Chapter 5 Signal-Space Analysis

Size: px
Start display at page:

Download "Chapter 5 Signal-Space Analysis"

Transcription

1 Chaper 5 Sgnal-Space Analy Sgnal pace analy provde a mahemacally elegan and hghly nghful ool for he udy of daa ranmon. 5. Inroducon o Sacal model for a genec dgal communcaon yem n eage ource: A pror probable for nformaon ource p P m for,,..., n ranmer: he ranmer ae he meage ource oupu m and en-code no a dnc gnal uable for ranmon over he channel. So: p P m P for,,..., Po-ng Chen@ece.ncu Chaper 5-

2 5. Inroducon o mu be a real-valued energy gnal.e., a gnal wh fne energy wh duraon. E d <. n Channel: he channel aumed n h ex lnear and wh a bandwdh wde enough o pa wh no doron. o A zero-mean addve whe Gauan noe AWG alo aumed o faclae he analy. Po-ng Chen@ece.ncu Chaper A ahemacal odel o We can hen mplfy he prevou yem bloc dagram o: o Upon he recepon of x wh a duraon of, he recever mae he be emae of m. We haven defned wha he be mean. Po-ng Chen@ece.ncu Chaper 5-4

3 5. Creron for he Be Decon o Be nmzaon of he average probably of ymbol error. e P p P mˆ m n I opmum n he mnmum probably of error ene. n Baed on h creron, we can begn o degn he recever ha can gve he be decon. m Po-ng Chen@ece.ncu Chaper Geomerc Repreenaon of Sgnal o Sgnal pace concep n Vecorzaon of he dcree or connuou gnal remove he redundancy n gnal, and provde a compac repreenaon for hem. n Deermnaon of he vecorzaon ba o Gram-Schmd orhogonalzaon procedure Po-ng Chen@ece.ncu Chaper 5-6 3

4 5. Gram-Schmd Orhogonalzaon Procedure Gven v!, v!,, v!, how o fnd an orhonormal ba for hem? ep Le u!! v v!. ep! u '! v! v! u! u. Se! u ep For 3, 4,...,! u '! u '. Le u! ' v! v! u! u! v! u! u! " v! u! u!. Se u!!' u u! '.! ep v hen u, u!,", u! form an orhonormal ba. Po-ng Chen@ece.ncu Chaper 5-7 Propere: vecor : v! v,,v n nner produc :"! v! v n v v orhogonal, f nner produc. v norm : v! v +#+ v n v orhonormal, f nner produc, and ndvual norm. v lnearly ndependen, f none can be repreened a a lnear combnaon of oher. v rangle nequaly : v! + v! v! + v!. Po-ng Chen@ece.ncu Chaper 5-8 4

5 v Cauchy Schwarz nequaly :! v! v! v! v wh equaly holdng f! v a! v. x norm quare :! v +! v! v +! v +! v! v. x Pyhagorean propery : If orhogonal,! v +! v! v +! v. x arx ranformaon w.r.. marx A :! v A! v. x egenvalue w.r.. marx A: [ ]. oluon λ of de A-λI x egenvecor w.r.. egenvalue λ : oluon! v of A! v λ! v. Po-ng Chen@ece.ncu Chaper Sgnal Space Concep for Connuou Funcon Propere for connuou funcon complex valued gnal : z b nner produc : < z, ẑ > zẑ * d. orhogonal, f nner produc. v norm : z a b z d v orhonormal, f nner produc, and ndvual norm. Po-ng Chen@ece.ncu Chaper 5- a v lnearly ndependen vecor, f none can be repreened a a lnear combnaon of oher. v rangle nequaly : z + ẑ z + ẑ. 5

6 v Cauchy Schwarz nequaly : < z, ẑ > z ẑ wh equaly holdng f z a ẑ, where a a complex number. x norm quare : z+ ẑ z + ẑ + < z, ẑ > + < ẑ, z >. x Pyhagorean propery : If orhogonal, z+ ẑ z + ẑ. x ranformaon w.r.. a funcon C,τ : b ẑ C,τ zτ dτ Recall v a n v. a x.a egenvalue and egenfuncon w.r.. a funcon C,τ : oluon λ and {φ } of λ φ C,τ φ τ dτ and C,τ can be repreened a C,τ φ λ φ τ. Po-ng Chen@ece.ncu Chaper 5- n a b x.b Gve a deermnc funcon {, [, }and a e of x.c If orhonormal ba{ ψ } where a ψ d. orhonormal e { ψ } poble ha ˆ < ha can pan. hen a ψ, <, K < K doe no pan he pace, hen a ψ for all choce of { a }. < K Po-ng Chen@ece.ncu Chaper 5-6

7 o Problem : How o mnmze he energy of e ˆ? o elec he coeffcen{ a } ha mnmze e d a K [ a ψ ] d a a ψ ψ d + ψ d + a K K e d, [ ψ ] a d ψ ψ d. a ψ d. Po-ng Chen@ece.ncu Chaper 5-3 a, ψ a e ˆ, ψ ψ ψ Hence, e, ˆ e ˆ d o Inerpreaon n a he proecon of ono he Ψ -ax. n a he energy-proecon of ono he Ψ -ax. Po-ng Chen@ece.ncu Chaper 5-4 7

8 a, ψ a e ˆ, ψ ψ ψ e d e [ ˆ ] d e d d d d K K a a K e d e d a ψ d ψ d e ˆ d [ ˆ ] d K a oably, ˆ d. Po-ng Chen@ece.ncu Chaper Sgnal Space Concep for Connuou Funcon o Compleene n If every fne energy gnal afe { ψ } K n Example. Fourer ere defned over [, ]. d a complee orhonormal e. Po-ng Chen@ece.ncu Chaper 5-6 K π π co, n a complee orhonormal e for gnal a, 8

9 5. Gram-Schmd Orhogonalzaon Procedure Gven v, v,, v, how o fnd an orhonormal ba for hem? ep Le u v v. ' ep u v v,u u. Se u u ' ' u. ep For 3, 4,..., Le u ' v v,u u! v,u u. Se u u ' u '. form an orhonormal ba. ep v hen u,u,!,u Po-ng Chen@ece.ncu Chaper Geomerc Repreenaon of Sgnal f, <,,,..., { f } orhonormal f d,,,...,,,..., Po-ng Chen@ece.ncu Chaper 5-8 9

10 5. Geomerc Repreenaon of Sgnal o hrough he gnal pace concep, where can be unambguouly repreened by an -dmenonal gnal vecor,,, over an -dmenonal gnal pace. o he degn of ranmer become he elecon of pon over he gnal pace, and he recever mae a gue abou whch of he pon wa ranmed. o In he -dmenonal gnal pace, n Lengh quare of he vecor energy of he gnal n angle beween vecor energy correlaon beween gnal, co θ d Po-ng Chen@ece.ncu Chaper Geomerc Repreenaon of Sgnal n he angle beween vecor ndependen of he ba ued. o From h vew, n he ranmer may be vewed a a ynhezer, whch ynheze he ranmed gnal by a ban of mulpler. n he recever may be vewed a an analyzer, whch correlae produc-negrae he common npu no ndvdual nformaonal gnal. Po-ng Chen@ece.ncu Chaper 5-

11 5. Geomerc Repreenaon of Energy Sgnal o Illuraon he geomerc repreenaon of gnal for he cae of and 3 Po-ng Chen@ece.ncu Chaper 5-5. Eucldean Dance o Afer vecorzaon, we can hen calculae he Eucldean dance beween wo gnal, whch he quared roo of: d, Kronecer dela funcon : δ, Applcaon : We may ay ha orhonormaly mean φ, φ δ. Po-ng Chen@ece.ncu Chaper 5-

12 Example 5. Schwarz Inequaly o Cauchy-Schwarz nequaly and angle beween gnal n Cauchy-Schwarz nequaly ad ha, wh equaly holdng f c. n Alo, he angle beween gnal gve ha, co θ n Hence, Cauchy-Schwarz nequaly can be equvalenly aed a: coθ wh equaly holdng f θ or π Po-ng Chen@ece.ncu Chaper Ba o he complee orhonormal ba for a gnal pace no unque! n So he ynhezer and analyzer for he ranmon of he ame nformaonal meage are no unque! o One way o deermne a e of orhonormal ba he Gram-Schmd orhogonalzaon procedure. o ry and pracce Example 5. yourelf! Po-ng Chen@ece.ncu Chaper 5-4

13 5.3 Converon of he Connuou AWG Channel no a Vecor Channel o Influence of he AWG noe o he gnal pace concep x + w where w zero-mean AWG wh PSD /. o Afer he correlaor a he recever, we oban: x x, f, f w, f + n Or equvalenly, x + w. n oably, here no nformaon lo by he gnal pace repreenaon.,, f f x x Po-ng Chen@ece.ncu Chaper Converon of he Connuou AWG Channel no a Vecor Channel o Sac of {w } x w!! +! x w o Snce { } deermnc, he drbuon of x a meanhf of he drbuon of w. o Oberve ha w Gauan drbued becaue w AWG. he drbuon of w can herefore be deermned by mean vecor and covarance marx. Po-ng Chen@ece.ncu Chaper 5-6 3

14 4 Po-ng Chaper Converon of he Connuou AWG Channel no a Vecor Channel o ean o Covarance [ ] ] [ ] [ d f w E d f w E w E [ ] d f f dd f f dd f f w w E d f w d f w E w w E δ δ ] [ ] [ Po-ng Chen@ece.ncu Chaper Converon of he Connuou AWG Channel no a Vecor Channel o A a reul, [w, w,, w ] are zero-mean..d. Gauan drbued wh varance /. o h how ha x ndependen Gauan drbued wh common varance / and mean vecor [,,, ]. Equvalenly, x f exp π x

15 5.3 Converon of he Connuou AWG Channel no a Vecor Channel o Remander erm n noe n I poble ha w' w w f n However, can be hown ha a an error erm w orhogonal o for. Hence, w wll no affec he decon error rae on meage. w', wh probably. Po-ng Chen@ece.ncu Chaper Lelhood Funcon o An equvalen gnal-pace channel model m m ˆ, c m x + w m d x { m,..., m } o he be decon funcon d ha mnmze he decon error : d x m, f P{ m arg m x} P{ m max P{ m x} { m,..., } m x}for all n h he maxmum a poeror probably AP decon rule. Po-ng Chen@ece.ncu Chaper 5-3 5

16 5.4 Lelhood Funcon o Wh equal pror probable, dx arg max P{m x} m {m,...,m } { } argmax P{m x}, P{m x},..., P{m x} argmax P{m x} f x, P{m x} f x,..., P{m x} f x Pm Pm Pm argmax P{m x} f x, P{m x} f x,..., P{m x} f x / / / { } argmax f x m, f x m,..., f x m fx m named he lelhood funcon gven m ranmed Hence, he above rule named he maxmum-lelhood L decon rule. Po-ng Chen@ece.ncu Chaper Lelhood Funcon o AP rule L rule, f equal pror probably aumed. o In pracce, more convenen o wor on he loglelhood funcon, defned by d x arg max arg max { f x m }, f x m,..., f x m { log f x m,log f x m,...,log f x m } o Why log-lelhood funcon are more convenen? he decon funcon become um of quared Eucldean dance n AWG channel. Po-ng Chen@ece.ncu Chaper 5-3 6

17 dx argmax argmax argmax argmn log f x m argmax log f x log exp x π logπ x x argmn x argmn x Upon he recepon of receved gnal pon x, fnd he gnal pon ha cloe n Eucldean dance o x. Po-ng Chen@ece.ncu Chaper Coheren Deecon of Sgnal n oe: axmum Lelhood Decodng o Sgnal conellaon n he e of gnal pon n he gnal pace o Example. Sgnal conellaon for BQ code decon regon for decon regon for decon regon for 3 decon regon for 4 Po-ng Chen@ece.ncu Chaper

18 5.5 Coheren Deecon of Sgnal n oe: axmum Lelhood Decodng o Decon regon for and 4 Po-ng Chen@ece.ncu Chaper Coheren Deecon of Sgnal n oe: axmum Lelhood Decodng o Uually,,,, are named he meage pon. o he receved gnal pon x hen wander abou he ranmed meage pon n a Gauan-drbued random fahon. Po-ng Chen@ece.ncu Chaper

19 5.5 Coheren Deecon of Sgnal n oe: axmum Lelhood Decodng o Conan-energy gnal conellaon n In uch cae, he L decon rule can be reduced o an nner-produc. d x arg mn x arg mn arg mn arg max x, x x, + x, + E, f E conan. Po-ng Chen@ece.ncu Chaper Correlaon Recever o If gnal do no have equal energy, we can ue d x arg max x, E. o mplemen he L rule. n he recever coheren becaue he recever requre o be n perfec ynchronzaon wh he ranmer more pecfcally, he negraon mu begn a exacly he rgh me nance. Po-ng Chen@ece.ncu Chaper

20 producnegraor or correlaor x x Correlaon recever x demodulaor or deecor decon maer Po-ng Chen@ece.ncu Chaper 5-39 producnegraor or correlaor x x x mached fler decon maer Po-ng Chen@ece.ncu Chaper 5-4

21 Po-ng Chaper Equvalence of Correlaon and ached Fler Recever o he correlaor and mached fler can be made equvalen. o Specfcally, τ τ τ φ d h x d x x f h ϕ and mplcly ϕ zero oude. Po-ng Chen@ece.ncu Chaper Probably of Symbol Error o Average probably of ymbol error { } Z c e d f m m m d P m m d P m P P P, ranmed mn Pr ranmed ranmed x x x x x x { }. mn : where, Z x x x R

22 5.7 Invarance of Probably of Symbol Error o Probably of ymbol error nvaran wh repec o ba change.e., roaon and ranlaon of he gnal pace. o Specfcally, he ymbol error rae SER only depend on he relave Eucldean dance beween he meage pon. { x mn x m ranmed} P Pr e, Po-ng Chen@ece.ncu Chaper Invarance of Probably of Symbol Error o Specfcally, f Q a reverble ranform marx, uch a roaon, hen { x R : x mn x }, { x R : Qx Q mn } Qx Q, n If a gnal conellaon roaed by an orhonormal ranformaon, where Q an orhonormal marx, hen he probably of ymbol error P e ncurred n maxmum lelhood gnal deecon over an AWG channel compleely unchanged. Po-ng Chen@ece.ncu Chaper 5-44

23 5.7 Invarance of Probably of Symbol Error o A par of gnal conellaon for llurang he prncple of roaonal nvarance. Po-ng Chen@ece.ncu Chaper Invarance of Probably of Symbol Error o he nvarance n SER for ranlaon can be lewe proved. n I he ranmon power he ame for boh conellaon? Po-ng Chen@ece.ncu Chaper

24 5.7 nmum Energy Sgnal o Snce SER nvaran o roaon and ranlaon, we may roae and ranlae he gnal conellaon o mnmze he ranmon power whou affecng SER. E g p Fnd a and Q uch ha E a, Q p g Q a n Snce Q doe no change he norm.e., ranmon power, we only need o deermne he rgh a. mnmzed. Po-ng Chen@ece.ncu Chaper nmum Energy Sgnal o Deermne he opmal a. E a g p p a a + a p p a opmal a and E a g p + a opmal p p Po-ng Chen@ece.ncu Chaper

25 5.7 nmum Energy Sgnal o So ubfgure a below ha mnmum average energy. Po-ng Chen@ece.ncu Chaper Unon Bound on Probably of Error o Unon bound P e Pr Pr, P A B P A + P B { x mn x m ranmed} x x " "! x x x > x " "! x > x Pr Pr, { x > x m ranmed} Po-ng Chen@ece.ncu Chaper 5-5 m ranmed m ranmed 5

26 5.7 Unon Bound on Probably of Error x > x x > x 3 x > x 4 { x > x } { } x > x { x > x } 3 4 Po-ng Chen@ece.ncu Chaper Unon Bound on Probably of Error where P, Pr oably, gven m Pe P,, { x > x m ranmed }. ranmed, x Gauan drbued wh mean. w Po-ng Chen@ece.ncu Chaper 5-5 6

27 5.7 Unon Bound on Probably of Error Snce x + w when wa ranmed, we have Pr x Pr Pr Pr > x + w w w Pr Pr n < where n ranmed > + w > w + > w + w < ranmed ranmed + w ranmed ranmed ranmed w. Chaper 5-53 Po-ng Chen@ece.ncu 5.7 Unon Bound on Probably of Error Oberve ha w zero-mean Gauan drbued wh covarance marx E[ww ] I, where I he deny marx. Hence, n w Gauan drbued wh E[n] E w E[n ] E E w E[w] ww w E ww I. h mple ha w n/ Gauan drbued Chaper 5-54 wh mean zero and varance /. 7

28 5.7 Unon Bound on Probably of Error A a reul, Pr n< ranmed Pr w< ranmed Pr w< ranmed Pr w> ranmed, where he la equaly vald becaue he probably deny funcon of a zero-mean Gauan random varable ymmerc wh repec o w hence, Pr[w >a]pr[w< a] for any a>. Po-ng Chen@ece.ncu Chaper Unon Bound on Probably of Error o Hence, n For, P, Pr { x > x m ranmed} Pr w > d / π d erfc v exp dv, where d, where erfcu π u exp z dz. Po-ng Chen@ece.ncu Chaper

29 5.7 Unon Bound on Probably of Error erfc d, where erfcu exp z dz. u π n he ame formula vald for any. w n For, P, Pr { x > x m ranmed} Pr w > d and w d / exp v dv π don' care, where d Po-ng Chen@ece.ncu Chaper Unon Bound on Probably of Error o Conequenly, he unon bound for ymbol error rae : Pe P, erfc,, o he above bound can be furher mplfed when addonal condon gven. n For example, f he gnal conellaon crcularly ymmerc n he ene ha {d, d,, d } a permuaon of {d, d,, d } for, hen Pe erfc Po-ng Chen@ece.ncu, Chaper 5-58 d d 9

30 Pe 5.7 Unon Bound on Probably of Error o Anoher mplfcaon of unon bound n Defne he mnmum dance of a gnal conellaon a: d mn d mn,, hen by he rc decreang propery of erfc funcon,, d erfc d erfc Po-ng Chen@ece.ncu Chaper 5-59 erfc d mn dmn erfc erfc mn, d 5.7 Unon Bound on Probably of Error o We may ue he bound for erfc funcon o realze he relaon beween SER and d mn. exp u erfc u for u >.683 π d mn d mn P e erfc exp, f dmn > π n Concluon: SER decreae exponenally a he quared mnmum dance grow. Po-ng Chen@ece.ncu Chaper 5-6 3

31 5.7 Relaon beween BER and SER o he nformaon b are ranmed n group of log b o form an -ary ymbol. o h gve he reul ha a large ymbol error rae SER may no caue a large b error rae BER. n For example, a ymbol error for large may be due o only b error. n Opmcally, f every ymbol error due o a ngle b error, hen aumng ha n ymbol are ranmed n SER SER SER BER. n log log In general, BER. log Po-ng Chen@ece.ncu Chaper Relaon beween BER and SER n Pemcally, f every ymbol error caue log b error, hen aumng ha n ymbol are ranmed n log SER BER n log SER. In general, BER SER. n Summary: SER BER SER log Po-ng Chen@ece.ncu Chaper 5-6 3

32 5.7 Relaon beween BER and SER o If he ac for number of b error paern ha caue one ymbol error nown, we can hen deermne he exac relaon beween BER and SER. n SER # b P b BER n log where #b number of ' n b, and b repreen a bnary permuaon of log b paern. Here, a n b mean a b error occurred n he correpondng poon; hence, all-zero paern excluded becaue repreen no ymbol error. Po-ng Chen@ece.ncu Chaper Relaon beween BER and SER o Example. If all b error paer are equally lely, hen n SER # b P b SER BER n log log SER log u log u u log log SER log / SER # b / log oe u u u. Po-ng Chen@ece.ncu Chaper

Chapter 6 DETECTION AND ESTIMATION: Model of digital communication system. Fundamental issues in digital communications are

Chapter 6 DETECTION AND ESTIMATION: Model of digital communication system. Fundamental issues in digital communications are Chaper 6 DEECIO AD EIMAIO: Fundamenal ssues n dgal communcaons are. Deecon and. Esmaon Deecon heory: I deals wh he desgn and evaluaon of decson makng processor ha observes he receved sgnal and guesses

More information

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 4

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 4 CS434a/54a: Paern Recognon Prof. Olga Veksler Lecure 4 Oulne Normal Random Varable Properes Dscrmnan funcons Why Normal Random Varables? Analycally racable Works well when observaon comes form a corruped

More information

Control Systems. Mathematical Modeling of Control Systems.

Control Systems. Mathematical Modeling of Control Systems. Conrol Syem Mahemacal Modelng of Conrol Syem chbum@eoulech.ac.kr Oulne Mahemacal model and model ype. Tranfer funcon model Syem pole and zero Chbum Lee -Seoulech Conrol Syem Mahemacal Model Model are key

More information

(,,, ) (,,, ). In addition, there are three other consumers, -2, -1, and 0. Consumer -2 has the utility function

(,,, ) (,,, ). In addition, there are three other consumers, -2, -1, and 0. Consumer -2 has the utility function MACROECONOMIC THEORY T J KEHOE ECON 87 SPRING 5 PROBLEM SET # Conder an overlappng generaon economy le ha n queon 5 on problem e n whch conumer lve for perod The uly funcon of he conumer born n perod,

More information

Chapter 6 DETECTION AND ESTIMATION: Model of digital communication system. Fundamental issues in digital communications are

Chapter 6 DETECTION AND ESTIMATION: Model of digital communication system. Fundamental issues in digital communications are Chaper 6 DCIO AD IMAIO: Fndaenal sses n dgal concaons are. Deecon and. saon Deecon heory: I deals wh he desgn and evalaon of decson ang processor ha observes he receved sgnal and gesses whch parclar sybol

More information

Lecture 11: Stereo and Surface Estimation

Lecture 11: Stereo and Surface Estimation Lecure : Sereo and Surface Emaon When camera poon have been deermned, ung rucure from moon, we would lke o compue a dene urface model of he cene. In h lecure we wll udy he o called Sereo Problem, where

More information

. The geometric multiplicity is dim[ker( λi. number of linearly independent eigenvectors associated with this eigenvalue.

. The geometric multiplicity is dim[ker( λi. number of linearly independent eigenvectors associated with this eigenvalue. Lnear Algebra Lecure # Noes We connue wh he dscusson of egenvalues, egenvecors, and dagonalzably of marces We wan o know, n parcular wha condons wll assure ha a marx can be dagonalzed and wha he obsrucons

More information

. The geometric multiplicity is dim[ker( λi. A )], i.e. the number of linearly independent eigenvectors associated with this eigenvalue.

. The geometric multiplicity is dim[ker( λi. A )], i.e. the number of linearly independent eigenvectors associated with this eigenvalue. Mah E-b Lecure #0 Noes We connue wh he dscusson of egenvalues, egenvecors, and dagonalzably of marces We wan o know, n parcular wha condons wll assure ha a marx can be dagonalzed and wha he obsrucons are

More information

Dynamic Team Decision Theory. EECS 558 Project Shrutivandana Sharma and David Shuman December 10, 2005

Dynamic Team Decision Theory. EECS 558 Project Shrutivandana Sharma and David Shuman December 10, 2005 Dynamc Team Decson Theory EECS 558 Proec Shruvandana Sharma and Davd Shuman December 0, 005 Oulne Inroducon o Team Decson Theory Decomposon of he Dynamc Team Decson Problem Equvalence of Sac and Dynamc

More information

( ) [ ] MAP Decision Rule

( ) [ ] MAP Decision Rule Announcemens Bayes Decson Theory wh Normal Dsrbuons HW0 due oday HW o be assgned soon Proec descrpon posed Bomercs CSE 90 Lecure 4 CSE90, Sprng 04 CSE90, Sprng 04 Key Probables 4 ω class label X feaure

More information

How about the more general "linear" scalar functions of scalars (i.e., a 1st degree polynomial of the following form with a constant term )?

How about the more general linear scalar functions of scalars (i.e., a 1st degree polynomial of the following form with a constant term )? lmcd Lnear ransformaon of a vecor he deas presened here are que general hey go beyond he radonal mar-vecor ype seen n lnear algebra Furhermore, hey do no deal wh bass and are equally vald for any se of

More information

A Demand System for Input Factors when there are Technological Changes in Production

A Demand System for Input Factors when there are Technological Changes in Production A Demand Syem for Inpu Facor when here are Technologcal Change n Producon Movaon Due o (e.g.) echnologcal change here mgh no be a aonary relaonhp for he co hare of each npu facor. When emang demand yem

More information

Normal Random Variable and its discriminant functions

Normal Random Variable and its discriminant functions Noral Rando Varable and s dscrnan funcons Oulne Noral Rando Varable Properes Dscrnan funcons Why Noral Rando Varables? Analycally racable Works well when observaon coes for a corruped snle prooype 3 The

More information

A. Inventory model. Why are we interested in it? What do we really study in such cases.

A. Inventory model. Why are we interested in it? What do we really study in such cases. Some general yem model.. Inenory model. Why are we nereed n? Wha do we really udy n uch cae. General raegy of machng wo dmlar procee, ay, machng a fa proce wh a low one. We need an nenory or a buffer or

More information

Scattering at an Interface: Oblique Incidence

Scattering at an Interface: Oblique Incidence Course Insrucor Dr. Raymond C. Rumpf Offce: A 337 Phone: (915) 747 6958 E Mal: rcrumpf@uep.edu EE 4347 Appled Elecromagnecs Topc 3g Scaerng a an Inerface: Oblque Incdence Scaerng These Oblque noes may

More information

Laplace Transformation of Linear Time-Varying Systems

Laplace Transformation of Linear Time-Varying Systems Laplace Tranformaon of Lnear Tme-Varyng Syem Shervn Erfan Reearch Cenre for Inegraed Mcroelecronc Elecrcal and Compuer Engneerng Deparmen Unvery of Wndor Wndor, Onaro N9B 3P4, Canada Aug. 4, 9 Oulne of

More information

Lecture 6: Learning for Control (Generalised Linear Regression)

Lecture 6: Learning for Control (Generalised Linear Regression) Lecure 6: Learnng for Conrol (Generalsed Lnear Regresson) Conens: Lnear Mehods for Regresson Leas Squares, Gauss Markov heorem Recursve Leas Squares Lecure 6: RLSC - Prof. Sehu Vjayakumar Lnear Regresson

More information

Chapter 6: AC Circuits

Chapter 6: AC Circuits Chaper 6: AC Crcus Chaper 6: Oulne Phasors and he AC Seady Sae AC Crcus A sable, lnear crcu operang n he seady sae wh snusodal excaon (.e., snusodal seady sae. Complee response forced response naural response.

More information

Cooling of a hot metal forging. , dt dt

Cooling of a hot metal forging. , dt dt Tranen Conducon Uneady Analy - Lumped Thermal Capacy Model Performed when; Hea ranfer whn a yem produced a unform emperaure drbuon n he yem (mall emperaure graden). The emperaure change whn he yem condered

More information

Robustness Experiments with Two Variance Components

Robustness Experiments with Two Variance Components Naonal Insue of Sandards and Technology (NIST) Informaon Technology Laboraory (ITL) Sascal Engneerng Dvson (SED) Robusness Expermens wh Two Varance Componens by Ana Ivelsse Avlés avles@ns.gov Conference

More information

John Geweke a and Gianni Amisano b a Departments of Economics and Statistics, University of Iowa, USA b European Central Bank, Frankfurt, Germany

John Geweke a and Gianni Amisano b a Departments of Economics and Statistics, University of Iowa, USA b European Central Bank, Frankfurt, Germany Herarchcal Markov Normal Mxure models wh Applcaons o Fnancal Asse Reurns Appendx: Proofs of Theorems and Condonal Poseror Dsrbuons John Geweke a and Gann Amsano b a Deparmens of Economcs and Sascs, Unversy

More information

Comb Filters. Comb Filters

Comb Filters. Comb Filters The smple flers dscussed so far are characered eher by a sngle passband and/or a sngle sopband There are applcaons where flers wh mulple passbands and sopbands are requred Thecomb fler s an example of

More information

CH.3. COMPATIBILITY EQUATIONS. Continuum Mechanics Course (MMC) - ETSECCPB - UPC

CH.3. COMPATIBILITY EQUATIONS. Continuum Mechanics Course (MMC) - ETSECCPB - UPC CH.3. COMPATIBILITY EQUATIONS Connuum Mechancs Course (MMC) - ETSECCPB - UPC Overvew Compably Condons Compably Equaons of a Poenal Vecor Feld Compably Condons for Infnesmal Srans Inegraon of he Infnesmal

More information

CS286.2 Lecture 14: Quantum de Finetti Theorems II

CS286.2 Lecture 14: Quantum de Finetti Theorems II CS286.2 Lecure 14: Quanum de Fne Theorems II Scrbe: Mara Okounkova 1 Saemen of he heorem Recall he las saemen of he quanum de Fne heorem from he prevous lecure. Theorem 1 Quanum de Fne). Le ρ Dens C 2

More information

H = d d q 1 d d q N d d p 1 d d p N exp

H = d d q 1 d d q N d d p 1 d d p N exp 8333: Sacal Mechanc I roblem Se # 7 Soluon Fall 3 Canoncal Enemble Non-harmonc Ga: The Hamlonan for a ga of N non neracng parcle n a d dmenonal box ha he form H A p a The paron funcon gven by ZN T d d

More information

Matrix reconstruction with the local max norm

Matrix reconstruction with the local max norm Marx reconrucon wh he local max norm Rna oygel Deparmen of Sac Sanford Unvery rnafb@anfordedu Nahan Srebro Toyoa Technologcal Inue a Chcago na@cedu Rulan Salakhudnov Dep of Sac and Dep of Compuer Scence

More information

Lecture VI Regression

Lecture VI Regression Lecure VI Regresson (Lnear Mehods for Regresson) Conens: Lnear Mehods for Regresson Leas Squares, Gauss Markov heorem Recursve Leas Squares Lecure VI: MLSC - Dr. Sehu Vjayakumar Lnear Regresson Model M

More information

Bayes rule for a classification problem INF Discriminant functions for the normal density. Euclidean distance. Mahalanobis distance

Bayes rule for a classification problem INF Discriminant functions for the normal density. Euclidean distance. Mahalanobis distance INF 43 3.. Repeon Anne Solberg (anne@f.uo.no Bayes rule for a classfcaon problem Suppose we have J, =,...J classes. s he class label for a pxel, and x s he observed feaure vecor. We can use Bayes rule

More information

Sklar: Sections (4.4.2 is not covered).

Sklar: Sections (4.4.2 is not covered). COSC 44: Dgal Councaons Insrucor: Dr. Ar Asf Deparen of Copuer Scence and Engneerng York Unversy Handou # 6: Bandpass Modulaon opcs:. Phasor Represenaon. Dgal Modulaon Schees: PSK FSK ASK APK ASK/FSK)

More information

An introduction to Support Vector Machine

An introduction to Support Vector Machine An nroducon o Suppor Vecor Machne 報告者 : 黃立德 References: Smon Haykn, "Neural Neworks: a comprehensve foundaon, second edon, 999, Chaper 2,6 Nello Chrsann, John Shawe-Tayer, An Inroducon o Suppor Vecor Machnes,

More information

Notes on the stability of dynamic systems and the use of Eigen Values.

Notes on the stability of dynamic systems and the use of Eigen Values. Noes on he sabl of dnamc ssems and he use of Egen Values. Source: Macro II course noes, Dr. Davd Bessler s Tme Seres course noes, zarads (999) Ineremporal Macroeconomcs chaper 4 & Techncal ppend, and Hamlon

More information

Lecture 11 SVM cont

Lecture 11 SVM cont Lecure SVM con. 0 008 Wha we have done so far We have esalshed ha we wan o fnd a lnear decson oundary whose margn s he larges We know how o measure he margn of a lnear decson oundary Tha s: he mnmum geomerc

More information

Fundamentals of PLLs (I)

Fundamentals of PLLs (I) Phae-Locked Loop Fundamenal of PLL (I) Chng-Yuan Yang Naonal Chung-Hng Unvery Deparmen of Elecrcal Engneerng Why phae-lock? - Jer Supreon - Frequency Synhe T T + 1 - Skew Reducon T + 2 T + 3 PLL fou =

More information

ANALYSIS AND MODELING OF HYDROLOGIC TIME SERIES. Wasserhaushalt Time Series Analysis and Stochastic Modelling Spring Semester

ANALYSIS AND MODELING OF HYDROLOGIC TIME SERIES. Wasserhaushalt Time Series Analysis and Stochastic Modelling Spring Semester ANALYSIS AND MODELING OF HYDROLOGIC TIME SERIES Waerhauhal Tme Sere Analy and Sochac Modellng Sprng Semeer 8 ANALYSIS AND MODELING OF HYDROLOGIC TIME SERIES Defnon Wha a me ere? Leraure: Sala, J.D. 99,

More information

Online Appendix for. Strategic safety stocks in supply chains with evolving forecasts

Online Appendix for. Strategic safety stocks in supply chains with evolving forecasts Onlne Appendx for Sraegc safey socs n supply chans wh evolvng forecass Tor Schoenmeyr Sephen C. Graves Opsolar, Inc. 332 Hunwood Avenue Hayward, CA 94544 A. P. Sloan School of Managemen Massachuses Insue

More information

Motion of Wavepackets in Non-Hermitian. Quantum Mechanics

Motion of Wavepackets in Non-Hermitian. Quantum Mechanics Moon of Wavepaces n Non-Herman Quanum Mechancs Nmrod Moseyev Deparmen of Chemsry and Mnerva Cener for Non-lnear Physcs of Complex Sysems, Technon-Israel Insue of Technology www.echnon echnon.ac..ac.l\~nmrod

More information

Linear Response Theory: The connection between QFT and experiments

Linear Response Theory: The connection between QFT and experiments Phys540.nb 39 3 Lnear Response Theory: The connecon beween QFT and expermens 3.1. Basc conceps and deas Q: ow do we measure he conducvy of a meal? A: we frs nroduce a weak elecrc feld E, and hen measure

More information

Graduate Macroeconomics 2 Problem set 5. - Solutions

Graduate Macroeconomics 2 Problem set 5. - Solutions Graduae Macroeconomcs 2 Problem se. - Soluons Queson 1 To answer hs queson we need he frms frs order condons and he equaon ha deermnes he number of frms n equlbrum. The frms frs order condons are: F K

More information

L N O Q. l q l q. I. A General Case. l q RANDOM LAGRANGE MULTIPLIERS AND TRANSVERSALITY. Econ. 511b Spring 1998 C. Sims

L N O Q. l q l q. I. A General Case. l q RANDOM LAGRANGE MULTIPLIERS AND TRANSVERSALITY. Econ. 511b Spring 1998 C. Sims Econ. 511b Sprng 1998 C. Sm RAD AGRAGE UPERS AD RASVERSAY agrange mulpler mehod are andard fare n elemenary calculu coure, and hey play a cenral role n economc applcaon of calculu becaue hey ofen urn ou

More information

Chapter Lagrangian Interpolation

Chapter Lagrangian Interpolation Chaper 5.4 agrangan Inerpolaon Afer readng hs chaper you should be able o:. dere agrangan mehod of nerpolaon. sole problems usng agrangan mehod of nerpolaon and. use agrangan nerpolans o fnd deraes and

More information

Communication with AWGN Interference

Communication with AWGN Interference Communcaton wth AWG Interference m {m } {p(m } Modulator s {s } r=s+n Recever ˆm AWG n m s a dscrete random varable(rv whch takes m wth probablty p(m. Modulator maps each m nto a waveform sgnal s m=m

More information

Epistemic Game Theory: Online Appendix

Epistemic Game Theory: Online Appendix Epsemc Game Theory: Onlne Appendx Edde Dekel Lucano Pomao Marcano Snscalch July 18, 2014 Prelmnares Fx a fne ype srucure T I, S, T, β I and a probably µ S T. Le T µ I, S, T µ, βµ I be a ype srucure ha

More information

Solution in semi infinite diffusion couples (error function analysis)

Solution in semi infinite diffusion couples (error function analysis) Soluon n sem nfne dffuson couples (error funcon analyss) Le us consder now he sem nfne dffuson couple of wo blocks wh concenraon of and I means ha, n a A- bnary sysem, s bondng beween wo blocks made of

More information

Advanced time-series analysis (University of Lund, Economic History Department)

Advanced time-series analysis (University of Lund, Economic History Department) Advanced me-seres analss (Unvers of Lund, Economc Hsor Dearmen) 3 Jan-3 Februar and 6-3 March Lecure 4 Economerc echnues for saonar seres : Unvarae sochasc models wh Box- Jenns mehodolog, smle forecasng

More information

Multiple Failures. Diverse Routing for Maximizing Survivability. Maximum Survivability Models. Minimum-Color (SRLG) Diverse Routing

Multiple Failures. Diverse Routing for Maximizing Survivability. Maximum Survivability Models. Minimum-Color (SRLG) Diverse Routing Mulple Falure Dvere Roung for Maxmzng Survvably One-falure aumpon n prevou work Mulple falure Hard o provde 100% proecon Maxmum urvvably Maxmum Survvably Model Mnmum-Color (SRLG) Dvere Roung Each lnk ha

More information

Multiple Regressions and Correlation Analysis

Multiple Regressions and Correlation Analysis Mulple Regreon and Correlaon Analy Chaper 4 McGraw-Hll/Irwn Copyrgh 2 y The McGraw-Hll Compane, Inc. All rgh reerved. GOALS. Decre he relaonhp eween everal ndependen varale and a dependen varale ung mulple

More information

Maximum Likelihood Estimation

Maximum Likelihood Estimation Mau Lkelhood aon Beln Chen Depaen of Copue Scence & Infoaon ngneeng aonal Tawan oal Unvey Refeence:. he Alpaydn, Inoducon o Machne Leanng, Chape 4, MIT Pe, 4 Saple Sac and Populaon Paaee A Scheac Depcon

More information

Bandlimited channel. Intersymbol interference (ISI) This non-ideal communication channel is also called dispersive channel

Bandlimited channel. Intersymbol interference (ISI) This non-ideal communication channel is also called dispersive channel Inersymol nererence ISI ISI s a sgnal-dependen orm o nererence ha arses ecause o devaons n he requency response o a channel rom he deal channel. Example: Bandlmed channel Tme Doman Bandlmed channel Frequency

More information

Online Supplement for Dynamic Multi-Technology. Production-Inventory Problem with Emissions Trading

Online Supplement for Dynamic Multi-Technology. Production-Inventory Problem with Emissions Trading Onlne Supplemen for Dynamc Mul-Technology Producon-Invenory Problem wh Emssons Tradng by We Zhang Zhongsheng Hua Yu Xa and Baofeng Huo Proof of Lemma For any ( qr ) Θ s easy o verfy ha he lnear programmng

More information

Part II CONTINUOUS TIME STOCHASTIC PROCESSES

Part II CONTINUOUS TIME STOCHASTIC PROCESSES Par II CONTINUOUS TIME STOCHASTIC PROCESSES 4 Chaper 4 For an advanced analyss of he properes of he Wener process, see: Revus D and Yor M: Connuous marngales and Brownan Moon Karazas I and Shreve S E:

More information

( ) () we define the interaction representation by the unitary transformation () = ()

( ) () we define the interaction representation by the unitary transformation () = () Hgher Order Perurbaon Theory Mchael Fowler 3/7/6 The neracon Represenaon Recall ha n he frs par of hs course sequence, we dscussed he chrödnger and Hesenberg represenaons of quanum mechancs here n he chrödnger

More information

Ordinary Differential Equations in Neuroscience with Matlab examples. Aim 1- Gain understanding of how to set up and solve ODE s

Ordinary Differential Equations in Neuroscience with Matlab examples. Aim 1- Gain understanding of how to set up and solve ODE s Ordnary Dfferenal Equaons n Neuroscence wh Malab eamples. Am - Gan undersandng of how o se up and solve ODE s Am Undersand how o se up an solve a smple eample of he Hebb rule n D Our goal a end of class

More information

st semester. Kei Sakaguchi. ee ac May. 10, 2011

st semester. Kei Sakaguchi. ee ac May. 10, 2011 0 s semeser IO Communcaon Sysems #4: Array Sgnal Processng Ke Sakaguc ee ac ay. 0, 0 Scedule s alf Dae Tex Conens # Apr. A-, B- Inroducon # Apr. 9 B-5, B-6 Fundamenals of wreless

More information

Discussion Session 2 Constant Acceleration/Relative Motion Week 03

Discussion Session 2 Constant Acceleration/Relative Motion Week 03 PHYS 100 Dicuion Seion Conan Acceleraion/Relaive Moion Week 03 The Plan Today you will work wih your group explore he idea of reference frame (i.e. relaive moion) and moion wih conan acceleraion. You ll

More information

Discrete Markov Process. Introduction. Example: Balls and Urns. Stochastic Automaton. INTRODUCTION TO Machine Learning 3rd Edition

Discrete Markov Process. Introduction. Example: Balls and Urns. Stochastic Automaton. INTRODUCTION TO Machine Learning 3rd Edition EHEM ALPAYDI he MI Press, 04 Lecure Sldes for IRODUCIO O Machne Learnng 3rd Edon alpaydn@boun.edu.r hp://www.cmpe.boun.edu.r/~ehem/ml3e Sldes from exboo resource page. Slghly eded and wh addonal examples

More information

Robust and Accurate Cancer Classification with Gene Expression Profiling

Robust and Accurate Cancer Classification with Gene Expression Profiling Robus and Accurae Cancer Classfcaon wh Gene Expresson Proflng (Compuaonal ysems Bology, 2005) Auhor: Hafeng L, Keshu Zhang, ao Jang Oulne Background LDA (lnear dscrmnan analyss) and small sample sze problem

More information

, the. L and the L. x x. max. i n. It is easy to show that these two norms satisfy the following relation: x x n x = (17.3) max

, the. L and the L. x x. max. i n. It is easy to show that these two norms satisfy the following relation: x x n x = (17.3) max ecure 8 7. Sabiliy Analyi For an n dimenional vecor R n, he and he vecor norm are defined a: = T = i n i (7.) I i eay o how ha hee wo norm aify he following relaion: n (7.) If a vecor i ime-dependen, hen

More information

HEAT CONDUCTION PROBLEM IN A TWO-LAYERED HOLLOW CYLINDER BY USING THE GREEN S FUNCTION METHOD

HEAT CONDUCTION PROBLEM IN A TWO-LAYERED HOLLOW CYLINDER BY USING THE GREEN S FUNCTION METHOD Journal of Appled Mahemacs and Compuaonal Mechancs 3, (), 45-5 HEAT CONDUCTION PROBLEM IN A TWO-LAYERED HOLLOW CYLINDER BY USING THE GREEN S FUNCTION METHOD Sansław Kukla, Urszula Sedlecka Insue of Mahemacs,

More information

OP = OO' + Ut + Vn + Wb. Material We Will Cover Today. Computer Vision Lecture 3. Multi-view Geometry I. Amnon Shashua

OP = OO' + Ut + Vn + Wb. Material We Will Cover Today. Computer Vision Lecture 3. Multi-view Geometry I. Amnon Shashua Comuer Vson 27 Lecure 3 Mul-vew Geomer I Amnon Shashua Maeral We Wll Cover oa he srucure of 3D->2D rojecon mar omograh Marces A rmer on rojecve geomer of he lane Eolar Geomer an Funamenal Mar ebrew Unvers

More information

V.Abramov - FURTHER ANALYSIS OF CONFIDENCE INTERVALS FOR LARGE CLIENT/SERVER COMPUTER NETWORKS

V.Abramov - FURTHER ANALYSIS OF CONFIDENCE INTERVALS FOR LARGE CLIENT/SERVER COMPUTER NETWORKS R&RATA # Vol.) 8, March FURTHER AALYSIS OF COFIDECE ITERVALS FOR LARGE CLIET/SERVER COMPUTER ETWORKS Vyacheslav Abramov School of Mahemacal Scences, Monash Unversy, Buldng 8, Level 4, Clayon Campus, Wellngon

More information

Advanced Machine Learning & Perception

Advanced Machine Learning & Perception Advanced Machne Learnng & Percepon Insrucor: Tony Jebara SVM Feaure & Kernel Selecon SVM Eensons Feaure Selecon (Flerng and Wrappng) SVM Feaure Selecon SVM Kernel Selecon SVM Eensons Classfcaon Feaure/Kernel

More information

China s Model of Managing the Financial System

China s Model of Managing the Financial System Chna odel of anagng he Fnancal Syem arku K Brunnermeer chael Sockn We Xong Inerne Appendx Th nerne appendx preen proof of he propoon n he man paper Proof of Propoon A We dere he perfec nformaon equlbrum

More information

The multivariate Gaussian probability density function for random vector X (X 1,,X ) T. diagonal term of, denoted

The multivariate Gaussian probability density function for random vector X (X 1,,X ) T. diagonal term of, denoted Appendx Proof of heorem he multvarate Gauan probablty denty functon for random vector X (X,,X ) px exp / / x x mean and varance equal to the th dagonal term of, denoted he margnal dtrbuton of X Gauan wth

More information

Mechanics Physics 151

Mechanics Physics 151 Mechancs Physcs 5 Lecure 0 Canoncal Transformaons (Chaper 9) Wha We Dd Las Tme Hamlon s Prncple n he Hamlonan formalsm Dervaon was smple δi δ Addonal end-pon consrans pq H( q, p, ) d 0 δ q ( ) δq ( ) δ

More information

Numerical Study of Large-area Anti-Resonant Reflecting Optical Waveguide (ARROW) Vertical-Cavity Semiconductor Optical Amplifiers (VCSOAs)

Numerical Study of Large-area Anti-Resonant Reflecting Optical Waveguide (ARROW) Vertical-Cavity Semiconductor Optical Amplifiers (VCSOAs) USOD 005 uecal Sudy of Lage-aea An-Reonan Reflecng Opcal Wavegude (ARROW Vecal-Cavy Seconduco Opcal Aplfe (VCSOA anhu Chen Su Fung Yu School of Eleccal and Eleconc Engneeng Conen Inoducon Vecal Cavy Seconduco

More information

The Signal, Variable System, and Transformation: A Personal Perspective

The Signal, Variable System, and Transformation: A Personal Perspective The Sgal Varable Syem ad Traformao: A Peroal Perpecve Sherv Erfa 35 Eex Hall Faculy of Egeerg Oule Of he Talk Iroduco Mahemacal Repreeao of yem Operaor Calculu Traformao Obervao O Laplace Traform SSB A

More information

Let s treat the problem of the response of a system to an applied external force. Again,

Let s treat the problem of the response of a system to an applied external force. Again, Page 33 QUANTUM LNEAR RESPONSE FUNCTON Le s rea he problem of he response of a sysem o an appled exernal force. Agan, H() H f () A H + V () Exernal agen acng on nernal varable Hamlonan for equlbrum sysem

More information

Response of MDOF systems

Response of MDOF systems Response of MDOF syses Degree of freedo DOF: he nu nuber of ndependen coordnaes requred o deerne copleely he posons of all pars of a syse a any nsan of e. wo DOF syses hree DOF syses he noral ode analyss

More information

Clustering (Bishop ch 9)

Clustering (Bishop ch 9) Cluserng (Bshop ch 9) Reference: Daa Mnng by Margare Dunham (a slde source) 1 Cluserng Cluserng s unsupervsed learnng, here are no class labels Wan o fnd groups of smlar nsances Ofen use a dsance measure

More information

Let us start with a two dimensional case. We consider a vector ( x,

Let us start with a two dimensional case. We consider a vector ( x, Roaion marices We consider now roaion marices in wo and hree dimensions. We sar wih wo dimensions since wo dimensions are easier han hree o undersand, and one dimension is a lile oo simple. However, our

More information

Should Exact Index Numbers have Standard Errors? Theory and Application to Asian Growth

Should Exact Index Numbers have Standard Errors? Theory and Application to Asian Growth Should Exac Index umbers have Sandard Errors? Theory and Applcaon o Asan Growh Rober C. Feensra Marshall B. Rensdorf ovember 003 Proof of Proposon APPEDIX () Frs, we wll derve he convenonal Sao-Vara prce

More information

Linear Algebra Primer

Linear Algebra Primer Linear Algebra rimer And a video dicuion of linear algebra from EE263 i here (lecure 3 and 4): hp://ee.anford.edu/coure/ee263 lide from Sanford CS3 Ouline Vecor and marice Baic Mari Operaion Deerminan,

More information

FTCS Solution to the Heat Equation

FTCS Solution to the Heat Equation FTCS Soluon o he Hea Equaon ME 448/548 Noes Gerald Reckenwald Porland Sae Unversy Deparmen of Mechancal Engneerng gerry@pdxedu ME 448/548: FTCS Soluon o he Hea Equaon Overvew Use he forward fne d erence

More information

Physics 240: Worksheet 16 Name

Physics 240: Worksheet 16 Name Phyic 4: Workhee 16 Nae Non-unifor circular oion Each of hee proble involve non-unifor circular oion wih a conan α. (1) Obain each of he equaion of oion for non-unifor circular oion under a conan acceleraion,

More information

Introduction to Boosting

Introduction to Boosting Inroducon o Boosng Cynha Rudn PACM, Prnceon Unversy Advsors Ingrd Daubeches and Rober Schapre Say you have a daabase of news arcles, +, +, -, -, +, +, -, -, +, +, -, -, +, +, -, + where arcles are labeled

More information

Comparison of Differences between Power Means 1

Comparison of Differences between Power Means 1 In. Journal of Mah. Analyss, Vol. 7, 203, no., 5-55 Comparson of Dfferences beween Power Means Chang-An Tan, Guanghua Sh and Fe Zuo College of Mahemacs and Informaon Scence Henan Normal Unversy, 453007,

More information

( t) Outline of program: BGC1: Survival and event history analysis Oslo, March-May Recapitulation. The additive regression model

( t) Outline of program: BGC1: Survival and event history analysis Oslo, March-May Recapitulation. The additive regression model BGC1: Survval and even hsory analyss Oslo, March-May 212 Monday May 7h and Tuesday May 8h The addve regresson model Ørnulf Borgan Deparmen of Mahemacs Unversy of Oslo Oulne of program: Recapulaon Counng

More information

In the complete model, these slopes are ANALYSIS OF VARIANCE FOR THE COMPLETE TWO-WAY MODEL. (! i+1 -! i ) + [(!") i+1,q - [(!

In the complete model, these slopes are ANALYSIS OF VARIANCE FOR THE COMPLETE TWO-WAY MODEL. (! i+1 -! i ) + [(!) i+1,q - [(! ANALYSIS OF VARIANCE FOR THE COMPLETE TWO-WAY MODEL The frs hng o es n wo-way ANOVA: Is here neracon? "No neracon" means: The man effecs model would f. Ths n urn means: In he neracon plo (wh A on he horzonal

More information

CHAPTER 5: MULTIVARIATE METHODS

CHAPTER 5: MULTIVARIATE METHODS CHAPER 5: MULIVARIAE MEHODS Mulvarae Daa 3 Mulple measuremens (sensors) npus/feaures/arbues: -varae N nsances/observaons/eamples Each row s an eample Each column represens a feaure X a b correspons o he

More information

Joint Channel Estimation and Resource Allocation for MIMO Systems Part I: Single-User Analysis

Joint Channel Estimation and Resource Allocation for MIMO Systems Part I: Single-User Analysis 624 IEEE RANSACIONS ON WIRELESS COUNICAIONS, VOL. 9, NO. 2, FEBRUARY 200 Jon Channel Esmaon and Resource Allocaon for IO Sysems Par I: Sngle-User Analyss Alkan Soysal, ember, IEEE, and Sennur Ulukus, ember,

More information

Model-Based FDI : the control approach

Model-Based FDI : the control approach Model-Baed FDI : he conrol approach M. Saroweck LAIL-CNRS EUDIL, Unver Llle I Olne of he preenaon Par I : model Sem, normal and no normal condon, fal Par II : he decon problem problem eng noe, drbance,

More information

CS344: Introduction to Artificial Intelligence

CS344: Introduction to Artificial Intelligence C344: Iroduco o Arfcal Iellgece Puhpa Bhaacharyya CE Dep. IIT Bombay Lecure 3 3 32 33: Forward ad bacward; Baum elch 9 h ad 2 March ad 2 d Aprl 203 Lecure 27 28 29 were o EM; dae 2 h March o 8 h March

More information

Equalization on Graphs: Linear Programming and Message Passing

Equalization on Graphs: Linear Programming and Message Passing Equalzaon on Graphs: Lnear Programmng and Message Passng Mohammad H. Taghav and Paul H. Segel Cener for Magnec Recordng Research Unversy of Calforna, San Dego La Jolla, CA 92093-0401, USA Emal: (maghav,

More information

DEEP UNFOLDING FOR MULTICHANNEL SOURCE SEPARATION SUPPLEMENTARY MATERIAL

DEEP UNFOLDING FOR MULTICHANNEL SOURCE SEPARATION SUPPLEMENTARY MATERIAL DEEP UNFOLDING FOR MULTICHANNEL SOURCE SEPARATION SUPPLEMENTARY MATERIAL Sco Wsdom, John Hershey 2, Jonahan Le Roux 2, and Shnj Waanabe 2 Deparmen o Elecrcal Engneerng, Unversy o Washngon, Seale, WA, USA

More information

Introduction ( Week 1-2) Course introduction A brief introduction to molecular biology A brief introduction to sequence comparison Part I: Algorithms

Introduction ( Week 1-2) Course introduction A brief introduction to molecular biology A brief introduction to sequence comparison Part I: Algorithms Course organzaon Inroducon Wee -2) Course nroducon A bref nroducon o molecular bology A bref nroducon o sequence comparson Par I: Algorhms for Sequence Analyss Wee 3-8) Chaper -3, Models and heores» Probably

More information

Math 128b Project. Jude Yuen

Math 128b Project. Jude Yuen Mah 8b Proec Jude Yuen . Inroducon Le { Z } be a sequence of observed ndependen vecor varables. If he elemens of Z have a on normal dsrbuon hen { Z } has a mean vecor Z and a varancecovarance marx z. Geomercally

More information

Econ107 Applied Econometrics Topic 5: Specification: Choosing Independent Variables (Studenmund, Chapter 6)

Econ107 Applied Econometrics Topic 5: Specification: Choosing Independent Variables (Studenmund, Chapter 6) Econ7 Appled Economercs Topc 5: Specfcaon: Choosng Independen Varables (Sudenmund, Chaper 6 Specfcaon errors ha we wll deal wh: wrong ndependen varable; wrong funconal form. Ths lecure deals wh wrong ndependen

More information

XMAP: Track-to-Track Association with Metric, Feature, and Target-type Data

XMAP: Track-to-Track Association with Metric, Feature, and Target-type Data XMAP: Track-o-Track Aocaon wh Merc, Feaure, Targe-ype Daa J. Ferry Meron, Inc. Reon, VA, U.S.A. ferry@mec.com Abrac - The Exended Maxmum A Poeror Probably XMAP mehod for rack-o-rack aocaon baed on a formal,

More information

Machine Learning 2nd Edition

Machine Learning 2nd Edition INTRODUCTION TO Lecure Sldes for Machne Learnng nd Edon ETHEM ALPAYDIN, modfed by Leonardo Bobadlla and some pars from hp://www.cs.au.ac.l/~aparzn/machnelearnng/ The MIT Press, 00 alpaydn@boun.edu.r hp://www.cmpe.boun.edu.r/~ehem/mle

More information

CS626 Speech, Web and natural Language Processing End Sem

CS626 Speech, Web and natural Language Processing End Sem CS626 Speech, Web and naural Language Proceng End Sem Dae: 14/11/14 Tme: 9.30AM-12.30PM (no book, lecure noe allowed, bu ONLY wo page of any nformaon you deem f; clary and precon are very mporan; read

More information

Symmetry and Asymmetry of MIMO Fading Channels

Symmetry and Asymmetry of MIMO Fading Channels Symmery and Asymmery of MIMO Fadng Channels mmanuel Abbe, mre Telaar, Member, I, and Lzhong Zheng, Member, I Absrac We consder ergodc coheren MIMO channels, and characerze he opmal npu dsrbuon under general

More information

Example: MOSFET Amplifier Distortion

Example: MOSFET Amplifier Distortion 4/25/2011 Example MSFET Amplfer Dsoron 1/9 Example: MSFET Amplfer Dsoron Recall hs crcu from a prevous handou: ( ) = I ( ) D D d 15.0 V RD = 5K v ( ) = V v ( ) D o v( ) - K = 2 0.25 ma/v V = 2.0 V 40V.

More information

Learning Objectives. Self Organization Map. Hamming Distance(1/5) Introduction. Hamming Distance(3/5) Hamming Distance(2/5) 15/04/2015

Learning Objectives. Self Organization Map. Hamming Distance(1/5) Introduction. Hamming Distance(3/5) Hamming Distance(2/5) 15/04/2015 /4/ Learnng Objecves Self Organzaon Map Learnng whou Exaples. Inroducon. MAXNET 3. Cluserng 4. Feaure Map. Self-organzng Feaure Map 6. Concluson 38 Inroducon. Learnng whou exaples. Daa are npu o he syse

More information

Lecture 9: Dynamic Properties

Lecture 9: Dynamic Properties Shor Course on Molecular Dynamcs Smulaon Lecure 9: Dynamc Properes Professor A. Marn Purdue Unversy Hgh Level Course Oulne 1. MD Bascs. Poenal Energy Funcons 3. Inegraon Algorhms 4. Temperaure Conrol 5.

More information

5.2 GRAPHICAL VELOCITY ANALYSIS Polygon Method

5.2 GRAPHICAL VELOCITY ANALYSIS Polygon Method ME 352 GRHICL VELCITY NLYSIS 52 GRHICL VELCITY NLYSIS olygon Mehod Velociy analyi form he hear of kinemaic and dynamic of mechanical yem Velociy analyi i uually performed following a poiion analyi; ie,

More information

Thruster Modulation for Unsymmetric Flexible Spacecraft with Consideration of Torque Arm Perturbation

Thruster Modulation for Unsymmetric Flexible Spacecraft with Consideration of Torque Arm Perturbation hruer Modulaon for Unymmerc Flexble Sacecraf wh onderaon of orue rm Perurbaon a Shgemune anwak Shnchro chkawa a Yohak hkam b a Naonal Sace evelomen gency of Jaan 2-- Sengen ukuba-h barak b eo Unvery 3--

More information

y z P 3 P T P1 P 2. Werner Purgathofer. b a

y z P 3 P T P1 P 2. Werner Purgathofer. b a Einführung in Viual Compuing Einführung in Viual Compuing 86.822 in co T P 3 P co in T P P 2 co in Geomeric Tranformaion Geomeric Tranformaion W P h f Werner Purgahofer b a Tranformaion in he Rendering

More information

January Examinations 2012

January Examinations 2012 Page of 5 EC79 January Examnaons No. of Pages: 5 No. of Quesons: 8 Subjec ECONOMICS (POSTGRADUATE) Tle of Paper EC79 QUANTITATIVE METHODS FOR BUSINESS AND FINANCE Tme Allowed Two Hours ( hours) Insrucons

More information

CHAPTER 10: LINEAR DISCRIMINATION

CHAPTER 10: LINEAR DISCRIMINATION CHAPER : LINEAR DISCRIMINAION Dscrmnan-based Classfcaon 3 In classfcaon h K classes (C,C,, C k ) We defned dscrmnan funcon g j (), j=,,,k hen gven an es eample, e chose (predced) s class label as C f g

More information

Consider processes where state transitions are time independent, i.e., System of distinct states,

Consider processes where state transitions are time independent, i.e., System of distinct states, Dgal Speech Processng Lecure 0 he Hdden Marov Model (HMM) Lecure Oulne heory of Marov Models dscree Marov processes hdden Marov processes Soluons o he hree Basc Problems of HMM s compuaon of observaon

More information