOR THESE LOW 3.2S THE DIdERENCE BETWEEN THE DIdERENT ESTIMATORS IS NOT INTERESTING FROM A DESIGN POINT OF VIEW AS ALL THE



0ART  0ERFORMANCE !NALYSIS OF #ODED /&$- ON &ADING #HANNELS

&IGURE  "IT ERROR RATE FOR THE MULTIPATH FADING CHANNEL &ROM BOTTOM TO TOP KNOWN CHANNEL ESTIMATOR !  MULTTONE AND ESTIMATOR "  MULTTONE  ESTIMATORS GIVE APPROXIMATELY THE SAME BIT ERROR RATE 4HUS THE INVESTIGATED METHOD SHOWS A HIGH POTENTIAL FOR USE IN THE DESIGN PHASE OF CODED /&$- SYSTEMS SINCE IT IS FAST SIMPLE AND ACCURATE )N THIS REPORT WE HAVE ONLY CONSIDERED THE EdECT OF CHANNEL ESTIMATION BUT OTHER PARAMETERS CAN ALSO BE INVESTIGATED SUCH AS PILOT DENSITY CHOICE OF CODE ETC

!

#HANNEL CORRELATION
D 8 v(

4HE FREQUENCY RESPONSE OF THE CHANNEL MODEL IN  IS G E S  mv S Ds {i~v 

AND THE FREQUENCY CORRELATION BETWEEN TWO ATTENUATIONS SPACED K SUBCARRIERS APART IS | t u t u} D 8 J J`K 1FREQ K  $ G S G S $ Jmv S J Ds {u~v (EQ - 3 -3 v( : D CQ  8  "D~v (~RMS Ds {u~v(EQ C~v +3 v(  r s CQ C  ` DWO ` ~RM S ` I {K E  ", a   CQ C ~RM S
I {K E

 WHERE THE NORMALIZATION COEbCIENT " IS CHOSEN SUCH THAT 1FREQ    4HE CHANNEL ESTIMATOR IS DESIGNED FOR A UNIFORM POWER DELAY PROLE WHICH CAN BE OBTAINED WHEN ~RMS  uu t t + ", 1FREQUNIFORM K    C  ` DWO `I {K I {K E 4HE TIME VARIATION OF THE CHANNEL FOLLOWS *AKES MODEL ;= WHOSE TIME CORRELATION IS 1 S  ) {%5 S  WHERE ) a IS THE ZEROTH ORDER "ESSEL FUNCTION OF THE RST KIND AND %5 IS THE MAXIMUM $OPPLER FREQUENCY 4HUS THE CORRELATION BETWEEN TWO ATTENUATIONS WITH A DISTANCE OF L /&$- SYMBOLS IS t t uu + 1TIME L  ) {%5 L -
+ 3  ) {E5 L 
  WHERE E5 IS THE RELATIVE TO THE INTERTONE SPACING MAXIMUM $OPPLER FREQUENCY

"

#HANNEL ESTIMATOR

4HE CHANNEL ESTIMATOR USED HERE IS A SEPARABLE 7IENER LTER ;= #ONSIDER RST THE CHANNEL ESTIMATION IN THE FREQUENCY DIRECTION )N THOSE /&$- SYMBOLS WHERE THERE ARE PILOTS THESE ESTIMATES ARE 8 E Gr J K  GLS J ` L-i ` H K  H       -i `  mrB 
v v

WHERE J K DENOTES SUBCARRIER J AT SYMBOL K SUPERSCRIPT H INDICATES THAT THE ESTIMATED ATTENUATION LIES H SUBCARRIERS FROM A PILOT AND -i IS THE DISTANCE BETWEEN PILOT SYMBOLS r .OTE THAT PILOTS ARE PLACED AT H   AND THAT WE NEED -i DIdERENT ESTIMATORS mv SINCE THE ESTIMATED ATTENUATIONS ARE PLACED DIdERENTLY RELATIVE TO THE PILOTS 4HESE PRELIMINARY ESTIMATES ARE THEN USED FOR THE ESTIMATION IN THE TIME DIRECTION 8 r&s B G  n s E Gr J K ` M- ` I  I       - ` 
t&u

8
v&w

r s B mv nw GLS J ` L-i ` H K ` M- ` I 



(ERE THE SUPERSCRIPT I DENOTES THE DISTANCE BETWEEN THE ESTIMATED ATTENUATION AND AN /&$- SYMBOL WITH PILOTS "Y PLACING ALL THE USED PILOTS IN A VECTOR O THE ESTIMATOR MAY BE EXPRESSED AS r&s B G  FQ Otr&us  
t&u r&s

WHERE

  r s B mvM IN nwM IN GLS J ` LMIN -i  K ` MMIN -  " # "    " # "   " r # " s B " mvM AX nw # " GLS J ` LMAX -i  K ` MMIN - MIN # Fr&s  "  Ot&u  " s # " mr " B " vM IN  nwMIN # " GLS J ` LMIN -i  K ` MMIN - " # "     !   r s B GLS J ` LMAX -i  K ` MMAX - mvMAX nwM AX

 # # # # # # # # !





0ART  0ERFORMANCE !NALYSIS OF #ODED /&$- ON &ADING #HANNELS

&ROM THIS DESCRIPTION IT IS STRAIGHTFORWARD TO OBTAIN THE CORRELATIONS NEEDED FOR THE ANALY SIS $UE TO THE PERIODIC PATTERN OF THE PILOTS THERE WILL BE -i - DIdERENT ESTIMATORS ALL WITH DIdERENT PERFORMANCE (ENCE WE AVERAGE OVER ALL ESTIMATED ATTENUATION TO GET A SCALAR PARAMETER &OR INSTANCE THE CROSS CORRELATION BETWEEN THE CHANNEL AND THE CHAN NEL ESTIMATE IS SET TO BE

  1qB q J K J  K 

 8 N Br&s O $ Gt&u Gt &u  -i - r&s

 



4HIS CAN BE JUSTIED IN THAT THERE IS ONLY A SMALL VARIATION AMONG THE ESTIMATED ATTENU ATIONS AND THEY HAVE ALL APPROXIMATELY THE SAME PROPERTIES .OTE THAT THE CORRELATIONS SHOULD BE CALCULATED AFTER DE INTERLEAVING AND THAT THE INDICES OF THE VARIABLES SHOULD BE ADJUSTED ACCORDINGLY

#

%RROR EVENT ENUMERATION

4HE ERROR STATE DIAGRAM FOR THE 

#
#  
# CODE WITH  0!- MODULATION CON SIDERED IN THIS REPORT IS DEPICTED IN &IGURE  4O CALCULATE THE TRANSFER FUNCTION OF THE ERROR STATE DIAGRAM WE USE A STATE SPACE DESCRIPTION ;=    z z !  z \      (+  z p (+  !  ! zIN  p 
p +  p + z !
  p (+  p
p + z  [Z  ] \ [Z ] !   z  ` a  p +  z !  zOUT  \ [Z ] z "

4HE TRANSFER FUNCTION IS zOUT " (` zIN


t  AND THAT ALL MATRICES CONTAIN THE SCALAR + THE

0 "Y NOTING THAT ( `    t( TRANSFER FUNCTION CAN BE WRITTEN AS

 zOUT 8 E E t E t  " !+ zIN t(



E AND "  +" E  (ENCE zOUT zIN IS A POLYNOMIAL IN + AND ALL ERROR WHERE  + E  !  +! E E t! E  4HIS CAN NOW BE USED TO ND ALL DOMINANT EVENTS OF LENGTH J
 CAN BE FOUND AS " ERROR EVENTS



$

,APLACE TRANSFORM h# R

4HIS DERIVATION OF THE ,APLACE TRANSFORM OF THE QUADRATIC FORM # FOLLOWS 4URIN ;= AND -AZO AND 3ALZ ;= 4HE QUADRATIC FORM #  Y9 ,Y

1 IS (ERMITIAN IE 1  19  &ROM THE NEW QUADRATIC FORM

( ( #  O9 1 ,1 O

IS EASIER TO WORK WITH IF N WE MAKE THE ELEMENTS IN Y UNCORRELATED IE O  1 Y .OW O h 9i ( ( 9 WE HAVE $ OO  $ 1 YY 1  ( WHICH IS THE IDENTITY MATRIX .OTE THAT 



 

WE DO AN EIGENVALUE DECOMPOSITION
( ( 1 ,1  4c49 

WHERE 4 CONTAINS THE EIGENVECTORS AND c  CH@F WITH THE EIGENVALUES 4HE QUADRATIC FORM IS NOW #  O9 4c49 O  P9 cP 
C 8

w a a a wC

 IS A DIAGONAL MATRIX

t(

wt JPt J 



WHERE P  49 O CONTAINS UNCORRELATED ELEMENTS SINCE 4 IS AN UNITARY MATRIX ;= 4HUS Pt ARE INDEPENDENT COMPLEX 'AUSSIAN VARIABLES WITH UNIT VARIANCE 3INCE THE REAL AND IMAGINARY PART OF Pt ARE INDEPENDENT wt JPt J IS   DISTRIBUTED WITH CORRESPONDING ,APLACE TRANSFORM 
wt R  ;= 4HE ,APLACE TRANSFORM OF A SUM OF INDEPENDENT VARIABLES IS THE PRODUCT OF THEIR INDIVIDUAL TRANSFORMS ;= AND HENCE h5 R    
wt R t(
( C 9



( ARE THE EIGENVALUES OF 1 , .OTE THAT THE EIGENVALUES wt OF 1 ,1 r s ` ( a ( ( ( CDS 1  ,1 ` w(  CDS 1 1 , ` w( 1 ` ( a ` ( a CDS 1 CDS 1 , ` w( CDS 1  CDS 1 , ` w(  



0ART  0ERFORMANCE !NALYSIS OF #ODED /&$- ON &ADING #HANNELS

5LTRASONIC )MAGING

0ART  %STIMATION OF THE 3PATIAL )MPULSE 2ESPONSE OF AN 5LTRASONIC 4RANSDUCER 5SING A 4OMOGRAPHIC !PPROACH

4HIS PART HAS BEEN PUBLISHED AS - 3ANDELL ! 'RENNBERG %STIMATION OF THE 3PATIAL )MPULSE 2ESPONSE OF AN 5LTRASONIC 4RANS DUCER 5SING A 4OMOGRAPHIC !PPROACH *OURNAL OF THE !COUSTICAL 3OCIETY OF !MERICA VOL  NO  PP   /CTOBER 



%STIMATION OF THE 3PATIAL )MPULSE 2ESPONSE OF AN 5LTRASONIC 4RANSDUCER 5SING A 4OMOGRAPHIC !PPROACH
!BSTRACT 4HE IMPULSE RESPONSE METHOD IS OFTEN USED TO STUDY THE ULTRASONIC ELD RADIATED BY PLANAR OR GENTLY CURVED TRANSDUCERS 4HESE INVESTIGATIONS CAN BE EXTENDED TO INCLUDE THE REECTED ECHOES FROM POINT REECTORS WHICH CAN BE USED IN SYNTESIZING ECHOES FROM REECTORS OF KNOWN SHAPES 4HE SINGLE POINT ECHO CAN BE MEASURED DIRECTLY BUT THERE ARE HOWEVER CERTAIN PROBLEMS ASSOCIATED WITH THIS 4HIS PAPER PROPOSES AN ALTERNATIVE METHOD TO ESTIMATE THE ECHO FROM A POINT LIKE REECTOR BY MEASURING THE ECHOES FROM SLIDING HALFPLANES 5SING A TOMOGRAPHIC APPROACH THE SINGLE POINT ECHO IS ESTIMATED FROM THESE MEASUREMENTS 4O PERFORM THE TOMOGRAPHIC INVERSION A TRUNCATED WEIGHTED 3INGULAR 6ALUE $ECOMPOSITION 36$ IS USED IN ORDER TO CATCH ESSENTIAL FEATURES OF THE ULTRASONIC ECHOES WHERE THE WEIGHT FUNCTIONS ARE CHOSEN SPECIALLY TO SUIT OUR APPLICATION .UMERICAL SOLUTIONS ARE GIVEN FOR THE CHOSEN WEIGHT FUNCTIONS AND AN ERROR ANALYSIS IS CARRIED OUT



)NTRODUCTION

4HE IMPULSE RESPONSE METHOD IS OFTEN USED TO STUDY THE ULTRASONIC ELD RADIATED BY PLANAR OR GENTLY CURVED TRANSDUCERS ;  = 4HESE INVESTIGATIONS CAN BE EXTENDED TO INCLUDE THE REECTED ECHO FROM A POINTLIKE REECTOR PLACED IN THE ULTRASONIC ELD 4HESE ECHOES CAN BE USEFUL IN PREDICTING THE ECHOES FROM REECTORS OF KNOWN SHAPES ;= 4HEY MAY ALSO BE OF USE IN IMPROVING THE PULSE ECHO IMAGING METHOD 4HIS METHOD MEASURES THE DISTANCE TO AN OBJECT BY ESTIMATING THE TIME OF IGHT 4/&  IE THE TIME BETWEEN TRANSMISSION OF AN ULTRASONIC PULSE AND THE ARRIVAL OF THE ECHO ; = 4HIS CAN BE DONE IN SEVERAL WAYS WHICH ARE DESCRIBED IN VARIOUS ARTICLES IN ; = (OWEVER THIS ASSUMES THAT THE TWO ECHOES HAVE THE SAME SIGNAL SHAPE !NY DIdERENCES IN THE SHAPES WILL DEGRADE THE 4/& ESTIMATE 4HIS MIGHT HOWEVER BE USED TO EXTRACT MORE INFORMATION IN THE ECHOES APART FROM THE ARRIVAL TIME SUCH AS THE SLOPE OF THE SURFACE AT THE MEASURED POINT 4WO ECHOES FROM A PLANE SURFACE WITH A SLOPING ANGLE OF p AND p RESPECTIVELY ARE SHOWN IN &IG  ILLUSTRATING THE FACT THAT THE SHAPE OF THE ECHO IS DEPENDENT ON THE ANGLE OF INCLINATION ;= 7HEN ULTRASOUND IS USED FOR MEASUREMENTS IN AIR A FOCUSED TRANSDUCER IS USUALLY USED DUE TO THE HIGH ATTENUATION )F THE REECTING OBJECT IS KEPT CLOSE TO THE FOCAL PLANE OF THE TRANSDUCER THE REECTING ECHO WILL ORIGINATE FROM A VERY SMALL AREA ON THE OBJECT 4HIS MIGHT BE MODELLED AS PLANE SLOPING SURFACE !PART FROM THE 4/& ESTIMATE WHICH GIVES THE AXIAL DISTANCE TO THE OBJECT AN ESTIMATE OF THE SLOPE MAY ALSO BE OBTAINED )F THIS ESTIMATE IS GOOD A SPARSER SCANNING GRID MAY BE USED WHILE MAINTAINING THE SAME ACCURACY OF THE ESTIMATED SURFACE PROLE "Y NEGLECTING NONLINEAR EdECTS IN THE MEDIUM AND THE TRANSDUCER THE ULTRASONIC ECHO



0ART  %STIMATION OF THE 3PATIAL )MPULSE 2ESPONSE

&IGURE  %CHOES FROM PLANE SURFACES WITH NORMALIZED AMPLITUDES 0ERPENDICULAR ABOVE AND p ANGLE OF INCLINATION BELOW  FROM A POINTLIKE REECTOR MAY BE MODELLED AS A LINEAR SYSTEM ;= SEE &IG  4HE TRANSDUCER IS ASSUMED TO BE CIRCULAR SYMMETRIC AND UNIFORMLY EXCITED &ROM THE RECI PROCITY PROPERTY ;= WE MAY PUT THE TRANSFER FUNCTIONS IN TRANSMISSION AND RECEPTION MODES EQUAL IE 3Q S  3I S  3 S AND GQ Q S  GI Q S  G Q S  4HE SPATIAL IMPULSE RESPONSE G Q S OF THE TRANSDUCER HAS BEEN CALCULATED ANALYTICALLY FOR A FOCUSED TRANSDUCER IN ;  = 4HE ELECTROMECHANICAL TRANSFER FUNCTION 3 S IS HOWEVER DIF CULT TO CALCULATE AND MUST BE ESTIMATED FROM MEASUREMENTS !T THE FOCAL POINT OF THE TRANSDUCER THE SPATIAL IMPULSE RESPONSE IS G  S  p S ` 1B  WHERE p a IS THE $IRAC DELTA FUNCTION 1 IS THE CURVATURE OF THE TRANSDUCER AND B IS THE SPEED OF SOUND IN THE MEDIUM

&IGURE  5LTRASONIC ECHO MODELLED AS A LINEAR SYSTEM 3INCE ALL ESTIMATED POINT ECHOES DEPEND ON THE FOCAL POINT ECHO IT IS ESSENTIAL THAT THE LATTER HAS HIGH ACCURACY 4O MEASURE THE FOCAL POINT ECHO DIRECTLY IS DIbCULT !N OBJECT WOULD HAVE TO BE VERY SMALL TO BE CONSIDERED POINTLIKE AND THIS WOULD RESULT IN A WEAK ECHO AND CONSEQUENTLY A LOW 3.2 )NCREASING THE SIZE WILL RESULT IN SPATIAL SMOOTHING AND DECONVOLUTION IS NECESSARY 4O AVOID THESE PRACTICAL LIMITATIONS WE PROPOSE AN INDIRECT METHOD INSTEAD BY USING A TOMOGRAPHIC APPROACH ;= 4HIS METHOD GIVES US AN INDEPEN DENT ESTIMATION OF ALL SINGLE POINT ECHOES IN THE ELD SIMULTANEOUSLY )N TOMOGRAPHY THE LINE INTEGRAL OF A TWO DIMENSIONAL FUNCTION IS NORMALLY MEASURED -ETHODS FOR INVERTING

 THESE MEASUREMENTS TO THE ORIGINAL FUNCTION ARE KNOWN AT LEAST IN THEORY ;= )N PRAC TICE RESTRICTIONS LIKE A NITE NUMBER OF PROJECTIONS LIMIT THE ACCURACY OF THESE INVERSIONS ;= (OWEVER WE CHOSE INSTEAD TO MEASURE THE ECHOES FROM HALFPLANES IE THE INTEGRALS OF THE LINE ECHOES 4HE REASON FOR THIS IS A BETTER 3.2 AND TO AVOID THE SPATIAL SMOOTHING THAT WILL OCCUR FROM A LINE REECTOR SINCE IT CAN NOT BE MADE INNITESIMAL SMALL




4HEORY
3INGLE POINT ECHOES

)N OUR EXPERIMENTS WE HAVE ASSUMED THAT OUR FOCUSED TRANSDUCER IS AXISYMMETRIC 4HE ASSUMPTION OF SYMMETRY IS FOR REASONS OF SIMPLICITY BUT THE METHOD CAN BE EXTENDED TO UNSYMMETRIC TRANSDUCERS 3UPPOSE THAT WE HAVE A PLANE WHICH IS PARALLEL TO THE FOCALPLANE AND IS LOCATED AT A HEIGHT Y FROM THE TRANSDUCER 4HE PLANE HAS A PERPENDICULAR DISTANCE O TO THE AXIS OF THE TRANSDUCER 4HE TRANSDUCER WORKS BOTH AS A TRANSMITTER AND A RECEIVER 4HE ECHO F FROM THE LINE WILL BE A FUNCTION OF DISTANCE O HEIGHT Y AND TIME S 5SING SUPERPOSITION WE CAN EXPRESS THE ECHO FROM A PLANE IN TERMS OF THE SINGLE POINT ECHOES 3INCE THE HEIGHT Y WAS CONSTANT IN OUR EXPERIMENT AND S IS ONLY A PARAMETER IN OUR CALCULATIONS WE WILL DROP THEM FROM NOW ON ,ET E Q DENOTE THE ECHO FROM A SINGLE POINT AT A DISTANCE Q FROM THE AXIS OF SYMMETRY 3INCE THE TRANSDUCER IS FOCUSED THE REECTION FROM A POINT FAR AWAY FROM THE AXIS OF SYMMETRY CAN BE NEGLECTED 4HIS MEANS THAT WE CAN PUT E Q   FOR JQ J   IN A NORMALIZED SCALE 7HAT SPATIAL DISTANCE Q   CORRESPONDS TO DEPENDS ON THE GEOMETRY OF THE TRANSDUCER AND HOW ACCURATE THE APPROXIMATION IS 4HE ECHO F O FROM A PLANE / WITH PERPENDICULAR DISTANCE O TO THE AXIS OF SYMMETRY WILL THEN BE THE INTEGRAL OF E Q OVER /  : F O 
G

: E Q CR 


* O Q E Q CQ



WHERE THE INTEGRATION AREA / IS DENED IN &IG  4HE INTEGRATION KERNEL * O Q IS

&IGURE  )NTEGRATION AREA FROM REECTING HALFPLANE IN THE FOCALPLANE



0ART  %STIMATION OF THE 3PATIAL )MPULSE 2ESPONSE

DERIVED IN !PPENDIX ! AND IS FOUND TO BE   {Q ` v O  `Q Q @QBBNR OQ `Q v O  Q   QvOv

* O Q 



4HE TRANSDUCER IS AXISYMMETRIC SO WE MAY ASSUME THE PLANE TO BE PARALLEL TO THE X AXIS FOR THE SAKE OF SIMPLICITY !S CAN BE SEEN FROM %Q   WE HAVE A MAPPING FROM THE SINGLE POINT ECHOES E Q TO THE HALFPLANE ECHOES F O VIA THE INTEGRAL KERNEL 3INCE THE PROJECTIONS ARE ALONG PLANES AND NOT LINES THIS IS A MODIED VERSION OF TOMOGRAPHY 5SING SOME TECHNIQUES FROM THAT AREA AN INVERSION FORMULA FOR THE SINGLE POINT ECHOES CAN BE CONSTRUCTED 4HERE IS SOME A PRIORI KNOWLEDGE OF E Q WHICH SHOULD BE USED 4HE WEIGHTED (ILBERT SPACES ' AND * IS RESPECTIVELY DENED AS '  + : < V9  WHICH CONTAIN THE SINGLE POINT ECHOES E Q  AND *  + :` < VB  WHICH CONTAINS THE HALFPLANE ECHOES F O  4HE SCALAR PRODUCT OF THESE SPACES ARE DENED AS : E  E 9  F  F B  :
   

E Q E Q V9 Q CQ F O F O VB O CO

 

4HE WEIGHT FUNCTIONS V9 Q AND VB O WILL BE DISCUSSED IN 3ECTION  4HUS WITH THE KERNEL IN %Q   AN INTEGRAL OPERATOR ! IS DENED MAPPING THE SINGLE POINT ECHOES E Q  ' TO THE HALFPLANE ECHOES F O  *  SEE &IG  F  !E !  '



&IGURE  (ALFPLANE ECHOES VIEWED AS A MAPPING BETWEEN (ILBERT SPACES





)NVERSION BY REGULARIZATION

)N ORDER TO ND THE SINGLE POINT ECHOES ONE MUST ND THE INVERSE OF THE OPERATOR ! OF %Q   4HIS EQUATION IS HOWEVER ILL POSED SINCE ! IS NOT CONTINUOUS IE THE SOLUTION E DOES NOT CONTINUOUSLY DEPEND ON F ;= 4HIS WILL CAUSE INSTABILITY SINCE A SMALL CHANGE IN F WILL RESULT IN A LARGE ERROR IN E  4HE GOAL IS TO APPROXIMATE THE SOLUTION E FROM A KNOWLEDGE OF A PERTURBED Fp WITH A KNOWN ERROR LEVEL KF p ` F KB v p  7HEN F BELONGS TO THE RANGE 1 '  F!E  E  ' G THEN THERE EXISTS A SOLUTION E OF !E  F  &OR A PERTURBED RIGHT HAND SIDE WE CANNOT EXPECT F p  1 ' IN GENERAL  5SING THE ERRONEOUS DATA F p  A REASONABLE APPROXIMATION E p TO THE EXACT SOLUTION E OF THE UNPERTURBED EQUATION !E  F IS SOUGHT /F COURSE THIS APPROXIMATION SHOULD BE STABLE IE E p DEPENDS CONTINUOUSLY ON THE ACTUAL DATA F p  4HEREFORE WE INTRODUCE THE APPROXIMATION OF THE UNBOUNDED INVERSE OPERATOR !  1 ' ' BY A BOUNDED LINEAR OPERATOR 2m  * '  ! FAMILY OF OPERATORS 2m  WITH THE REGULARIZATION PARAMETER m   IS CALLED A REGULARIZATION SCHEME ;= FOR THE OPERATOR ! IF
m

KHL 2m !E  E

E  '



4HE REGULARIZATION SCHEME APPROXIMATES THE SOLUTION E OF !E  F BY THE REGULARIZED SOLUTION

p  2 Fp Em m





3INGULAR VALUE DECOMPOSITION

! REGULARIZATION SCHEME 2m CAN BE CONSTRUCTED BY USING THE SINGULAR VALUE DECOMPOSITION 36$ OF THE OPERATOR ! &ROM THE 36$ TWO ORTHONORMAL SYSTEMS FEw G FFw G FOR '  * RESPECTIVELY ARE CONSTRUCTED ;= 4HESE TWO SYSTEMS ARE RELATED BY !Ew  }w Fw ! Fw  }w Ew  

WHERE ! IS THE ADJOINT OPERATOR OF ! AND }w ARE SINGULAR VALUES OF ! WHICH ARE ASSUMED TO BE ORDERED IE } w } w    4HE ADJOINT OPERATOR ! IS0 DENED FROM E ! F 9  !E F B  E  ' F  *  )T CAN BE SHOWN ;= THAT IF F  w @w Fw  THEN THE SPECTRAL CUT Od 2E F 
E 8  @w Ew } w w(



DESCRIBES A REGULARIZATION SCHEME WITH PARAMETER - AND K2E K  }E 



7EIGHTED 36$

"Y STUDYING THE 2ADON TRANSFORM ;= SUITABLE WEIGHT FUNCTIONS CAN BE DERIVED 4HE 2ADON TRANSFORM MAPS REAL VALUED FUNCTIONS IN 2E TO THEIR INTEGRALS OVER - `  DIMENSIONAL HYPERPLANES &OR -   THIS MEANS LINE INTEGRALS AND IN OUR CASE THE



0ART  %STIMATION OF THE 3PATIAL )MPULSE 2ESPONSE

2ADON TRANSFORM CORRESPONDS TO THE MAPPING BETWEEN THE POINT ECHOES TO THE LINE ECHOES )N ;= A COMPLETE SINGULAR SYSTEM FOR THE 2ADON TRANSFORM AS A MAPPING BETWEEN WEIGHTED (ILBERT SPACES HAS BEEN CALCULATED !SSUMING CIRCULAR SYMMETRY THE 2ADON TRANSFORM REDUCES TO

 2  + : < 4y
4y Q 5y O

+ :` < 5y



WHERE THE WEIGHT FUNCTIONS ARE  Q  ` Q  y  ` ay     ` O  

.OTE THAT AN EXTRA Q IS INTRODUCED IN 4y Q COMPARED TO ;= SINCE A CHANGE FROM #ARTESIAN TO POLAR COORDINATES IS IMPLIED 4HE PARAMETER y IS USED TO DETERMINE IF THE WEIGHT SHOULD BE LARGE AT THE BOUNDARY y   OR AT THE CENTRE y    &ROM THE NATURE OF THE ECHOES WE KNOW THAT THEIR AMPLITUDE DECREASE AS THE DISTANCE FROM THE AXIS OF SYMMETRY INCREASES "Y PUTTING A LARGE WEIGHT AT THE BOUNDARY WE CAN CREATE (ILBERT SPACES WHOSE FUNCTIONS WILL DECAY SUbCIENTLY FAST OTHERWISE THEIR NORM WOULD NOT BE NITE IE THEY WOULD NOT BELONG TO THE (ILBERT SPACE  4HIS WAY WE CAN CONSTRUCT TWO ORTHONORMAL BASES WHOSE BASE FUNCTIONS HAVE THE SAME SHAPE AS THE ECHOES &ROM THE MEASUREMENT THE DECAY OF THE ECHOES CAN BE ESTIMATED AND APPROPRIATE WEIGHT FUNCTIONS DESIGNED 3INCE OUR OPERATOR ! IS THE INTEGRAL OF THE 2ADON2TRANSFORM 2 WE DENE THE  WEIGHT FUNCTION FOR THE HALFPLANE ECHO SPACE * AS 6y O  y 5y W CW WHICH RESULTS IN THE HALFPLANE ECHO SPACE

 *  + :` < 6y



4HE CHOICE OF y MUST BE SUCH THAT E  ' !E  * )F THE HALFPLANE ECHOES DECAY AS  ` O WHEN O  IT IS SUbCIENT THAT y   `  ;= &OR THE TRANSDUCER USED IN THE EXPERIMENT SEE 3EC  WE FOUND {  AND WE MADE THE CHOICE y   4O SUMMARIZE WE DENE THE OPERATOR ! WHICH MAPS THE POINT ECHOES E Q  WHICH BELONG TO THE (ILBERT SPACE '  TO THE HALFPLANE ECHOES F O IN (ILBERT SPACE *  F  !E ` a !  + : <  4 WHERE ` a  Q  ` Q :   6 O  5 W CW  @QBBNR O  y ` a O P `  ` O  ` O
O ` O!  4 Q  ` a :` <  6
 

+

 








%RROR ESTIMATES
%RROR MODEL

!S WAS FOUND IN 3EC  THE REGULARIZATION SCHEME APPROXIMATES THE SOLUTION E OF !E  p  2 F p  "Y WRITING E p ` E  2 F p ` 2 F
2 !E ` E  F BY THE REGULARIZED SOLUTION Em m m m m m USING THE TRIANGLE INEQUALITY AND KF p ` F KB v p  AN ESTIMATE OF THE ERROR CAN BE WRITTEN AS
p ` E K v p K2 K
K2 !E ` E K KEm 9 m m 9



4HIS DECOMPOSITION ILLUSTRATES THE FACT THAT WE HAVE TWO TYPES OF ERROR THE RST TERM REECTS MAINLY THE INUENCE OF THE INCORRECT DATA AND THE SECOND TERM IS DUE TO THE APPROXIMATION ERROR BETWEEN 2m AND !  4HE LATTER WILL DECREASE AS m INCREASES BUT THE NORM K2m K WILL INCREASE UNBOUNDED SINCE 2m ! AS m  AND ! IS UNBOUNDED (ENCE THE CHOICE OF m WILL BE A TRADE Od BETWEEN APPROXIMATION ERROR AND DATA ERROR



$ATA ERROR

5SING THE SPECTRAL CUT Od METHOD THE MEASURED ECHO F p IS DECOMPOSED USING - TERMS IE FE 
E 8 w(

@w Fw



!SSUME THAT THE DATA ERROR CONSISTS OF ADDITIVE ZERO MEAN WHITE 'AUSSIAN NOISE T O WITH VARIANCE p   3INCE T O CAN NOT BE EXPECTED TO BELONG TO *  THE MEASURED ECHO Fp  F
T DOES NOT LIE IN *  )N ORDER TO MINIMIZE THE EdECT OF THE NOISE THE COEbCIENTS @w SHOULD BE CHOSEN SO THAT THE RESIDUAL F p ` F E IS MINIMIZED IN THE STANDARD NORM 4HE ORTHOGONALITY PRINCIPLE STATES THAT @w SHOULD BE CHOSEN SUCH THAT Fp `
E 8 w(

@w Fw  Ft

 

J        -



4HIS GIVES US THE FOLLOWING SYSTEM OF EQUATIONS   F  F  a a a F  FE  @ " # "         &@  !    FE  F  a a a FE  FE  @E  ` p a     F  F  F F  # " # " "      ! !
  ` p a F  FE  F FE   F
T

 # ! T F    


# ! 

T FE



0ART  %STIMATION OF THE 3PATIAL )MPULSE 2ESPONSE

&ROM &@  F
T IMPLYING @  & F
& T THE COVARIANCE MATRIX FOR @ IS FOUND TO BE H I 1dd  $ @`$ :@< @`$ :@< Q b c  $ & TTQ &  & 1 &  b Qc 4HE TERM Qrs IN THE COVARIANCE MATRIX 1 $ TT IS b c Qrs  $ T Fr  T Fs   v:  w :   $ T O Fr O CO T W Fs W CW   :   p  Fr O Fs O CO  p  Fr  Fs   4HIS MEANS THAT 1 p  & AND THAT 1dd p  &  4HE RECONSTRUCTED SINGLE POINT ECHO IS EE 
E 8  @w Ew  :E    EE <f :@    @E <Q } w( w

 E Q f @  ` a      }E   )T SHOULD BE NOTED HERE THAT E E IS NOT REALLY A SCALAR WHERE f DIAG } BUT A FUNCTION OF RADIUS Q 4HE EXPECTANCE AND VARIANCE OF THE RECONSTRUCTED SINGLE POINT ECHO E E ARE c b c b  $ E E  $ E Q f @  E Q f $ :@<  E Q f & F H` b c b ca I 6AR E E  $ E E ` $ E E H` aa I `  $ E Q f @ ` & F H` a ` Q   aQ I Q    $ E f & T E f & T b c  $ E Q f & TTQ &Q fQ E  E Q f & 1 & f E  p  E Q f & f E 

WHERE wMIN IS THE SMALLEST EIGENVALUE OF & 3INCE }E DECREASES FOR INCREASING - WE CAN EXPECT VARIANCE OF E E TO GROW FOR LARGER -  )N &IG  THE VARIANCE FOR -     IS PLOTTED 4HE ERROR VARIANCE ALSO DEPENDS ON THE SMALLEST EIGENVALUE wMIN OF & &OR THE STANDARD SCALAR PRODUCT IE THE WEIGHT FUNCTION VB O   IT IS ONE SINCE & BECOMES THE IDENTITY MATRIX 4HE MORE WEIGHT ONE PUTS ON THE FUNCTIONS AS O  THE LESS DIAGONAL THE MATRIX BECOMES AND THUS THE SMALLEST EIGENVALUE DECREASES

4HE VARIANCE CAN BE ESTIMATED BY ` aQ ` a b c 6AR E E  p  f E & f E p p v K f E K v K E K  wMIN wMIN }E





&IGURE  %RROR VARIANCE 1ii Q FOR DIdERENT VALUES OF - 



!PPROXIMATION ERROR

4HE SECOND PART OF %Q  IS THE APPROXIMATION ERROR 4O ANALYSE THE EdECT OF USING A REGULARIZED INVERSE THE DATA ERROR IS IGNORED AND ONLY THE APPROXIMATION ERROR IS TREATED 4HE REGULARIZATION PARAMETER IS -  THE NUMBER OF TERMS IN THE SPECTRAL CUT Od "Y USING E Ew 9   E ! Fw 9 }w    !E Fw B  F Fw B }w }w



THE ERROR DUE TO 2E CAN BE WRITTEN AS o o o2E F E ` E o 9 oE o  o8  o 8 o o  o @w Ew ` E Ew 9 Ew o o w( }w o w( 9 oE o  o8 @ ` F F o 8 o o w w B  o Ew

E Ew 9 Ew o o w( o }w w(E  9 V UE t u  8 U8 @w ` F Fw B  T 
E Ew  9 }w w( w(E 



4HE RST TERM IS AN ERROR WE MAKE WHEN NOT CHOOSING @w  F Fw B  3INCE FFw G w( IS AN ORTHONORMAL SET WITH RESPECT TO THE SCALAR PRODUCT IN *  THIS IS HOW THE OPTIMAL SET OF PARAMETERS F@w GE w( SHOULD BE CHOSEN TO MINIMIZE THE RESIDUAL IN THE * NORM "UT WE CHOSE TO MINIMIZE IT WITH RESPECT TO THE STANDARD NORM IN ORDER TO KEEP THE NOISE VARIANCE LOW SEE 3ECTION  4HE SECOND TERM OF %Q  IS THE RESIDUAL CAUSED BY PROJECTING F ONTO THE SUBSPACE SPANNED BY FFw GE w( 



0ART  %STIMATION OF THE 3PATIAL )MPULSE 2ESPONSE



%XPERIMENTS

7E HAVE MADE OUR EXPERIMENTS ON A BRASS BLOCK USING A FOCUSED TRANSDUCER FOR AIRBORNE ULTRASOUND 4HE TRANSDUCER WAS DESIGNED BY (ANS 7 0ERSSON ,UND )NSTITUTE OF 4ECH NOLOGY 3WEDEN FOR THE INVESTIGATION PRESENTED IN ;= AND IS ACOUSTICALLY ADAPTED FOR AIR )T HAS A RADIUS OF CURVATURE OF  MM A CROSS SECTION RADIUS OF  MM AND A CENTRE FREQUENCY  -(Z WHICH MAKES THE WAVELENGTH w   MM )N ORDER TO MEASURE F O  THE TRANSDUCER WAS PLACED JUST ABOVE THE EDGE OF A BRASS BLOCK AND MOVED OVER THE EDGE USING A COORDINATE TABLE 4O INCREASE THE 3.2 AN AVERAGING OF  ECHOES AT POSITION O WAS DONE (OWEVER THESE ECHOES CAN NOT BE AVERAGED IN THE USUAL WAY BECAUSE THEY ALL CONTAIN A SMALL RANDOM TIME SHIFT 4HIS TIME SHIFT IS CAUSED BY SMALL HETEROGENEITIES IN THE MEDIUM TEMPERATURE GRADIENTS AND AIR TURBULENCE IN OUR CASE  !N INVESTIGATION PERFORMED BY US SHOW THAT THERE ARE SMALL RANDOM TIME SHIFTS IN THE SIZE RANGE OF  `  NS IN THE ECHOES 4HE SAME PROBLEMS HAVE BEEN REPORTED IN ;= !VERAGING THE ECHOES IN THE NORMAL WAY WILL BOTH DISTORT THE SHAPE AND THE ARRIVAL TIME OF THE ECHO )NSTEAD AN AVERAGING PROCESS DESCRIBED IN ;= WAS USED 4HIS PROCESS ESTIMATES ALL THE TIME SHIFTS BETWEEN THE  ECHOES INTERPOLATES THE ECHOES SO THAT THEY ARE ALIGNED IN TIME AND THEN AVERAGES THEM !N AVERAGE OF THESE ECHOES WAS THEN STORED IN THE COMPUTER 4HE SAMPLING FREQUENCY WAS  -(Z AND WE TOOK  SAMPLES WHICH MAKES THE SWEEP LENGTH  xS 4HIS TIME IS SUbCIENT FOR THE ECHO TO DISAPPEAR ALMOST COMPLETELY "ETWEEN EACH MEASUREMENT OF A SINGLE ECHO WE MOVED THE TRANSDUCER  xM 4HE BRASS BLOCK WAS PLACED PERPENDICULAR TO THE TRANSDUCERS MOVEMENT 5SING THE DENITION OF THE FOCUS WIDTH AS THE DISTANCE BETWEEN THE TWO RST INTENSITY MINIMA IN THE FOCAL PLANE;= THE FOCUS WIDTH CWIDTH   MM  !FTER THE MEASUREMENTS WE FOUND THAT THE ECHO WAS APPROXIMATELY ZERO WHEN THE BRASS BLOCK WAS FURTHER AWAY THAN  MM FROM THE AXIS OF SYMMETRY .OTING THAT THIS IS MORE THAN THE DISTANCE TO RST INTENSITY MINIMA WHICH IS CWIDTH    MM WE NEGLECT THE CONTRIBUTION OF POINTS FURTHER AWAY 7E NORMALIZE THIS  MM DISTANCE TO O   IN OUR CALCULATIONS



2ESULTS

&OR THE CHOSEN WEIGHT FUNCTIONS DESCRIBED IN THE 3ECTION  THE 36$ WAS CALCULATED 3INCE THE %UCLIDIAN NORM IS NOT USED A MODICATION OF THE STANDARD 36$ WAS MADE TO NUMERICALLY CALCULATE THE SINGULAR SYSTEM SEE !PPENDIX " 4HE INTEGRAL OPERATOR ! WAS DIGITIZED INTO  POINTS IN Q AND  IN O IE a  ay   4HE RST  SINGULAR FUNCTIONS Fw  * AND Ew  ' ARE PLOTTED IN &IG  AND &IG  RESPECTIVELY 4HE INVERSION OF %Q  WAS THEN DONE BY USING THE SPECTRAL CUT Od METHOD "Y TRIAL AND ERROR WE FOUND A CHOICE OF THE REGULARIZATION PARAMETER -  IE THE NUMBER OF TERMS IN THE SUM IN %Q   TO BE  4HE INVERSION OF THE HALFPLANE ECHOES F O S WAS THEN PERFORMED IE ONE INVERSION FOR EACH VALUE OF S 4HE RESULTING SINGLE POINT ECHO E Q S IS PLOTTED AS A FUNCTION OF TIME IN &IG  FOR  DIdERENT VALUES OF Q  .OTE THAT THE AMPLITUDE IS NORMALIZED TO ENHANCE THE DIdERENCES IN SIGNAL SHAPE !LSO NOTE THAT POINTS AWAY FROM THE AXIS OF SYMMETRY YIELD ECHOES WHICH START EARLIER AND HAVE A



&IGURE  4HE RST  BASIS FUNCTIONS FOR (ILBERT SPACE ' 

&IGURE  4HE RST  BASIS FUNCTIONS FOR (ILBERT SPACE *  LONGER DURATION 4HIS IS CAUSED BY THE FACT THAT THESE POINTS WILL HAVE A DISTRIBUTION OF ULTRASONIC PATH LENGTHS THAT INCLUDE BOTH SHORTER AND LONGER ONES THAN THE FOCAL POINT WHICH IS EQUIDISTANT TO ALL POINTS ON THE SURFACE OF THE TRANSDUCER 4HE INTENSITY DISTRIBUTION IN THE FOCAL PLANE OF A FOCUSED TRANSDUCER CAN BE CALCULATED ANALYTICALLY ; = 4HE &OURIER TRANSFORM ' Q OF SPATIAL IMPULSE RESPONSE IN THE FOCAL PLANE IS FOUND IN ;= TO BE ' Q ` @ s (IfI(f ) @Q1B D 1 @Q1B 

WHERE ) a IS THE RST ORDER "ESSEL FUNCTION OF THE RST KIND 5NDER THE ASSUMPTION THAT THE ELD POINT INTENSITY IS PROPORTIONAL TO J' Q  J  WITH  BEING THE CENTER FREQUENCY AND USING THE RECIPROCITY PROPERTY ;= THE REECTED ENERGY FOR A NARROWBAND PULSE WILL



0ART  %STIMATION OF THE 3PATIAL )MPULSE 2ESPONSE

&IGURE  0OINT ECHOES AT DIdERENT RADIAL DISTANCES BE ( Q z J' Q



n n ) J zn n


n @Q1B n n n @Q1B 




)N &IG  THE REECTED ENERGY OF THE CALCULATED POINT ECHOES IS COMPARED TO %Q   !S CAN BE SEEN THESE TWO CURVES SHOW A SIMILAR SHAPE AND THE WIDTH OF THE FOCUS

&IGURE  2EECTED ENERGY FROM ELD POINT ECHOES ESTIMATED SINGLE POINT ECHOES SOLID AND THEORETICAL DASHED  CORRESPONDS WELL WITH THE CALCULATED VALUE IN 3EC 



3UMMARY

)N THIS PAPER WE HAVE DEALT WITH THE ECHO FROM A POINT LIKE REECTOR THE SINGLE POINT ECHO +NOWING THIS QUANTITY A MODEL OF THE ECHO FROM AN REECTOR OF ARBITRARY SHAPE

 CAN BE MADE !SSUMING THAT THE SUPERPOSITION PRINCIPLE ;= IS VALID THE ECHO CAN BE CONSIDERED AS THE SUM OF ALL THE ECHOES FROM POINTS THAT MAKE UP THE SURFACE 4HE SINGLE POINT ECHO FOR AN ULTRASONIC TRANSDUCER CAN BE MEASURED DIRECTLY BUT THERE ARE SOME PRACTICAL PROBLEMS ATTACHED TO IT 4HE REECTOR HAS TO BE VERY SMALL TO BE CONSIDERED A POINT LIKE REECTOR WHICH RESULTS IN A WEAK ECHO IE LOW 3.2 )F THE SIZE IS INCREASED ON THE OTHER HAND THERE WILL BE SPATIAL SMOOTHING /UR SOLUTION TO THIS WAS TO USE AN INDIRECT METHOD 4HE ECHOES FROM SLIDING HALFPLANES ARE MEASURED AND BY USING A TOMOGRAPHIC WAY OF REASONING THE SINGLE POINT ECHOES CAN BE ESTIMATED FROM THESE MEASUREMENTS !N OPERATOR ! MAPPING THE SINGLE POINT ECHOES FROM THE WEIGHTED (ILBERT SPACE ' TO THE HALFPLANE ECHOES IN THE WEIGHTED (ILBERT SPACE * WAS DENED 4HE REASON WHY A WEIGHT FUNCTION WAS USED IS THAT THE A PRIORI INFORMATION OF THE ECHOES SHOULD BE USED 4HESE WEIGHT FUNCTIONS WERE DETERMINED HEURISTICLY FROM METHODS USED IN TOMOGRAPHY ;= 4HE INVERSE OF THE OPERATOR ! IS NOT CONTINUOUS WHICH WILL RESULT IN A FATAL AMPLICATION OF DATA ERROR SUCH AS MEASUREMENT NOISE )NSTEAD WE CREATED A FAMILY OF REGULARIZED INVERSES 2m WHICH CAN BE MADE ARBITRARILY CLOSE TO ! AS THE REGULARIZATION PARAMETER m  3INCE THE RESULTING DATA ERROR INCREASES FOR INCREASING m THIS LEADS TO A TRADE Od BETWEEN APPROXIMATION ERROR AND DATA ERROR 4HE REGULARIZATION FAMILY USED WAS THE SPECTRAL CUT Od METHOD "Y DOING A SINGULAR VALUE DECOMPOSITION 36$ OF THE OPERATOR ! WITH RESPECT TO THE NORMS IN ' AND *  WE CAN CREATE AN ORTHONORMAL BASE FOR THE (ILBERT SPACES "Y PROJECTING THE MEASUREMENTS TO AN - DIMENSIONAL SUBSPACE OF *  SPANNED BY THE - RST BASIS FUNCTIONS WE CAN CONSTRUCT A REGULARIZED INVERSE TO ! THAT IS CONTINUOUS 4HE BASIS FUNCTIONS AND THE RESULTING SINGLE POINT ECHOES WERE SHOWN FOR THE CHOSEN WEIGHT FUNCTIONS

!

4HE INTEGRATION KERNEL

4HE ECHO FROM A HALFPLANE WITH PERPENDICULAR DISTANCE O   SEE &IG  FROM THE AXIS OF SYMMETRY CAN BE WRITTEN AS : :  : dffxy( F O  E Q QCQCt   QE Q CtCQ : 
G y
 

Q @QBBNR OQ E Q CQ



4HE RELATION F O
F `O  F `  F  CAN BE USED SINCE THE SUM OF TWO COMPLE MENTARY PARTS OF THE UNIT CIRCLE EQUALS THE WHOLE CIRCLE 4HUS FOR O   F O  F  ` F `O :  :  {   Q E Q CQ ` Q @QBBNR `OQ E Q CQ   y :  :   {QE Q CQ
Q @QBBNR OQ ` { E Q CQ  : y
 

y :

{QE Q CQ

Q @QBBNR OQ E Q CQ





0ART  %STIMATION OF THE 3PATIAL )MPULSE 2ESPONSE

4HE RESULT FROM %Q  AND %Q  CAN BE COMBINED TO : F O 
 

* O Q E Q CQ



WHERE THE INTEGRATION KERNEL IS DENED AS  {Q ` v O  `Q  Q @QBBNR OQ `Q v O  Q * O Q    QvOv



"

36$ WITH WEIGHTED NORMS

)N ORDER TO NUMERICALLY CALCULATE THE 36$ WE DIGITIZE THE INTEGRAL OPERATOR ! DESCRIBED IN 3EC  7E HAVE A NON %UCLIDIAN NORM SINCE THE INNER PRODUCTS IN THE SPACES ' AND * ARE DENED AS
Q 4E E  E 9  E  Q F  F B  F 6F

 

WHERE Er IS A - b  VECTOR AND Fs IS A , b  VECTOR CONTAINING SAMPLES FROM THE FUNCTIONS Er Q  ' AND Fs O  *  4HE SQUARE MATRICES 4 AND 6 ARE DIAGONAL AND CONTAINS SAM PLES FROM THE WEIGHT FUNCTIONS 4 Q AND 6 O RESPECTIVELY 3INCE THE WEIGHT FUNCTIONS ARE NON NEGATIVE 4( AND 6( EXIST .OW WE MAKE A CHANGE OF COORDINATES  4( E o  6( F 2EWRITING THE MAPPING OF E TO F 4 ( 4 (
 



6(

E      ! 

F 6( o o o

   

4HE 36$ OF THE MATRIX ! IS !  :o    oE <f:    D <Q  `fhQ 

WHERE f DIAG }  }    AND }w ARE THE SINGULAR VALUES OF ! 4HIS IS A STANDARD 36$ IE FOR THESE MATRICES WE HAVE hQ h  (D `Q `  (E  

 4HESE BASIS VECTORS WHICH ARE ORTHOGONAL WITH THE NORMAL %UCLIDIAN INNER PRODUCT CAN NOW BE TRANSFORMED TO OBTAIN BASIS VECTORS FOR ' AND *  $ENOTING THE BASIS VECTORS FOR THESE SPACES Ew AND Fw  WE HAVE FROM %Q  %  :E E    EE <  :4(  4(     4( E <  4( h &  :F F    FD <  :6( o 6( o    6( oD <  6( ` 4HESE MATRICES ARE ORTHOGONAL TO THE RESPECTIVE INNER PRODUCTS IN ' AND *  ` aQ ` a %Q 4%  4( h 4 4( h  hQ 4( 44( h  hQ h  (D ` ( aQ ` a 6 ` 6 6( `  &Q 6& 





 `Q 6( 66( `  `Q `  (E

 WITH RESPECT            

7HAT IS LEFT TO SHOW IS THAT Ew AND Fw REALLY ARE THE SINGULAR VECTORS OF TO THE NORMS IN ' AND *  4HE ADJOINT OPERATOR ! OF ! IS DENED AS E ! F 9  !E F B E  ' F  * 4HE ADJOINT OPERATOR OF IS FOUND FROM EQ 4

 

E Q 6F  E Q Q 6F Q6  4 Q 6

!w  }w ow ! ow  }w w WITH !  !Q WHICH IS THE CASE WITH %UCLIDIAN NORM  WE HAVE 6

(

5SING THE FOLLOWING RELATIONSHIP OF THE SINGULAR VECTORS

4 ( w 4 ( w
 

!w      Ew 

}w ow }w ow }w 6( ow }w Fw     }w w }w w }w 4( w }w Ew

AND 4  !Q ow 4( Q 6( ow Q 6 `6( o a w Fw

7E SEE THAT Ew AND Fw ARE THE SINGULAR VECTORS OF AND THAT THE SINGULAR VALUES ARE THE SAME AS FOR ! 4HIS MEANS THAT THE STANDARD ALGORITHMS FOR 36$ CAN BE USED TO CALCULATE THE SINGULAR SYSTEM OF WHEN THE NORMS ARE NON %UCLIDIAN



0ART  %STIMATION OF THE 3PATIAL )MPULSE 2ESPONSE

0ART  %STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES IN .ARROWBAND 5LTRASONIC %CHOES 5SING THE (ILBERT 4RANSFORM #ORRELATION

4HIS PART HAS BEEN PUBLISHED AS ! 'RENNBERG - 3ANDELL %STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES IN .ARROWBAND 5LTRASONIC %CHOES 5SING THE (ILBERT 4RANSFORM #ORRELATION )%%% 4RANSACTIONS ON 5LTRASONICS &ERROELECTRICS AND &REQUENCY #ONTROL VOL  NO  PP   3EPTEMBER 



%STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES IN .ARROWBAND 5LTRASONIC %CHOES 5SING THE (ILBERT 4RANSFORM #ORRELATION
!BSTRACT )N MANY AREAS THE TIME DELAY OF ARRIVAL 4$/! IS DESIRED )N THE CASE OF NARROWBAND SIGNALS WE PROPOSE A FAST AND SIMPLE METHOD TO ESTIMATE SMALL TIME DELAYS 4HIS METHOD IS SHOWN TO HAVE THE SAME OR BETTER ACCURACY AS THE CROSS CORRELATION METHODS FOR SMALL DELAYS IN THE ORDER OF FRACTIONS OF THE SAMPLE INTERVAL )T IS BASED ON USING THE (ILBERT TRANSFORM IN CORRELATION BETWEEN TWO SIGNALS AND CONSISTS OF ONLY ONE SCALAR PRODUCT WHICH MAKES IT FAST )T MAY ALSO BE USED IN APPLICATIONS WITH NARROWBAND SIGNALS WHERE THE MEASUREMENTS ARE REPEATABLE SUCH AS ULTRASONIC IMAGING AND NONDESTRUCTIVE TESTING )N ULTRASONIC APPLICATIONS DUE TO UCTUATIONS IN THE INSONIED MEDIA A SMALL RANDOM TIME SHIFT MAY BE PRESENT CAUSING THE SIGNALS TO BE MISALIGNED IN TIME !VERAGING SIGNALS UNDER THESE CONDITIONS WILL RESULT IN A DISTORTION OF THE SIGNAL SHAPE 7E PROPOSE AN AVERAGING METHOD TO AVOID THIS AND TO ACCOMPLISH A HIGHER 3.2 WITHOUT THE DISTORTION 3IMULATIONS AND EXPERIMENTS FROM ULTRASONIC APPLICATIONS ARE PRESENTED



)NTRODUCTION

4HE TIME DELAY OF ARRIVAL 4$/! IS REQUIRED IN MANY AREAS EG ULTRASONIC IMAGING ;= %STIMATORS OF THE 4$/! ARE DESIGNED IN DIdERENT WAYS AS CAN BE SEEN IN ;  = AND VARIOUS ARTICLES IN ; = BUT THEY HAVE A FEW THINGS IN COMMON 5SUALLY THEY HAVE AN 3.2 THRESHOLD UNDER WHICH THEY PERFORM SUBSTANTIALLY LESS WELL ;= &OR CORRELATOR BASED ESTIMATORS THIS IS DUE TO PEAK AMBIGUITY IE THE POSSIBILITY OF SELECTING THE WRONG PEAK OF THE CORRELATION 4HESE ESTIMATORS ALSO NEED SOME KIND OF INTERPOLATION TO OBTAIN SUBSAMPLE RESOLUTION EG A PARABOLIC T ;  = )N ;= A COMPARISON OF SEVERAL METHODS IS MADE ! SPECIAL CASE OF TIME DELAY ESTIMATION IS WHEN THE SIGNALS ARE NARROWBAND ;= 5NDER THIS ASSUMPTION WE PROPOSE A NOVEL METHOD TO ESTIMATE THE 4$/! FOR SMALL TIME DELAYS 4HIS METHOD WILL BE SHOWN TO BE FAST AND SIMPLE !PART FROM THE INCREASED RESOLUTION IN 4$/! IT CAN BE USED FOR OTHER PURPOSES /NE SUCH APPLICATION IS THE PROBLEM OF AVERAGING )N AREAS WHERE THE MEASUREMENTS ARE REPEATABLE SUCH AS ULTRASONIC IMAGING AND NONDESTRUCTIVE EVALUATION .$%  THE PROCESS OF AVERAGING CAN BE USED TO IMPROVE THE 3.2 )F THE USEFUL PART OF THE SIGNAL IS THE SAME FOR ALL MEASUREMENTS IT WILL REMAIN THE SAME AFTER THE AVERAGING WHILE THE DISTURBANCES EG NOISE AND ARTIFACTS WILL BE ATTENUATED SINCE THEY OCCUR RANDOMLY IN EACH SEPARATE MEASUREMENT (OWEVER IF THE USEFUL PART OF THE SIGNAL HAS A SMALL RANDOM TIME SHIFT DUE TO PHYSICAL REASONS OR INACCURACIES IN THE MEASUREMENT EQUIPMENT THE AVERAGING WILL DISTORT THIS SIGNAL 4HIS PHENOMENON CAN APPEAR EG IN ULTRASONIC IMAGING !IRBORNE ULTRASOUND SUdERS FROM INHOMOGENEITIES IN THE MEDIUM SUCH AS TEMPERATURE GRADIENTS AND AIR TURBULENCE ;= 5SING THE ABOVE MENTIONED METHOD WE CAN ESTIMATE THESE SMALL



0ART  %STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES

RANDOM TIME SHIFTS AND DESIGN AN AVERAGING ALGORITHM THAT TIME ALIGNS IE CAUSES THE USEFUL PART OF THE SIGNALS TO HAVE THE SAME POSITION IN TIME SO THAT VERY LITTLE SIGNAL SHAPE DISTORTION OCCURS WHEN THE AVERAGING IS CARRIED OUT



$ERIVATION OF THE ESTIMATOR

7E MODEL ALL RECEIVED ECHOES AS HAVING THE SAME SIGNAL SHAPE AND ONLY DIdERING IN THE ARRIVAL TIME 4HUS USING Q S AS A REFERENCE ECHO WE ND THAT THE RECEIVED ECHO IS R S  Q S`t 

WHERE THE TIME DELAY t IS CONSIDERED TO BE AN UNKNOWN DETERMINISTIC PARAMETER )N MANY TIME DELAY ESTIMATORS ;  = THE CROSS CORRELATION :  R S Q S
~ CS  1 ~ 

BETWEEN THE REFERENCE ECHO Q S AND THE RECEIVED ECHO R S IS CONSIDERED 4HE TIME LAG MAXIMIZING THIS CROSS CORRELATION IS THE ESTIMATED TIME DELAY (OWEVER #ABOT ;= SUGGESTED THAT THE CROSS CORRELATION BETWEEN R S AND Q v S  THE (ILBERT TRANSFORM OF Q S  SHOULD BE USED INSTEAD 4HIS CROSS CORRELATION IS SHOWN TO BE v 1 v ~  1 ~ 

)N THIS CASE THE CROSS CORRELATION WILL NOT HAVE A MAXIMUM AT THE TIME LAG t BUT A ZERO CROSSING )T CAN BE ARGUED THAT IT IS EASIER TO ND A ZERO CROSSING THAN A PEAK IN A NOISY SIGNAL )T INTRODUCES HOWEVER AN AMBIGUITY SINCE THERE CAN BE SEVERAL ZERO CROSSING PARTICULARLY IF THE CROSS CORRELATION IS NARROWBAND 4HE (ILBERT TRANSFORM IS DENED AS ;=  Q v S  ( FQ S G  { :

 QW CW  G S  W ` S

cQ S



WHERE THE INTEGRAL IS A #AUCHY 0RINCIPAL 6ALUE #06 AND c DENOTES CONVOLUTION 4HE  (ILBERT KERNEL IS DENOTED BY G S  ` {  7E OBSERVE THAT IF Q S  + `  THEN IT  FOLLOWS THAT Q v S  + `   SEE ;= #ONSIDER THE CROSS CORRELATION BETWEEN R S AND Q v S AT TIME LAG ZERO AND USE 0ARSEVALS THEOREM :  :  1  Q S`t Q v S CS  Q S ` t :G S c Q S < CS v    :  :   Ds {it 1 E ' E 1 E CE  Ds {it ' E J1 E J CE 

(ERE WE LET ' E  I RFM E DENOTE THE &OURIER TRANSFORM OF G S ;= !SSUMING THAT Q S IS NARROWBAND IE ITS ENERGY IS CONCENTRATED IN FREQUENCY INTERVALS OF SIZE ! AROUND

 fE  AND THAT !t ~  1 v  CAN BE APPROXIMATED AS 1 v  { { { : i 3(



i3( i 3( : i 3( : i 3( s {i s {i t t J1 E J CE
ID J1 E `ID i3( i 3( a $ ` s {i t
Ds {it  `$ RHM {E t  `$ RHM I `D
         

Ds {it ' E J1 E J CE

: i 3(

Ds {it ' E J1 E J CE J CE


2 WHERE $   Q S CS IS THE ENERGY OF Q S AND   {E  7E HAVE ASSUMED THE EXPONENTIAL FUNCTION TO BE CONSTANT OVER THE INTEGRATION INTERVAL 4HIS GIVES US 2 `  R S Q v S CS `1 v  RHM  t {  2   $  Q S CS 4HIS ESTIMATOR CAN BE EXPECTED TO WORK WELL ONLY FOR NARROWBAND SIGNALS AND SMALL TIME DELAYS t 4HE ASSUMPTION THAT Ds {it IS CONSTANT OVER THE INTEGRATION INTERVAL WILL NOT BE VALID OTHERWISE !S WE WORK WITH SAMPLED SIGNALS WE GIVE THE DISCRETE VERSION OF   4HE OBSERVATION INTERVAL 3  ASSUMED TO BE LARGER THAN THE DURATION OF THE ECHOES IS SAMPLED AT - POINTS WITH INTERVAL 3  3 -  $ENOTING THESE SAMPLES Rw  R M3  WE MAY APPROXIMATE THE INTEGRALS IN  WITH SUMS AND YIELD B t 




@QBRHM

2Q 2Q
 

RSQ v S CS Q  S CS




@QBRHM

`3

0E  Rt Q vt ( 0t E   3 t( Qt



b cQ b cQ 5SING VECTORS AND DENOTING THEM R  R R a a a RE   v Q Q v Q v a a a Q vE  b cQ AND Q  Q Q a a a QE   WE HAVE B t 



RQ v Q @QBRHM ` Q Q Q

u 

3INCE Q S IS A REFERENCE ECHO WE MAY CALCULATE `v QQQ Q  A ONE TIME FOR ALL AND ARRIVE AT THE FAST AND SIMPLE TIME DELAY ESTIMATOR ` a B   @QBRHM RQ A t




)F THE ANGULAR CENTRE FREQUENCY  IS NOT KNOWN IT MAY BE ESTIMATED BY INDUCING OUR OWN TIME DELAY )NTRODUCING A CONSTRUCTED SIGNAL W S  Q S ` 3  WE HAVE t  3  )N b c Q Q VECTOR NOTATION THIS IS EQUAL TO W   Q  &OR THIS KNOWN t WE CAN REWRITE  TO ESTIMATE  t v wu b c A  Q  Q B  @QBRHM   3



0ART  %STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES




0ERFORMANCE OF THE ESTIMATOR

4HE INUENCE OF BANDWIDTH

)N THE DERIVATION OF THE ESTIMATOR WE ASSUMED AN INNITESIMALLY SMALL BANDWIDTH 4HIS IS OF COURSE NOT THE CASE FOR AN ARBITRARY NARROWBAND SIGNAL 4O ANALYZE THE INUENCE OF BANDWIDTH WE ASSUME THAT THE Q S HAS THE FOLLOWING &OURIER TRANSFORM Q 6 JE f E J  !  3 1E    OTHERWISE 4HIS SIGNAL HAS ENERGY $ AND SINGLE SIDED BANDWIDTH !  )N !PPENDIX ! WE SHOW THAT THE BANDWIDTH INTRODUCES A BIAS IN THE FORM OF A MULTIPLICATIVE TERM 2 `  R S Q v S CS B 2  RHM  t   RHMB !t RHM  t  Q S CS 7E SEE THAT IN ORDER TO KEEP THIS BIAS LOW THE PRODUCT OF BANDWIDTH AND TIME DELAY MUST BE SMALL 7E ASSUME THAT THE DELAY IS SMALL JtJ   i  RESULTING IN J  tJ  { (ENCE 3  WE HAVE J!tJ v i  n WHERE n  E !  &OR NARROWBAND SIGNALS n IS IN THE ORDER OF   CAUSING THE BIAS FACTOR RHMB !t TO BE CLOSE TO 



.OISE ERROR

!SSUME THAT Q S AND R S ARE CORRUPTED BY INDEPENDENT ADDITIVE WHITE 'AUSSIAN NOISE 3INCE BOTH Q S AND R S ARE BANDLIMITED WE CAN ALSO ASSUME THAT THE NOISE HAS A LIMITED BANDWIDTH )F THIS IS NOT THE CASE THE RECEIVED SIGNALS CAN ALWAYS BE LOWPASS LTERED WITHOUT ANY LOSS OF INFORMATION .OTE THAT THE (ILBERT TRANSFORM OF A WHITE 'AUSSIAN PROCESS M S IS ALSO WHITE AND 'AUSSIAN
 2w v E  2w E J' E J  2w E



(ERE WE LET 2w E DENOTE THE POWER SPECTRAL DENSITY OF THE STOCHASTIC PROCESS M S  IE THE &OURIER TRANSFORM OF THE AUTOCORRELATION FUNCTION 1w ~  )F WE INCLUDE THE NOISE IN  WE GET 2Q `  R S
M S Q v S
M v  S CS  B t @QBRHM  2Q   Q S
M S CS   WHERE THE REFERENCE ECHO IS Q S
M S AND THE RECEIVED ECHO R S
M S  4HE SUPERSCRIPT v a DENOTES (ILBERT TRANSFORMATION AND CAN NUMERICALLY BE CALCULATED IN THE &OURIER DOMAIN USING &&4 ;= &ROM THE DISCRETE VERSION OF THE ESTIMATOR   WE HAVE 0E  t u ` 3 R
M Q v
M v   ! t t t  t t ( B t @QBRHM  @QBRHM  0E   3 t( Qt
Mt   4HE SAMPLED PROCESSES Mt AND Mt ARE WHITE AND 'AUSSIAN ;= WITH VARIANCE }  3INCE OUR ESTIMATOR IS NONLINEAR IT IS DIbCULT TO ANALYZE (OWEVER THE MEAN IS MUCH LARGER

 THAN THE STANDARD DEVIATION FOR THE DENOMINATOR AND WE MAY LINEARIZE THE ESTIMATOR IN   SEE !PPENDIX " TO OBTAIN t u N O  x3 B  $ t { @QBRHM x2  |r t t u u s }   x  x3 }3
x }2 2 3 B $ t`t {
@QBRHM `t     x x2  2 x2 ` x3  WHERE x2 x3  }2  }3     $
3 }  $ RHM  t $ 3 }
3 3 } $ 3 }
3 3 }  u


   

4HE MEAN OF THE ESTIMATE IS t u t N O  $ RHM  t  2-1 B RHM $ t  @QBRHM  @QBRHM  $

3 } 2-1
   WHERE WE DENE 2-1  $ 3 }





4HIS DENITION OF THE 3.2 IS THE RATIO BETWEEN THE ENERGIES OF THE SIGNAL AND THE NOISE IN THE OBSERVED TIME INTERVAL )T IS EQUIVALENT TO THE DENITION IN ;= WHICH IS 2-1  1  }  WHERE 1 ~  $ FQ S Q S
~ G &OR AN ERGODIC STOCHASTIC PROCESS 1 ~ 2Q  CAN BE APPROXIMATED WITH Q Q S Q S
~ CS IF THE OBSERVATION INTERVAL IS LONG ENOUGH  (ENCE FOR LARGE 3 THE DENITIONS OF 2-1S ARE EQUIVALENT !S WE SEE FROM   THE ESTIMATOR CAN BE EXPECTED TO BE ASYMPTOTICALLY UNBIASED FOR LARGE 2-1 BUT CONTAIN A BIAS FOR SMALL 2-1 !S FOR THE ERROR VARIANCE IT CAN BE SHOWN FROM  THAT |r s }  B` t KHL $ t 
t   PEI D" -  |r s } B` t KHL $ t    PEI D" &ROM THIS WE SEE THAT THE STANDARD DEVIATION OF THE ERROR FOR VERY LOW 2-1 WILL BE APPROXIMATELY EQUAL TO THE DELAY t 'ENERALLY FOR AN ARBITRARY 2-1 WE WILL HAVE AN INCREASE IN ERROR VARIANCE FOR LARGER t 



.UMERICAL SIMULATIONS

)N ORDER TO VERIFY THE PERFORMANCE GIVEN BY THE PREVIOUS SECTION WE SIMULATED  NUMERICALLY BY USING A SYNTHETIC ECHO IN THE FORM OF Q S  Dd BNR



S

S3





0ART  %STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES

&IGURE  3YNTHETIC ECHO USED IN SIMULATIONS WITH THE PARAMETERS @   S    {! RADS S   a ! S AND THE SAMPLE FREQUENCY %   -(Z WHICH CORRESPONDS TO OUR EQUIPMENT 4HE NUMBER OF SAMPLES WERE -   CORRESPONDING TO A MEASURING TIME OF 3   xS 4HE SYNTHETIC ECHO IS SHOWN IN &IG  3INCE WE USED %   -(Z AND THE CENTRE FREQUENCY E   -(Z WE HAVE  SAMPLES PER PERIOD IN THE ECHO &OR EACH VALUE OF THE 3.2 WE SIMULATED  ECHOES AND CALCULATED THE MEAN OF THE ESTIMATOR   7E ADDED NOISE TO BOTH Q S AND R S  4HE SIMPLIED EXPRESSION OF THE MEAN  WAS COMPARED TO THE MEAN OF THE SIMULATION )N &IG  WE SHOW THESE CURVES AS A FUNCTION OF 3.2 WITH t  3  &OR

&IGURE  3IMULATED SOLID AND THEORETICAL DASHED MEAN OF THE ESTIMATOR FOR t  3 

 B IS COMPARISON WE GIVE THE ERROR VARIANCE FOR #ABOTS ESTIMATOR ;= IE THE ESTIMATE t THE INTERPOLATED ZEROS CROSSING OF 1 E ~  |r s }  B $ t#ABOT ` t   !w 3 o4 A WHERE A   o4  2 J{E J J2 E J CE  2  J2 E J CE
 

 

o
 o 

AND o  Q6 } IS THE 3.2 FOR Q S AND !w  %  IS THE SINGLE SIDED BANDWIDTH OF M S  .OTE THAT FOR A NARROWBAND SYSTEM WE HAVE A {   4HE ERROR VARIANCE IN  CAN BE REWRITTEN AS |r s }  2-1
 B#ABOT ` t  $ t  - 2-1 A

$UE TO APPROXIMATIONS IN ;= THIS ERROR VARIANCE IS HOWEVER ONLY VALID FOR 2-1  ` D" IN OUR SIMULATION )N &IG  THE SIMULATED AND THEORETICAL ERROR VARIANCE OF OUR ESTIMATOR ARE PLOTTED TOGETHER WITH #ABOTS 4HE CURVES ARE HARD TO TELL FROM EACH OTHER FOR 2-1   D" 4HE REASON WHY THE ERROR VARIANCE FOR THE ESTIMATOR  BELOW THIS

&IGURE  3IMULATED SOLID AND THEORETICAL DASHED ERROR VARIANCE OF THE ESTIMATOR FOR t  3  4HESE ARE HARD TO DISTINGUISH THE DOTTED CURVE IS THE ERROR VARIANCE OF #ABES ESTIMATOR 2-1 IS LESS THAN #ABOTS IS THAT IT BECOMES BIASED FOR SMALL 2-1 AND HENCE NOT QUITE COMPARABLE



0ART  %STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES



!PPLICATION EXAMPLE AVERAGING ALGORITHM

5SING THE ESTIMATOR FOR SMALL TIME DELAYS   WE CAN PROCEED TO FORMULATE AN AVERAGING ALGORITHM 4HIS METHOD BECOMES USEFUL WHEN THE ECHOES HAVE SMALL RANDOM TIME SHIFTS 4HESE COULD ORIGINATE FROM UCTUATIONS IN THE INSONIED MEDIUM )N OUR LINE OF RESEARCH WHICH CONTAINS SURFACE PROLING BY USING AIRBOURNE ULTRASOUND ;= THESE UCTUATIONS TAKE THE FORM OF TEMPERATURE VARIATIONS AND AIR TURBULENCE ! CHANGE IN THE AIR TEMPER ATUR OF p # RESULTS IN A  MS CHANGE OF SOUND SPEED ;= 5SING OUR TRANSDUCER WITH A RADIUS OF CURVATURE OF  MM IE A TOTAL PATH LENGTH OF  MM FOR A SURFACE IN THE FOCAL PLANE THIS MEANS A TIME DELAY OF  NS 3 IN OUR CASE WHERE 3 IS THE SAMPLE INTERVAL 4HESE SMALL RANDOM TIME SHIFTS HAVE TO BE TAKEN INTO ACCOUNT WHEN AN AVERAGING PROCESS IS APPLIED /THERWISE WE WILL INTRODUCE A DISTORTION OF THE SIGNAL DUE TO THE MISALIGNMENT OF THE ECHOES !SSUME THAT WE WANT TO AVERAGE OVER , ECHOES Rr S  Q S ` tr
Mr S H  aaa, `   WHERE WE DENE R S  Q S
M S  )N   tr ARE SMALL INDEPENDENT RANDOM TIME SHIFTS AND Mr S WHITE 'AUSSIAN ADDITIVE NOISE WHICH ARE INDEPENDENT OF tr  4HE RST THING IS TO ESTIMATE THE INDIVIDUAL TIME DELAYS BETWEEN THE ECHOES 4HIS CAN BE DONE AS DESCRIBED IN 3ECTION  7E USE THE RST ECHO R S AS THE REFERENCE ECHO 4O BE ABLE TO AVERAGE OVER THE , ECHOES WE NEED TO ALIGN THE ECHOES IN TIME 4HIS HAS BEEN DONE BY USING AN INTERPOLATION FORMULA WHICH IS APPROXIMATELY AN ALL PASS LTER WITH LINEAR PHASE CORRESPONDING TO THE TIME DELAY  3INCE WE HAVE SAMPLES OF THE ECHOES THE PROBLEM IS EXPRESSED AS HAVING Rr M3 AND WANTING Rr M3
tr  7E USE THE SAME NOTATION AS BEFORE IE THE SAMPLES ARE Rrw  Rr M3  ! SIMPLE INTERPOLATION METHOD IS ,AGRANGES FORMULA ;= OF THE ORDER -
 E W
Sa 
E 8 t( E

WHICH USES THE FACT THAT E W IS KNOWN IN THE EQUALLY SPACED POINTS Wt  W ` J a 4HE WEIGHTS Gt DEPEND ON S WHICH IS THE FRACTION OF THE DISTNACE a FOR WHICH WE WANT TO ND E  4HE INTERPOLATION  CAN BE SEEN AS A LTER WITH TAPS Gt  7E CAN NOW ALIGN Rrw WITH R BY FORMING Rrw c Gw tr WHERE c DENOTES CONVOLUTION "Y DOING THIS FOR ALL ECHOES Rrw WE CAN AVERAGE OVER THE , ECHOES
D   8  Rw  R , r( rw

Gt S E W ` J a





4HE AVERAGED ECHO Rw WILL HAVE THE SAME POSITION IN TIME AS Rw SINCE THIS IS USED AS A REFERENCE ECHO )N ORDER TO ND AN ABSOLUTE POSITION IN TIME WE AVERAGE OVER ALL ESTIMATED TIME DELAYS
D   8B t tr , r(



4HIS GIVES US AN ESTIMATION OF THE CORRECT RELATIVE POSITION IN TIME OF Rw  4O OBTAIN THE NAL AVERAGED SIGNAL WE INTERPOLATE WITH THE ,AGRANGE INTERPOLATION LTER WITH TAPS Gt WHICH ARE DETERMINED BY t





%XPERIMENTS

!PART FROM THE DIbCULTIES OF DENING SIGNAL SHAPE DISTORTION THERE ARE ALSO PRACTICAL PROBLEMS ! NOISE LESS REFERENCE ECHO IS USUALLY NOT AVAILABLE AND THUS IT IS DIbCULT TO MEASURE THE EdECT OF THE SIGNAL SHAPE DISTORTION )NSTEAD WE CHOSE TO GET ANOTHER QUANTITATIVE MEASURE OF THE PERFORMANCE OF THE ALGORITHM /NE AREA WHERE IT CAN BE USED IS AIRBORNE ULTRASOUND 7E USED A FOCUSED TRANSDUCER WITH A RADIUS OF CURVATURE OF  MM A DIAMETER OF  MM AND A CENTRE FREQUENCY OF APPROXIMATELY  -(Z )T WAS DESIGNED BY (7 0ERSSON ,UND )NSTITUTE OF 4ECHNOLOGY FOR THE INVESTIGATION PRESENTED IN ;= AND IS ACOUSTICALLY ADAPTED FOR AIR 7E APPLIED IT TO A PHANTOM IN THE FORM OF A MASSIVE BRASS BLOCK WITH SMOOTH SURFACES 4HE BLOCK WAS PLACED PERPENDICULAR TO THE TRANSDUCER THE POSITION OF WHICH WAS XED THROUGHOUT THE WHOLE EXPERIMENT 7E REGISTERED  ECHOES WHICH TOOK ABOUT  SECONDS WITH OUR EQUIPMENT 4HESE ECHOES WHICH SHOULD BE IDENTICAL SINCE THEY WERE MEASURED AT THE SAME SPOT WERE THEN INVESTIGATED ! MEASURED ECHO IS SHOWN IN &IG  AND THE 2-1 WAS ESTIMATED TO BE APPROXIMATELY  D"

&IGURE  -EASURED ULTRASONIC ECHO FROM THE PHANTOM



2ESULTS

)F WE PLOT A CROSS SECTION OF THE REGISTERED ECHOES IE FOR A XED VALUE OF THE TIME S WE PLOT WE HOW THE AMPLITUDE VARIES FOR THE DIdERENT MEASUREMENTS WE CAN SEE HOW THE MISALIGNMENT EdECT CAN CAUSE PROBLEMS WHEN WE AVERAGE THE ECHOES 4HIS CROSS SECTION PLOT IS SHOWN IN &IG  )F THE ECHOES DID NOT HAVE A RANDOM TIME SHIFT THESE CURVES WOULD BE STRAIGHT HORIZONTAL LINES APART FROM THE MEASUREMENT NOISE 4HE UPPERMOST CURVE CORRESPONDS TO A POINT IN TIME WHERE THE REFERENCE ECHO HAS A PEAK IE SMALL DERIVATIVE AND THE LOWERMOST TO A POINT IN TIME WHEN THE REFERENCE ECHO HAS A ZERO



0ART  %STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES

&IGURE  #ROSS SECTION IF THE MEASURED ECHOES FOR TWO VALUES OF S 4HE LOWERMOST CURVE CORRESPONDS TO S  S WHICH IS A ZERO CROSSING 4OTAL TIME FOR THE  MEASUREMENTS WAS  SECONDS CROSSING IE LARGE DERIVATE ,OOKING AT THE ESTIMATIONS OF THE TIME SHIFTS RELATIVE TO THE REFERENCE IE RST ECHO &IG  WE CAN SEE THAT IT HAS THE SAME DRIFT AS THE CROSS SECTION 4HE MAGNITUDE OF THE DERIVATIVE OF THE ECHO IN &IG  AT THE ZERO CROSSINGS IS

&IGURE  %STIMATED TIME SHIFTS RELATIVE TO THE RST ECHO FOR THE MEASUREMENTS IN &IG  ESTIMATED TO BE ALMOST  6xS &ROM &IG  AND &IG  WE SEE THAT THE DELAY OF ONE SAMPLE IE  NS AT A ZERO CROSSING RESULTS IN AN AMPLITUDE DIdERENCE OF APPROXIMATELY  6 )F WE LINEARIZE THE ECHO AT THE ZERO CROSSING WE GET AN AMPLITUDE OF  6xS b  NS   6 &ROM &IG  WE CAN SEE THAT IT HAS A LOW FREQUENCY SHAPE WHICH CORRESPONDS TO THE UCTUATIONS IN THE MEDIUM ;= THAT APPEAR DURING THE TOTAL MEASUREMENT TIME OF  SECONDS 4O SEE THE EdECTS OF USING OUR AVERAGING ALGORITHM WE LOOK AT THE CROSS SECTION VARIANCE OF R S FOR EACH POINT OF TIME IE U S  5@Q FR S G 7E COMPARE THE VARIANCE FOR THE ORIGINAL ECHOES WITH THE ONE WHERE WE ESTIMATED THE INDIVIDUAL TIME SHIFTS AND

 INTERPOLATED THE ECHOES IN ORDER FOR THEM TO BE TIME ALIGNED WITH THE RST ECHO !S WE CAN SEE IN &IG  THE ALIGNMENT IS VERY GOOD 4HE PEAKS IN VARIANCE FOR THE ORIGINAL

&IGURE  6ARIANCE OF THE CROSS SECTION BEFORE SOLID AND AFTER DASHED TIME ALIGNMENT ECHOES CORRESPOND TO POINTS IN TIME WHERE THE ECHOES DERIVATIVE IS LARGE THUS YIELDING LARGE VARIATIONS IN THE CROSS SECTION 4HE VARIANCE AFTER THE TIME ALIGNMENT HAS A CONSTANT AMPLITUDE AND DOES NOT VARY WITH TIME THE SMALL VARIATIONS THAT ARE STILL PRESENT ARE MAINLY DUE TO ADDITIVE NOISE



$ISCUSSION

7E HAVE PROPOSED A NEW METHOD FOR ESTIMATING SMALL TIME DELAYS IN NARROWBAND SIGNALS #ABOT ;= PROPOSED TO USE THE CORRELATION WITH THE (ILBERT TRANSFORM OF THE REFERENCE ECHO INSTEAD OF THE ORIGINAL ECHO /UR PROPOSED ESTIMATOR USES HOWEVER ONLY THE (ILBERT TRANSFORM CORRELATION FOR TIME LAG ZERO &OR NARROWBAND SIGNALS WE HAVE SHOWN THAT THIS CROSS CORRELATION HAS THE SHAPE OF A SINE WAVE IN THE NEIGHBOURHOOD OF THE TRUE TIME DELAY t &OR THE STANDARD CROSS CORRELATION THIS WILL BE A COSINE BUT KNOWING ONLY ONE POINT ON THAT CURVE RESULTS IN AN AMBUIGITY OF THE SIGN SINCE COSINE IS AN EVEN FUNCTION 4HUS THE (ILBERT TRANSFORMING IS NECESSARY &OR LARGE 2-1 IT CAN BE SHOWN FROM  THAT |r s } 3  B` t $ t z a   3  BNR  t 2-1 WHILE #ABOTS ESTIMATOR HAS THE ASYMPTOTIC SHAPE ;= |r s } 3  B#ABOT ` t $ t z a  3  2-1





0ART  %STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES

(ENCE FOR LARGE 2-1 AND SMALL TIME DELAYS t THESE TWO METHODS HAVE THE SAME ACCURACY &OR SMALL TIME DELAYS t A BIAS IS INTRODUCED IN OUR METHOD .OTICE THE LACK OF THE USUAL BUMPS IN THE ERROR VARIANCE CURVE 4HIS IS DUE TO THE EdECT THAT NO PEAK AMBUIGITY OCCURS SINCE t IS ASSUMED TO BE SMALLER THAN A QUARTER OF A WAVELENGTH IE  i  "Y USING ONLY ONE CROSS CORRELATION VALUE THE ESTIMATOR CONSISTS OF A SCALAR PRODUCT WITH THE NORMALIZED (ILBERT TRANSFORM OF THE REFERENCE ECHO MAKING IT SUITABLE FOR APPLICATIONS WHERE MULTIPLE ECHOES ARE TO BE COMPARED TO A REFERENCE ECHO 3UCH APPLICATIONS INCLUDE EG SENSOR ARRAYS AND THE PROPOSED AVERAGING ALGORITHM 4HE LATTER CAN BE USED WHEN THE INDIVIDUAL SIGNALS HAVE A SMALL RANDOM TIME SHIFT BETWEEN THEMSELVES )N THIS CASE WE CAN HAVE THE ADVANTAGE OF USING THE ABOVE ESTIMATION METHOD TO ALIGN ALL SIGNALS IN TIME BEFORE AVERAGING THEM 4HE MEAN OF THE ESTIMATED TIME SHIFTS CAN ALSO BE USED TO ND A BETTER ABSOLUTE POSITION IN TIME FOR THE AVERAGED SIGNAL ! TYPICAL AREA OF APPLICATION FOR THIS METHOD IS ULTRASONIC MEASUREMENTS

!

4HE EdECT OF BANDWIDTH



!SSUMING THAT THE NARROWBANDED SIGNAL Q S HAS THE &OURIER TRANSFORM Q 6 JE f E J  !  3 1E   OTHERWISE

WHERE $ DENOTES THE ENERGY AND ! THE BANDWIDTH WE HAVE USING 0ARSEVAL THEOREM AND THE MODEL R S  Q S ` t :  :  RSQ v S CS  Ds {it 1 E I RFM E 1 E CE

: i 3(

4HUS WE HAVE 2 `  R S Q v S CS $ RHMB !t RHM 2  $  Q S CS

v wi 3( v wi 3( $ Ds {it $ Ds {it   `I

I ! I {t i 3( ! I {t i3( $ s {it $ s {i t  `I D ! RHMB !t
I D ! RHMB !t ! ! $  I ! RHMB !t I RHM {E t  `$ RHMB !t RHM {E t !


i3(

`IDs {it J1 E J CE

: i 3(
i 3(

IDs {it J1 E J CE



 RHMB !t RHM



"

,INEARIZING THE ESTIMATOR

)N ORDER TO LINEARIZE THE ESTIMATOR  WE USE THE 4AYLOR SERIES OF THE FUNCTION t u  W E W X  @QBRHM X 

  E W  X
W ` W 4HE PARTIAL DERIVATIVES ARE E  W E X  


n n E n E n n n
X ` X
aaa W n&( & X n&( &



`   W   `W R P `  r s  X   X X ` W `

   r s X   PX  ` W





 )NTRODUCING 3  WHICH IS A ZERO MEAN STOCHASTIC VARIABLE WITH VARIANCE }3  5@Q F! G AND x3  $ F! G WE CAN APPROXIMATE THE ESTIMATOR  BY PUTTING !  x3
3 AND CORRESPONDINGLY  x2
2 AND USE THE ABOVE LINEARIZATION AT THE POINT x3  x2 t u  ! B t  @QBRHM  t u x3  x  P3 { @QBRHM
3 P  ` 2    x2 x2 ` x3 x    x2 2 ` x3

B IS NOW FOUND TO BE 4HE MEAN OF t t u N O  x3 B  $ t @QBRHM x2 



4HE ERROR VARIANCE IS    |r t u    s } x3 3 2 x3 B` t P $ t  $ @QBRHM `t

P  `     x2 x2 ` x x   x2 3 2 ` x3 t u t u x3 }  x }  @QBRHM  ` t
  3 
  3 2  x2   x2 ` x3  x2 x2 ` x3 t t u u      x3 x }
x }  @QBRHM ` t
2  3  3 2   x2   x2 x2 ` x3 IF WE ASSUME THAT 2 AND 3 ARE UNCORRELATED 4HE MEAN OF ! IS FOUND TO BE E  8 $ F! G  $ `3 R :J < Q v :J <
R :J < M v  :J <
M :J < Q v :J <
M :J< M v  :J < : { `

 RSQ v S CS { $ RHM 
3


t(

 FOLLOWS FROM

SINCE M :J< AND M v  :J < ARE INDEPENDENT AND HAVE ZERO MEAN 4HE MEAN OF $F G  $ : {
E  8 t(

Q :J<
Q :J< M :J<
M  :J < 

 Q 

S CS
3 -}  $
3 }

 &OR THE VARIANCES OF

0ART  %STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES AND ! WE HAVE h ! ` x3



  5@Q F! G  $ }3

 3 $

   E 8 
t(

R :J < M v  :J <
M :J < Q v :J <
M :J< M v  :J <

  

 3 $


E E  8 8 t( u(

R :J< M v  :J< R :K< M v  :K<

M :J < Q v :J < M :K< Q v :K<
M :J< M v  :J < M :K< M v  :K<

R :J < M v  :J< M :K< Q v :K<

R :J < M v  :J < M :K< M v  :K<
M :J < Q v :J< M :K< M v  :K< t  3 AND
  5@Q F G  $ }2 

$  $  }
}
- }} 3 3

u  $ 3 } 
- 3 }   $ 3 } 
3 3 }     E 8 
t(



` x2

 3 $

Q :J < M :J <
M :J < ` }

  

 3 $

E E  8 8 t( u(

Q :J< M :J< Q :K< M :K<

M :J < M :K<
} 

Q :J< M :J< M :K< ` Q :J< M :J< } ` M :J < } t u a  $  `        3  }

-
- }
- } ` - } } 3  $ 3 }
- 3 }   $ 3 } 
3 3 }




(ERE WE HAVE USED THE FOURTH MOMENT OF A ZERO MEAN 'AUSSIAN STOCHASTIC VARIABLE VARIABLE  $ FW G  }  ;= GIVING US $
E E  8 8 t( u(

M :J< M :K<

 $  `


8
t (u

` a - ` - }
- }  - 
- }
 

M :J < M :K<
a

8
t(u

M :J< 

AND WITH THE THIRD MOMENT $ FW G   ;= $

E E  8 8 t( u(

M :J< M :K <

$

8
t (u

M :J < M :K<

8
t (u

M :J <





"IBLIOGRAPHY
;= 2ADIO BROADCASTING SYSTEMS $IGITAL !UDIO "ROADCASTING $!" TO MOBILE PORTABLE AND XED RECEIVERS %43   %43) %UROPEAN 4ELECOMMUNICATIONS 3TANDARDS )NSTITUTE 6ALBONNE &RANCE &EB  ;= $IGITAL BROADCASTING SYSTEMS FOR TELEVISION SOUND AND DATA SERVICES %UROPEAN 4ELECOM MUNICATIONS 3TANDARD PR%43   $RAFT VERSION   !PR  ;= 4RANSMISSION AND RECEPTION 4ECHNICAL 2EPORT '3- 2ECOMMENDATION  VERSION  %43) 6ALBONNE &RANCE -AR  ;= 7ORKING DOCUMENT TOWARDS %423-'  SELECTION PROCEDURES FOR THE CHOICE OF RADIO TRANSMISSION TECHNOLOGIES OF THE UNIVERSAL MOBILE TELECOMMUNICATIONS SYSTEM 5-43  4ECHNICAL 2EPORT $423-'  %43) 6ALBONNE &RANCE  ;= - !BRAMOVITZ AND ) 3TEGUN (ANDBOOK OF MATHEMATICAL FUNCTIONS WITH FORMULAS GRAPHS AND MATHEMATICAL TABLES .UMBER  IN !PPLIED -ATH 3ERIES .AT "UREAU OF 3TAND 7ASHINGTON $# 53!  ;= - !LARD AND 2 ,ASSALLE 0RINCIPLES OF MODULATION AND CHANNEL CODING FOR DIGITAL BROAD CASTING FOR MOBILE RECEIVERS %"5 2EVIEW 4ECHNICAL   !UG  ;= * " !NDERSEN AND " , !NDERSEN &IRST ORDER FREQUENCY SELECTIVE EdECTS ON PHASE MODULATIONS IN A FADING CHANNEL 4ECHNICAL 2EPORT #/34  4$   %52/#/34 &IRENZE *AN  ;= - !RDITI & &ORSTER AND * (UNT 4RANSIENT ELDS OF CONCAVE ANNULAR ARRAYS 5LTRASON )MAG   ;= * * VAN DE "EEK %STIMATION OF 3YNCHRONIZATION 0ARAMETERS ,IC THESIS ,ULE 5NIVERSITY OF 4ECHNOLOGY ,ULE 3WEDEN  ;= * * VAN DE "EEK / %DFORS 0 / "RJESSON - 7AHLQVIST AND # STBERG ! CONCEPTUAL STUDY OF /&$- BASED MULTIPLE ACCESS SCHEMES 0ART  #HANNEL ESTIMATION IN THE UPLINK 4ECHNICAL 2EPORT 4DOC  %43) 34# 3-' MEETING NO  (ELSINKI &INLAND -AY  ;= * * VAN DE "EEK / %DFORS 0 / "RJESSON - 7AHLQVIST AND # STBERG ! CON CEPTUAL STUDY OF /&$- BASED MULTIPLE ACCESS SCHEMES 0ART  0ERFORMANCE EVALUATION OF A CODED SYSTEM 4ECHNICAL 2EPORT 4DOC  %43) 34# 3-' MEETING NO  $SSELDORF 'ERMANY 3EPT 





0ART  %STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES

;= * * VAN DE "EEK / %DFORS - 3ANDELL 3 + 7ILSON AND 0 / "RJESSON /N CHANNEL ESTIMATION IN /&$- SYSTEMS )N 0ROC )%%% 6EHIC 4ECHNOL #ONF VOLUME  PAGES  #HICAGO ), *ULY  ;= * * VAN DE "EEK - 3ANDELL AND 0 / "RJESSON -, ESTIMATION OF TIMING AND FRE QUENCY OdSET IN MULTICARRIER SYSTEMS 2ESEARCH 2EPORT 45,%!  $IVISION OF 3IGNAL 0ROCESSING ,ULE 5NIVERSITY OF 4ECHNOLOGY  ;= * * VAN DE "EEK - 3ANDELL - )SAKSSON AND 0 / "RJESSON ,OW COMPLEX FRAME SYNCHRONIZATION IN /&$- SYSTEMS )N 0ROC )NT #ONF 5NIVERSAL 0ERSONAL #OMMUN PAGES  4OKYO *APAN .OV  ;= $ "ENGTSSON AND $ ,ANDSTRM #ODING IN A DISCRETE MULTITONE MODULATION SYSTEM -ASTERS THESIS ,ULE 5NIVERSITY OF 4ECHNOLOGY !PR  ;= % "IGLIERI $ $IVSALAR 0 * -C,ANE AND - + 3IMON )NTRODUCTION TO TRELLIS CODED MODULATION WITH APPLICATIONS -ACMILLAN .EW 9ORK  ;= * ! # "INGHAM -ULTICARRIER MODULATION FOR DATA TRANSMISSION !N IDEA WHOSE TIME HAS COME )%%% #OMMUN -AG    -AY  ;= 2 "OUCHER AND * (ASSAB !NALYSIS OF DISCRETE IMPLEMENTATION OF A GENERALIZED CROSS CORRELATOR )%%% 4RANS !COUST 3PEECH 3IGNAL 0ROCESS     ;= 2 "RACEWELL 4HE &OURIER TRANSFORM AND ITS APPLICATIONS -C'RAW (ILL  ED EDITION  ;= * "RADLEY AND 2 +IRLIN $ELAY ESTIMATION BY EXPECTED VALUE )%%% 4RANS !COUST 3PEECH 3IGNAL 0ROCESS     ;= 0 "RJESSON . ' (OLMER + ,INDSTRM AND ' 3ALOMONSSON .OTES ON ULTRASOUND ECHOES FROM ANGULAR SURFACES 4ECHNICAL REPORT 42  %LECTRICAL %NGINEERING $EPART MENT 5NIVERSITY OF ,UND 3EPT  ;= 2 #ABOT ! NOTE ON THE APPLICATION OF THE (ILBERT TRANSFORM TO TIME DELAY ESTIMATION )%%% 4RANS !COUST 3PEECH 3IGNAL 0ROCESS     ;= ' #ARTER EDITOR )%%% 4RANS !COUST 3PEECH 3IGNAL 0ROCESSING 3PECIAL ISSUE ON TIME DELAY ESTIMATION VOLUME   ;= ' #ARTER #OHERENCE AND TIME DELAY ESTIMATION 0ROC )%%%    &EB  ;= ' #ARTER EDITOR #OHERENCE AND TIME DELAY ESTIMATION )%%% 0RESS .EW 9ORK  ;= % & #ASAS AND # ,EUNG /&$- FOR DATA COMMUNICATION OVER MOBILE RADIO &- CHANNELS 0ART ) !NALYSIS AND EXPERIMANTAL RESULTS )%%% 4RANS #OMMUN    -AY  ;= % & #ASAS AND # ,EUNG /&$- FOR DATA COMMUNICATION OVER MOBILE RADIO &- CHANNELS 0ART )) 0ERFORMANCE IMPROVEMENT )%%% 4RANS #OMMUN    !PR  ;= * #AVERS AND 0 (O !NALYSIS OF THE ERROR PERFORMANCE OF TRELLIS CODED MODULATION IN 2AYLEIGH FADING CHANNELS )%%% 4RANS #OMMUN    *AN 


;= * + #AVERS !N ANALYSIS OF PILOT SYMBOL ASSISTED MODULATION FOR 2AYLEIGH FADING CHAN NELS )%%% 4RANS 6EHIC 4ECHNOL    .OV  ;= 2 7 #HANG 3YNTHESIS OF BAND LIMITED ORTHOGONAL SIGNALS FOR MULTICHANNEL DATA TRANS MISSION "ELL 3YSTEM 4ECH *  $EC  ;= 2 7 #HANG AND 2 ! 'IBBY 4HEORETICAL STUDY OF PERFORMANCE OF AN ORTHOGONAL MULTIPLEXING DATA TRANSMISSION SCHEME )%%% 4RANS #OMMUN #/-    !UG  ;= & #HAPEAU "LONDEAU AND * 'REENLEAF ! THEORETICAL MODEL OF ACOUSTOELECTRIC TRANSDUCER WITH A NONUNIFORM DISTRIBUTION OF PIEZOELECTRIC COEbCIENT !PPLICATION TO TRANSDUCER OPTIMIZATION * !COUST 3OC !M     ;= ! #HINI -ULTICARRIER MODULATION IN FREQUENCY SELECTIVE FADING CHANNELS 0H$ THESIS #ARLETON 5NIVERSITY /TTAWA #ANADA  ;= ! #HINI - 3 %L 4ANANY AND 3 ! -AHMOUD 4RANSMISSION OF HIGH RATE !4- PACKETS OVER INDOOR RADIO CHANNELS )%%% * 3ELECT !REAS #OMMUN    !PR  ;= 0 3 #HOW "ANDWIDTH OPTIMIZED DIGITAL TRANSMISSION TECHNIQUES FOR SPECTRALLY SHAPED CHANNELS WITH IMPULSE NOISE 0H$ THESIS 3TANFORD 5NIVERSITY #! -AY  ;= 0 3 #HOW . !L $HAHIR AND * - #IOb ! MULTICARRIER % ($3, TRANSCEIVER SYSTEM WITH CODED MODULATION %UR 4RANS 4ELECOMMUN 2EL 4ECHNOL    -AY  ;= 0 3 #HOW * - #IOb AND * ! # "INGHAM ! PRACTICAL DISCRETE MULTITONE TRANSCEIVER LOADING ALGORITHM FOR DATA TRANSMISSION OVER SPECTRALLY SHAPED CHANNELS )%%% 4RANS #OMMUN    &EB  ;= 3 + #HOW AND 0 3CHULTHEISS $ELAY ESTIMATION USING NARROWBAND PROCESSES )%%% 4RANS !COUST 3PEECH 3IGNAL 0ROCESSING     ;= , * #IMINI !NALYSIS AND SIMULATION OF A DIGITAL MOBILE CHANNEL USING ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING )%%% 4RANS #OMMUN #/-    *ULY  ;= * - #IOb 0ERSONAL COMMUNICATION  ;= & #LASSEN AND ( -EYR &REQUENCY SYNCHRONIZATION ALGORITHMS FOR /&$- SYSTEMS SUIT ABLE FOR COMMUNICATION OVER FREQUENCY SELECTIVE FADING CHANNELS )N 0ROC )%%% 6EHIC 4ECHNOL #ONF VOLUME  PAGES  3TOCKHOLM 3WEDEN *UNE  ;= 7 #OBB &REQUENCY DOMAIN METHOD FOR THE PREDICTION OF THE ULTRASONIC ELD PATTERNS OF PULSED FOCUSED RADIATORS * !COUST 3OC !M     ;= , #OHEN 4IME FREQUENCY ANALYSIS 0RENTICE (ALL .EW *ERSEY  ;= $ #OLTON AND 2 +RESS )NVERSE ACOUSTIC AND ELECTROMAGNETIC SCATTERING THEORY 3PRINGER 6ERLAG  ;= * 7 #OOK 7IDEBAND IMPULSIVE NOISE SURVEY OF THE ACCESS NETWORK "4 4ECHNOL *OURN    *ULY 



0ART  %STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES

;= 4 DE #OUASNON 2 -ONNIER AND * " 2AULT /&$- FOR DIGITAL 46 BROADCASTING 3IGNAL 0ROC    3EPT  ;= & $AdARA AND / !DAMI ! NEW FREQUENCY DETECTOR FOR ORTHOGONAL MULTICARRIER TRANSMIS SION TECHNIQUES )N 0ROC )%%% 6EHIC 4ECHNOL #ONF VOLUME  PAGES  #HICAGO ), *ULY  ;= & $AdARA AND ! #HOULY -AXIMUM LIKELIHOOD FREQUENCY DETECTORS FOR ORTHOGONAL MUL TICARRIER SYSTEMS )N 0ROC )NTERN #ONF #OMMUN PAGES  'ENEVA 3WITZERLAND -AY  ;= 3 $EANS 4HE 2ADON TRANSFORM AND SOME OF ITS APPLICATIONS *OHN 7ILEY 3ONS 

;= 2 $INIS 0 -ONTEZUMA AND ! 'USMO 0ERFORMANCE TRADE OdS WITH QUASI LINEARLY AM PLIED /&$- THROUGH A TWO BRANCH COMBINING TECHNIQUE )N 0ROC )%%% 6EHIC 4ECHNOL #ONF VOLUME  PAGES  !TLANTA '! !PR  ;= $ $IVSALAR AND - + 3IMON 4RELLIS CODED MODULATION FOR   BITSS TRANSMISSION OVER A FADING MOBILE SATELLITE CHANNEL )%%% * 3ELECT !REAS #OMMUN 3!#     &EB  ;= $ $IVSALAR AND - + 3IMON 4HE DESIGN OF TRELLIS CODED -03+ FOR FADING CHANNELS 0ERFORMANCE CRITERIA )%%% 4RANS #OMMUN    3EPT  ;= $ $IVSALAR AND - + 3IMON 4HE DESIGN OF TRELLIS CODED -03+ FOR FADING CHANNELS 3ET PARTITIONING FOR OPTIMUM CODE DESIGN )%%% 4RANS #OMMUN    3EPT  ;= $ $IVSALAR AND - + 3IMON -ULTIPLE SYMBOL DIdERENTIAL DETECTION OF -03+ )%%% 4RANS #OMMUN    -AR  ;= $ $IVSALAR AND - + 3IMON -AXIMUM LIKELIHOOD DIdERENTIAL DETECTION OF UNCODED AND TRELLIS CODED AMPLITUDE PHASE MODULATION OVER !7'. AND FADING CHANNELS -ETRICS AND PERFORMANCE )%%% 4RANS #OMMUN    *AN  ;= ( $JELOUAH * "ABOUX AND - 0ERDRIX %XPERIMENTAL DETERMINATION OF THE IMPULSE RESPONSE OF AXISYMMETRIC FOCUSED ULTRASONIC TRANSDUCERS * 0HYS $ !PPL 0HYS    ;= * $U AND " 6UCETIC 4RELLIS CODED  1!- FOR FADING CHANNELS %UR 4RANS 4ELECOM MUN 2EL 4ECHNOL    -AY  ;= $ % $UDGEON AND 2 - -ERSEREAU -ULTIDIMENSIONAL DIGITAL SIGNAL PROCESSING 0RENTICE (ALL %NGLEWOOD #LIdS .*  ;= ( $YM AND ( -C+EAN &OURIER SERIES AND INTEGRALS !CADEMIC 0RESS .EW 9ORK  ;= / %DFORS - 3ANDELL * * VAN DE "EEK $ ,ANDSTRM AND & 3JBERG !N INTRODUCTION TO ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING 2ESEARCH 2EPORT 45,%!  $IV OF 3IGNAL 0ROCESSING ,ULE 5NIVERSITY OF 4ECHNOLOGY 3EPT  ;= / %DFORS - 3ANDELL * * VAN DE "EEK 3 + 7ILSON AND 0 / "RJESSON !NALYSIS OF $&4 BASED CHANNEL ESTIMATORS FOR /&$- 2ESEARCH 2EPORT 45,%!  $IV OF 3IGNAL 0ROCESSING ,ULE 5NIVERSITY OF 4ECHNOLOGY 3EPT 


;= / %DFORS - 3ANDELL * * VAN DE "EEK 3 + 7ILSON AND 0 / "RJESSON /&$CHANNEL ESTIMATION BY SINGULAR VALUE DECOMPOSITION )N 0ROC )%%% 6EHIC 4ECHNOL #ONF PAGES  !TLANTA '! !PR  ;= / %DFORS - 3ANDELL * * VAN DE "EEK 3 + 7ILSON AND 0 / "RJESSON /&$CHANNEL ESTIMATION BY SINGULAR VALUE DECOMPOSITION 2ESEARCH 2EPORT 45,%!  $IV OF 3IGNAL 0ROCESSING ,ULE 5NIVERSITY OF 4ECHNOLOGY 3EPT  ;= 6 %NGELS AND ( 2OHLING $IdERENTIAL MODULATION TECHNIQUES FOR A  -"ITS RADIO CHANNEL USING ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING 7IRELESS 0ERS #OMMUN      ;= 6 %NGELS AND ( 2OHLING -ULTILEVEL DIdERENTIAL MODULATION TECHNIQUES  $!03+ FOR MULTICARRIER TRANSMISSION SYSTEMS %UR 4RANS 4ELECOMMUN 2EL 4ECHNOL    .OV  ;= ( %RIKSSON $IGITAL "LOCK 4RANSMISSION AND 4IME OF &LIGHT %STIMATION 2ECEIVER $ESIGN AND 0ERFORMANCE !NALYSIS 0H$ THESIS ,ULE 5NIVERSITY OF 4ECHNOLOGY ,ULE 3WEDEN  ;= + &AZEL 0ERFORMANCE OF CONVOLUTIONALLY CODED #$-!/&$- IN A FREQUENCY TIME SELEC TIVE FADING CHANNEL AND ITS NEAR FAR RESISTANCE )N 0ROC )NTERN #ONF #OMMUN VOLUME  PAGES  .EW /RLEANS ,! .OV  ;= - &INK AND * & #ARDOSO $IdRACTION EdECTS IN PULSE ECHO MEASUREMENT )%%% 4RANS 3ONIC 5LTRASON 35     ;= 2 & &ISCHER AND * " (UBER ! NEW LOADING ALGORITHM FOR DISCRETE MULTITONE TRANSMIS SION )N 0ROC 'LOBECOM ,ONDON 5+  4O APPEAR ;= * &OX " +HURI 9AKUB AND ' +INO (IGH FREQUENCY ACOUSTIC WAVE MEASUREMENTS IN AIR )N 0ROC )%%% 5LTRASONIC 3YMPOSIUM PAGES  !TLANTA '!  ;= 0 &RENGER AND ! 3VENSSON ! DECISION DIRECTED COHERENT DETECTOR FOR /&$- )N 0ROC )%%% 6EHIC 4ECHNOL #ONF VOLUME  PAGES  !TLANTA '! !PR  ;= & 'AGNON AND $ (ACCOUN "OUNDS ON THE ERROR PERFORMANCE OF CODING FOR NONINDEPEN DENT 2ICIAN FADING CHANNELS )%%% 4RANS #OMMUN    &EB  ;= - 'HOSH !NALYSIS OF THE EdECT OF IMPULSE NOISE ON MULTICARRIER AND SINGLE CARRIER 1!SYSTEMS )%%% 4RANS #OMMUN    &EB  ;= ' 'OLUB AND # VAN ,OAN -ATRIX #OMPUTATIONS .ORTH /XFORD !CADEMIC *OHNS (OPKINS !CADEMIC 0RESS ND EDITION  ;= ! 'RENNBERG AND - 3ANDELL %XPERIMENTAL DETERMINATION OF THE ULTRASONIC ECHO FROM A POINTLIKE REECTOR USING A TOMOGRAPHIC APPROACH )N 0ROC OF THE )%%% 5LTRASONIC 3YMPOSIUM PAGES  4UCSON !: /CT  ;= ! 'RENNBERG AND - 3ANDELL %STIMATION OF SUBSAMPLE TIME DELAY DIdERENCES IN NARROW BAND ULTRASONIC ECHOES USING THE (ILBERT TRANSFORM CORRELATION )%%% 4RANS 5LTRASON &ERROELEC &REQ #ONTR    



0ART  %STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES

;= ' 'RIMMETT AND $ 3TIRZAKER 0ROBABILITY AND RANDOM PROCESSES /XFORD 5NIVERSITY 0RESS /XFORD %NGLAND ND EDITION  ;= 2 'ROSS AND $ 6EENEMAN #LIPPING DISTORTION IN $-4 !$3, SYSTEMS %LECTRON ,ETT    .OV  ;= - 'UDMUNDSON AND 0 / !NDERSSON !DJACENT CHANNEL INTERFERENCE IN AN /&$- SYS TEM )N 0ROC )%%% 6EHIC 4ECHNOL #ONF VOLUME  PAGES  !TLANTA '! !PR  ;= 2 (AAS !PPLICATION DES TRANSMISSIONS PORTEUSES MULTIPLES AUX COMMUNICATIONS RADIO MOBILES 0HD THESIS %COLE .ATIONAL 3UPRIEURE DES 4LCOMMUNICATIONS 0ARIS &RANCE *AN  )N %NGLISH ;= 0 ( (ALPERN /PTIMUM NITE DURATION .YQUIST SIGNALS )%%% 4RANS #OMMUN #/-    *UNE  ;= * (AVLICE AND * 4AENZER -EDICAL ULTRASONIC IMAGING !N OVERVIEW OF PRINCIPLES AND INSTRUMENTATION 0ROC )%%%  !PR  ;= ) (EIN AND 7 /"RIEN #URRENT TIME DOMAIN METHODS FOR ASSESSING TISSUE MOTION BY ANALYSIS FROM REECTED ULTRASOUND ECHOES ! REVIEW )%%% 4RANS 5LTRASON &ERROELEC &REQ #ONTR     ;= $ (ERTZ 4IME DELAY ESTIMATION BY COMBINING EbCIENT ALGORITHMS AND GENERALIZED CROSS CORRELATION METHODS )%%% 4RANS !COUST 3PEECH 3IGNAL 0ROCESSING     ;= # (ILL 4HE GENERATION AND STRUCTURE OF ACOUSTIC ELDS )N # (ILL EDITOR 0HYSICAL PRINCIPLES IN MEDICAL ULTRASONIC PAGES  %LLIS (ORWOOD  ;= 0 (O * #AVERS AND * 6ARALDI 4HE EdECTS OF CONSTELLATION DENSITY ON TRELLIS CODED MODULATION IN FADING CHANNELS )%%% 4RANS 6EHIC 4ECHNOL    !UG  ;= 0 (O AND $ &UNG %RROR PERFORMANCE OF INTERLEAVED TRELLIS CODED 03+ MODULATIONS IN CORRELATED 2AYLEIGH FADING CHANNELS )%%% 4RANS #OMMUN    $EC  ;= 0 (O AND $ &UNG %RROR PERFORMANCE OF MULTIPLE SYMBOL DIdERENTIAL DETECTION OF 03+ SIGNALS TRANSMITTED OVER CORRELATED 2AYLEIGH FADING CHANNELS )%%% 4RANS #OMMUN    /CT  ;= * (UANG AND , , #AMPBELL 4RELLIS CODED -$03+ IN CORRELATED AND SHADOWED 2ICIAN FADING CHANNELS )%%% 4RANS #OMMUN    .OV  ;= 0 (HER 4#- ON FREQUENCY SELECTIVE LAND MOBILE FADING CHANNELS )N 0ROC 4IRRENIA )NT 7ORKSHOP $IGITAL #OMMUN 4IRRENIA )TALY 3EPT  ;= 0 (HER ! STATISTICAL DISCRETE TIME MODEL FOR THE 73353 MULTIPATH CHANNEL )%%% 4RANS #OMMUN    .OV  ;= 0 (HER * (AGENAUER % /dER # 2APP AND ( 3CHULZE 0ERFORMANCE OF AN 2#0# CODED /&$- BASED DIGITAL AUDIO BRAODCASTING $!" SYSTEM )N 0ROC 'LOBECOM VOL UME  PAGES  


;= ' *ACOVITTI AND ' 3CARANO $ISCRETE TIME TECHNIQUES FOR TIME DELAY ESTIMATION )%%% 4RANS 3IGNAL 0ROCESS     ;= 7 # *AKES -ICROWAVE MOBILE COMMUNICATIONS #LASSIC 2EISSUE )%%% 0RESS 0ISCATAWAY .EW *ERSEY  ;= 3 ( *AMALI AND 4 ,E .GOC ! NEW  STATE  03+ 4#- SCHEME FOR FAST FADING SHADOWED MOBILE RADIO CHANNELS )%%% 4RANS 6EHIC 4ECHNOL    &EB  ;= 3 ( *AMALI AND 4 ,E .GOC #ODED MODULATION TECHNIQUES FOR FADING CHANNELS +LUWER !CADEMIC 0UBLISHERS  ;= * *ENSEN ! MODEL FOR THE PROPAGATION AND SCATTERING OF ULTRASOUND IN TISSUE * !COUST 3OC !M     ;= ! *ONES AND 4 7ILKINSON #OMBINED CODING FOR ERROR CONTROL AND INCREASED ROBUSTNESS TO SYSTEM NONLINEARITIES IN /&$- )N 0ROC )%%% 6EHIC 4ECHNOL #ONF VOLUME  PAGES  !TLANTA '! !PR  ;= ! *ONES 4 7ILKINSON AND 3 "ARTON "LOCK CODING SCHEME FOR REDUCTION OF PEAK TO MEAN ENVELOPE POWER RATIO OF MULTICARRIER TRANSMISSION SCHEMES %LECTRON ,ETT     $EC  ;= ) +ALET 4HE MULTITONE CHANNEL )%%% 4RANS #OMMUN    &EB  ;= ) +ALET AND 3 3HAMAI /N THE CAPACITY OF A TWISTED WIRE PAIR 'AUSSIAN MODEL )%%% 4RANS #OMMUN    -AR  ;= + 7 +ANG * !NN AND ( 3 ,EE $ECISION DIRECTED MAXIMUM LIKELIHOOD ESTIMATION OF /&$- FRAME SYNCHRONIZATION OdSET %LECTRON ,ETT    $EC  ;= 4 +ELLER ET AL 2EPORT ON DIGITAL AUDIO RADIO LABORATORY TESTS 4ECHNICAL REPORT %LECTRONIC )NDUSTRIES !SSOCIATION -AY  ;= ' +INO !COUSTIC WAVES 0RENTICE (ALL  ;= , +INSLER ! &REY ! #OPPENS AND * 3ANDERS &UNDAMENTALS OF ACOUSTICS *OHN 7ILEY 3ONS .EW 9ORK 53!  ;= " +HLER )MPULSE RESPONSE OF A PIEZOELECTRIC LAYER !COUSTICA   ;= ' +OSSOd !NALYSIS OF FOCUSING ACTION OF SPHERICALLY CURVED TRANSDUCERS 5LTRASOUND IN -ED AND "IOL   ;= ( * ,ANDAU AND ( / 0OLLAK 0ROLATE SPHERIODAL WAVE FUNCTIONS &OURIER ANALYSIS AND UNCERTAINTY ))) 4HE DIMENSION OF THE SPACE OF ESSENTIALLY TIME AND BAND LIMITED SIGNALS "ELL 3YSTEM 4ECH *   ;= " ,E &LOCH - !LARD AND # "ERROU #ODED ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING 0ROC )%%%    *UNE  ;= " ,E &LOCH 2 (ALBERT ,ASSALLE AND $ #ASTELAIN $IGITAL SOUND BROADCASTING TO MOBILE RECEIVERS )%%% 4RANS #ONSUMER %LECTRONICS    !UG 



0ART  %STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES

;= ! ,HMERY )MPULSE RESPONSE METHOD TO PREDICT ECHO RESPONSES FROM TARGETS OF COMPLEX GEOMETRY 0ART  * !COST 3OC !M     ;= ( ,I AND * + #AVERS !N ADAPTIVE LTERING TECHNIQUE FOR PILOT AIDED TRANSMISSION SYSTEMS )%%% 4RANS 6EHIC 4ECHNOL    !UG  ;= ! ,OUIS /RTHOGONAL FUNCTION SERIES EXPANSIONS AND THE NULL SPACE OF THE 2ADON TRANS FORM 3)!- * -ATH !NAL   ;= " -ARTI %UROPEAN ACTIVITIES ON DIGITAL TELEVISION BROADCASTING FROM COMPANY TO CO OPERATIVE PROJECTS %"5 2EVIEW 4ECHNICAL    ;= * -AZO AND * 3ALZ 0ROBABILITY OF ERROR FOR QUADRATIC DETECTORS "ELL 3YSTEM 4ECH *    .OV  ;= 2 ' -C+AY 0 * -C,ANE AND % "IGLIERI %RROR BOUNDS FOR TRELLIS CODED -03+ ON A FADING MOBILE SATELLITE CHANNEL )%%% 4RANS #OMMUN    $EC  ;= 3 -C,AREN AND * 7EIGHT 4RANSMIT RECEIVE MODE RESPONSES FROM NITE SIZED TARGETS IN UID MEDIA * !COUST 3OC !M     ;= 2 -ODDEMEIJER /N THE DETERMINATION OF THE POSITION OF EXTREMA OF SAMPLED CORRELATORS )%%% 4RANS 3IGNAL 0ROCESS     ;= - -OENECLAEY AND - VAN "LADEL $IGITAL ($46 BROADCASTING OVER THE #!46 DISTRIB UTION SYSTEM 3IGNAL 0ROC )MAGE #OMMUN    $EC  ;= - , -OHER AND * ( ,ODGE 4#-0 ! MODULATION AND CODING STRATEGY FOR 2ICIAN FADING CHANNELS )%%% * 3ELECT !REAS #OMMUN    $EC  ;= 0 -OOSE ! TECHNIQUE FOR ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING FREQUENCY OdSET CORRECTION )%%% 4RANS #OMMUN    /CT  ;= ! -LLER /&$- TRANSMISSION OVER TIME VARIANT CHANNELS )N 0ROC )NT "ROADC #ONV NUMBER  PAGES  !MSTERDAM .ETHERLANDS 3EPT  ;= & .ATTERER 4HE MATHEMATICS OF COMPUTERIZED TOMOGRAPHY *OHN 7ILEY 3ONS 

;= * .ILSSON #ODING TO CONTROL THE SIGNAL WAVEFORM IN - ARY 03+ MULTICARRIER COMMUNI CATIONS )N 0ROC 2ADIOVETEN +ONF PAGES  ,ULE 3WEDEN *UNE  ;= 2 VAN .OBELEN AND $ 0 4AYLOR !NALYSIS OF THE PAIRWISE ERROR PROBABILITY OF NONIN TERLEAVED CODES ON THE 2AYLEIGH FADING CHANNEL )%%% 4RANS #OMMUN    !PR  ;= ! /PPENHEIM AND 2 3CHAFER $ISCRETE TIME SIGNAL PROCESSING 0RENTICE (ALL  ;= ! 0APOULIS 0ROBABILITY RANDOM VARIABLES AND STOCHASTIC PROCESSES -C'RAW (ILL 3INGA PORE ND EDITION  ;= ! 0ELED AND ! 2UIZ &REQUENCY DOMAIN DATA TRANSMISSION USING REDUCED COMPUTATIONAL COMPLEXITY ALGORITHMS )N 0ROC )%%% )NT #ONF !COUST 3PEECH 3IGNAL 0ROCESSING PAGES  $ENVER #/ 


;= ! 0ENTTINEN AND - ,UUKKALA 4HE IMPULSE RESPONSE AND PRESSURE NEARELD OF A CURVED ULTRASONIC RADIATOR * 0HYS $ !PPL 0HYS   ;= " 0ICINBONO 2ANDOM SIGNALS AND SYSTEMS 0RENTICE (ALL %NGLEWOOD #LIdS .*  ;= 4 0OLLET AND - -OENECLAEY 3YNCHRONIZABILITY OF /&$- SIGNALS )N 0ROC 'LOBECOM VOLUME  PAGES  3INGAPORE .OV  ;= 4 0OLLET 0 3PRUYT AND - -OENECLAEY 4HE "%2 PERFORMANCE OF /&$- SYSTEMS USING NON SYNCHRONIZED SAMPLING )N 0ROC 'LOBECOM VOLUME  PAGES  3AN &RANCISCO #! .OV  ;= 4 0OLLET - VAN "LADEL AND - -OENECLAEY "%2 SENSITIVITY OF /&$- SYSTEMS TO CARRIER FREQUENCY OdSET AND 7IENER PHASE NOISE )%%% 4RANS #OMMUN     &EB-AR!PR  ;= * 0ROAKIS $IGITAL COMMUNICATIONS 0RENTICE (ALL RD EDITION  ;= + 2AMCHANDRAN ! /RTEGA + - 5Z AND - 6ETTERLI -ULTIRESOLUTION BROADCAST FOR DIGITAL ($46 USING JOINT SOURCECHANNEL CODING )%%% * 3ELECT !REAS #OMMUN    *AN  ;= - 2EDWOOD 4RANSIENT PERFORMANCE OF A PIEZOELECTRIC TRANSDUCER * !COUST 3OC !M     ;= # 2EINERS AND ( 2OHLING -ULTICARRIER TRANSMISSION TECHNIQUE IN CELLULAR MOBILE COM MUNICATION SYSTEMS )N 0ROC )%%% 6EHIC 4ECHNOL #ONF PAGES  3TOCKHOLM 3WEDEN *UNE  ;= * 2INNE AND - 2ENFORS 4HE BEHAVIOUR OF ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING IN AN AMPLITUDE LIMITING CHANNEL )N 0ROC )%%% 6EHIC 4ECHNOL #ONF VOLUME  PAGES  .EW /RLEANS ,! -AY  ;= ( 2OHLING + "RNINGHAUS AND 4 -LLER 0ERFORMANCE OF COHERENT /&$- #$-! FOR BROADBAND MOBILE COMMUNICATION )N 0ROC 2!#% -OBILE #OMMUN 3UMMIT PAGES  #ASCAIS .OV  ;= ( 2OHLING AND 2 'RNHEID -ULTICARRIER TRANSMISSION TECHNIQUE IN MOBILE COMMUNICA TION SYSTEMS )N 0ROC 2!#% -OBILE #OMMUN 3UMMIT PAGES  #ASCAIS .OV  ;= ! 2UIZ * - #IOb AND 3 +ASTURIA $ISCRETE MULTIPLE TONE MODULATION WITH COSET CODING FOR THE SPECTRALLY SHAPED CHANNEL )%%% 4RANS #OMMUN    *UNE  ;= - 2USSELL AND ' 3TBER )NTERCHANNEL INTERFERENCE ANALYSIS OF /&$- IN A MOBILE EN VIRONMENT )N 0ROC )%%% 6EHIC 4ECHNOL #ONF VOLUME  PAGES  #HICAGO ), *ULY  ;= - 3ABLATASH 4RANSMISSION OF ALL DIGITAL ADVANCED TELEVISION 3TATE OF THE ART AND FUTURE DIRECTIONS )%%% 4RANS "ROADC    *UNE 



0ART  %STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES

;= " 2 3ALTZBERG 0ERFORMANCE OF AN EbCIENT PARALLEL DATA TRANSMISSION SYSTEM )%%% 4RANS #OMMUN #/-    $EC  ;= - 3ANDELL 3PATIAL DECOMPOSITION OF ULTRASONIC ECHOES ,IC THESIS ,ULE 5NIVERSITY OF 4ECHNOLOGY ,ULE 3WEDEN  ;= - 3ANDELL AND / %DFORS ! COMPARATIVE STUDY OF PILOT BASED CHANNEL ESTIMATORS FOR WIRELESS /&$- 2ESEARCH 2EPORT 45,%!  $IV OF 3IGNAL 0ROCESSING ,ULE 5NIVERSITY OF 4ECHNOLOGY 3EPT  ;= - 3ANDELL AND ! 'RENNBERG %STIMATION OF THE SPATIAL IMPULSE RESPONSE OF AN ULTRASONIC TRANSDUCER USING A TOMOGRAPHIC APPROACH * !COUST 3OC !M     ;= - 3ANDELL * * VAN DE "EEK AND 0 / "RJESSON 4IMING AND FREQUENCY SYNCHRONIZATION IN /&$- SYSTEMS USING THE CYCLIC PREX )N )NTERN 3YMP 3YNCH PAGES  %SSEN 'ERMANY $EC  ;= - 3ANDELL 3 + 7ILSON AND 0 / "RJESSON 0ERFORMANCE ANALYSIS OF CODED /&$ON FADING CHANNELS WITH NON IDEAL INTERLEAVING AND CHANNEL KNOWLEDGE 2ESEARCH 2EPORT 45,%!  $IV OF 3IGNAL 0ROCESSING ,ULE 5NIVERSITY OF 4ECHNOLOGY 3EPT  ;= ' 3ANTELLA /&$- WITH GUARD INTERVAL AND SUBCHANNEL EQUALIZATION IN A  RESOLUTION TRANSMISSION SCHEME FOR DIGITAL TELEVISION BROADCASTING )N 0ROC )NTERN #ONF #OMMUN VOLUME  PAGES  .EW /RLEANS ,! -AY  ;= ( 3ARI ' +ARAM AND ) *EANCLAUDE 4RANSMISSION TECHNIQUES FOR DIGITAL TERRISTRIAL 46 BROADCASTING )%%% #OMMUN -AG    &EB  ;= , , 3CHARF 3TATISTICAL SIGNAL PROCESSING $ETECTION ESTIMATION AND TIME SERIES ANALYSIS !DDISON 7ESLEY  ;= # 3CHLEGEL 4RELLIS CODED MODULATION ON TIME SELECTIVE FADING CHANNELS )%%% 4RANS #OMMUN     ;= # 3CHLEGEL AND $ * #OSTELLO "ANDWIDTH EbCIENT CODING FOR FADING CHANNELS #ODE CONSTRUCTION AND PERFORMANCE ANALYSIS )%%% * 3ELECT !REAS #OMMUN    $EC  ;= 7 & 3CHREIBER !DVANCED TELEVISION SYSTEMS FOR TERRISTRIAL BROADCASTING 3OME PROBLEMS AND SOME PROPOSED SOLUTIONS 0ROC )%%%    *UNE  ;= # 3CHUELER ( ,EE AND ' 7ADE &UNDAMENTALS OF DIGITAL ULTRASONIC IMAGING )%%% 4RANS 3ONICS 5LTRASON 35     ;= - + 3IMON AND $ $IVSALAR 4HE PERFORMANCE OF TRELLIS CODED MULTILEVEL $03+ ON A FADING MOBILE SATELLITE CHANNEL )%%% 4RANS 6EHIC 4ECHNOL    -AY  ;= + 3ISTANIZADEH 0 #HOW AND * #IOb -ULTITONE TRANSMISSION FOR ASYMMETRIC DIGITAL SUBSCRIBER LINES !$3,  )N 0ROC )NTERN #ONF #OMMUN PAGES  -AY  ;= 3 " 3LIMANE AND 4 ,E .GOC 4IGHT BOUNDS ON THE ERROR PROBABILITY OF CODED MODULATION SCHEMES IN 2AYLEIGH FADING CHANNELS )%%% 4RANS 6EHIC 4ECHNOL    &EB 


;= ' 3TRANG ,INEAR !LGEBRA AND )TS !PPLICATIONS !CADEMIC 0RESS ND EDITION  ;= ' , 3TBER AND - 2USSELL 4ERRESTRIAL DIGITAL VIDEO BROADCASTING FOR MOBILE RECEPTION USING /&$- )N 0ROC 'LOBECOM VOLUME  PAGES  3INGAPORE .OV  ;= # % 7 3UNDBERG AND . 3ESHADRI #ODED MODULATION FOR FADING CHANNELS !N OVERVIEW %UR 4RANS 4ELECOMMUN 2EL 4ECHNOL    -AY  ;= . 3UNDSTRM 0 / "RJESSON . ' (OLMER , /LSSON AND ( 7 0ERSSON 2EGISTRATION OF SURFACE STRUCTURES USING AIRBORNE FOCUSED ULTRASOUND 5LTRASOUND -ED "IOL     ;= # 4ELLAMBURA 1 7ANG AND 6 + "HARGAVA ! PERFORMANCE ANALYSIS OF TRELLIS CODED MODULATION SCHEMES OVER 2ICIAN FADING CHANNELS )%%% 4RANS 6EHIC 4ECHNOL     .OV  ;= # 4ELLAMBURA 1 7ANG AND 6 + "HARGAVA 0ERFORMANCE OF TRELLIS CODED MODULA TION SCHEMES ON SHADOWED MOBILE SATELLITE COMMUNICATION CHANNELS )%%% 4RANS 6EHIC 4ECHNOL    &EB  ;= 0 * 4OURTIER 2 -ONNIER AND 0 ,OPEZ -ULTICARRIER MODEM FOR DIGITAL ($46 TERRESTRIAL BROADCASTING 3IGNAL 0ROC )MAGE #OMMUN    $EC  ;= ( VAN 4REES $ETECTION ESTIMATION AND MODULATION THEORY VOLUME  *OHN 7ILEY 3ONS .EW 9ORK 53!  ;= * # 4U 4HEORY DESIGN AND APPLICATION OF MULTI CHANNEL MODULATION FOR DIGITAL COMMU NICATIONS 0H$ THESIS 3TANFORD 5NIVERSITY #! *UNE  ;= ' , 4URIN 4HE CHARACTERISTIC FUNCTION OF (ERMITIAN QUADRATIC FORMS IN COMPLEX NORMAL VARIABLES "IOMETRIKA   ;= ! 6AHLIN AND . (OLTE /PTIMAL NITE DURATION PULSES FOR /&$- )%%% 4RANS #OM MUN    *AN  ;= - 6ETTERLI AND * +OVACEVIC 7AVELETS AND SUBBAND CODING 0RENTICE (ALL %NGLEWOOD #LIdS .*  ;= % 6ITERBO AND + &AZEL (OW TO COMBAT LONG ECHOES IN /&$- TRANSMISSION SCHEMES 3UB CHANNEL EQUALIZATION OR MORE POWERFUL CHANNEL CODING )N 0ROC 'LOBECOM VOLUME  PAGES  3INGAPORE .OV  ;= " 6UCETIC AND * $U 4HE EdECTS OF PHASE NOISE ON TRELLIS CODED MODULATION OVER 'AUSSIAN AND FADING CHANNELS )%%% 4RANS #OMMUN     ;= - 7AHLQVIST 2 ,ARSSON AND # STBERG 4IME SYNCHRONIZATION IN THE UPLINK OF AN /&$- SYSTEM )N 0ROC )%%% 6EHIC 4ECHNOL #ONF VOLUME  PAGES  !TLANTA '! !PR  ;= - 7AHLQVIST # STBERG * * VAN DE "EEK / %DFORS AND 0 / "RJESSON ! CONCEPTUAL STUDY OF /&$- BASED MULTIPLE ACCESS SCHEMES 0ART  !IR INTERFACE REQUIREMENTS 4ECHNICAL 2EPORT 4DOC  %43) 34# 3-' MEETING NO  (ELSINKI &INLAND -AY 



0ART  %STIMATION OF 3UBSAMPLE 4IME $ELAY $IdERENCES

;= 7 $ 7ARNER AND # ,EUNG /&$-&- FRAME SYNCHRONIZATION FOR MOBILE RADIO DATA COMMUNICATION )%%% 4RANS 6EHIC 4ECHNOL    !UG  ;= , 7EI AND # 3CHLEGEL 3YNCHRONIZATION REQUIREMENTS FOR MULTI USER /&$- ON SATELLITE MOBILE AND TWO PATH 2AYLEIGH FADING CHANNELS )%%% 4RANS #OMMUN     &EB-AR!PR  ;= * 7EIGHT AND ! (AYMAN /BSERVATIONS OF THE PROPAGATION OF VERY SHORT ULTRASONIC PULSES AND THEIR REECTION BY SMALL TARGETS * !COUST 3OC !M     ;= 3 " 7EINSTEIN AND 0 - %BERT $ATA TRANSMISSION BY FREQUENCY DIVISION MULTIPLEXING USING THE DISCRETE &OURIER TRANSFORM )%%% 4RANS #OMMUN #/-    /CT  ;= 0 7ELLS "IOMEDICAL ULTRASONICS !CADEMIC 0RESS ,ONDON  ;= * * 7ERNER 4HE ($3, ENVIRONMENT )%%% * 3ELECT !REAS #OMMUN 3!#     !UG  ;= 3 ' 7ILSON $IGITAL MODULATION AND CODING 0RENTICE (ALL .EW *ERSEY 53!  ;= 3 + 7ILSON $IGITAL AUDIO BROADCASTING IN A FADING AND DISPERSIVE CHANNEL 0H$ THESIS 3TANFORD 5NIVERSITY #! !UG  ;= 3 + 7ILSON 2 % +HAYATA AND * - #IOb  1!- MODULATION WITH ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING IN A 2AYLEIGH FADING ENVIRONMENT )N 0ROC )%%% 6EHIC 4ECHNOL #ONF VOLUME  PAGES  3TOCKHOLM 3WEDEN *UNE  ;= 9 7U AND " #ARON $IGITAL TELEVISION TERRESTRIAL BROADCASTING )%%% #OMMUN -AG    -AY  ;= ' 9OUNG + 4 &OSTER AND * 7 #OOK "ROADBAND MULTIMEDIA DELIVERY OVER COPPER "4 4ECHNOL *OURN    /CT  ;= % :EHAVI AND * + 7OLF /N THE PERFORMANCE EVALUATION OF TRELLIS CODES )%%% 4RANS )NFORM 4HEORY )4    -AR  ;= ! :EIRA AND 0 3CHULTHEISS 2EALIZABLE LOWER BOUNDS FOR TIME DELAY ESTIMATION )%%% 4RANS 3IGNAL 0ROCESS     ;= 4 . :OGAKIS * 4 * !SLANIS AND * - #IOb !NALYSIS OF A CONCATENATED CODING SCHEME FOR A DISCRETE MULTITONE MODULATION SYSTEM )N 0ROC )%%% -ILITARY #OMMUN #ONF VOLUME  PAGES  ,ONG "RANCH .* /CT  ;= 4 . :OGAKIS 0 3 #HOW * 4 !SLANIS AND * - #IOb )MPULSE NOISE MITIGATION STRATEGIES FOR MULTICARRIER MODULATION )N 0ROC )NTERN #ONF #OMMUN PAGES  'ENEVA -AY  ;= 4 . :OGAKIS AND * - #IOb 4HE EdECTS OF TIMING JITTER ON THE PERFORMANCE OF A DISCRETE MULTITONE SYSTEM )%%% 4RANS #OMMUN    *ULY  ;= 7 9 :OU AND 9 7U #/&$- !N OVERVIEW )%%% 4RANS "ROADC    -AR