A Novel Hierarchical Method for Digital Signal Type Classification

Size: px
Start display at page:

Download "A Novel Hierarchical Method for Digital Signal Type Classification"

Transcription

1 Proceedngs of the 6th WSEAS Internatona Conference on Apped Informatcs and Communcatons, Eounda, Greece, August 8-0, 006 (pp ) A Nove Herarchca Method for Dgta Sgna ype Cassfcaton AAOLLAH EBRAHIMZADEH, SEYED ALIREZA SEYEDIN Department of Eectrca and Computer Engneerng Nousrvan Insttute of echnoogy, Ferdows Unversty of Mashad IRAN Abstract: - Automatc sgnas type cassfcaton (ASC) s a technque for recognzng the moduaton scheme of an ntercepted sgna and has seen ncreasng demand n both mtary and cvan. Most of prevous technques can dentfy ony a few knds of sgna types. Aso, usuay, they request hgh SNR to acheve a mnmum acceptabe performance and don t ncude hgher order and new types of sgnas (e.g. QAM types).he work presented here proposes nove hgh performance method for dentfcaton of dgta sgna types. In ths method t s proposed a herarchca mutcass cassfer based on support vector machne (SVM). he nputs of ths cassfer are hgher order moments and cumuants (up to eght). Genetc agorthm (GA) s used to seect the best mode of SVMs n order to mprove the performance. Smuaton resuts show that proposed method can recognze a ot of dgta sgna types and acheves a hghy probabty of cassfcaton even at ow sgna to nose ratos (SNRs). Key-Words: - Pattern recognton, moduaton, machne earnng, support vector machne, mode seecton, eght order statstcs. Introducton Automatc sgnas type cassfcaton s a technque for recognzng the moduaton scheme of an ntercepted sgna. It pays an mportant roe for varous appcatons ad purposes. For exampe, n mtary appcaton, moduaton cassfcaton can be empoyed for eectronc surveance, nterference dentfcaton, montorng; n cv appcatons, t can be used for spectrum management, network traffc admnstraton, dfferent data rate aocaton, sgna confrmaton, nterference dentfcaton, software rados, mutdrop networks, ntegent modem, etc. In the past, sgna type recognton reed mosty on operators scannng the rado frequency spectrum wth a wde-band recever and checkng t vsuay on some sort of dspay. Ceary, these methods reed very much on the operators' sks and abtes. hese mtatons then ed to the deveopment of more automated moduaton recognzers. One semautomatc approach was to run the receved sgna through a number of demoduators and then have an operator determne the moduaton format by stenng to the output of each demoduator. hs approach s however not very practca anymore due to the new dgta technques that transfer both voce and data. hen technques for automatc sgna type cassfcaton (ASC) started to emerge. Whst eary researches ASC were concentrated on anaogue moduatons and ower orders of some types of dgta moduatons, such as frequency shft keyng (FSK), amptude shft keyng (ASK), the recent contrbutons n the subect focus more on hgher order and new types of dgta sgnas ke phase shft keyng (PSK), amptude shft keyng (QAM). Prmary, ths s due to ncreasng usage of such moduatons n many nove appcatons. ASC technques usuay can be categorzed n two man prncpes: the decson theoretc (D) and the pattern recognton (PR). D approaches use probabstc and hypothess testng arguments to formuate the recognton probem and to obtan the cassfcaton rue [-3]. he maor drawbacks of these approaches are ther very hgh computatona compexty, dffcuty wthn the mpementaton of decson rue, and ack of robustness to mode msmatch. Aso, these methods have dffcutes to set the correct threshod vaues. PR approaches, however, do not need such carefu treatment. hese methods are smpe to mpement.hey can be further dvded nto two subsystems: the feature extracton subsystem and the cassfer subsystem. he former extracts promnent characterstcs from the raw data, whch are caed features, and the cassfer dentfes the type of sgna based on the extracted features [4-]. It s showed that n moduaton dentfcaton area, ANNs especay mutayer perceptron (MLP) outperform other cassfers [8-]. However, wth regard to effectveness of ANNs, there are some probems worth mentonng. he tradtona ANNs have mtatons on generazaton gvng rse to modes that can overft to the tranng data. ANNs show

2 Proceedngs of the 6th WSEAS Internatona Conference on Apped Informatcs and Communcatons, Eounda, Greece, August 8-0, 006 (pp ) poor performances n ow SNRs and hgher number of moduaton schemes usuay requres a more compex neura network and hgher tranng tme [3, 4].hs defcences are due to the optmzaton agorthms used n ANNs for seecton of parameters and the statstca measures used to seect the mode. Recenty, support vector machnes (SVMs), based on statstca earnng theory are ganng appcatons n the area of machne earnng and pattern recognton because of hgh accuracy and exceent generazaton capabty. In [3, 4], the authors showed that usng SVM n content of sgna type dentfcaton acheves hghy success rate. hus n ths paper we use SVM as a cassfer. Seecton the proper mode of cassfer can mprove a substanta amount of s performance. In ths paper we have used the GA for SVM s mode seecton. Penty of ASC methods are abe to recognze ony a few knds of sgnas types and/or ower order of moduatons; aso, usuay, they request hgh SNR for havng a mnmum acceptabe performance. hs s due to facts: cassfer and features. Athough the cassfer has an mportant roe n cassfcaton, t shoud be mentoned that the features have vta roe. In ths work we have used hgher order moments and cumuants as features. he work presented here, proposes a hgh effcent ASC method, whch s abe to recognze dfferent type of receved sgnas. he paper organzed as foows. Secton presents feature extracton. Aso the consdered moduaton set s ntroduced n ths secton. Secton 3, descrbes the cassfer. Secton 4, ntroduced the GA that s used for mode seecton of SVMs. Secton 5, shows some smuaton studes. Fnay, n secton 6 concusons are presented. Feature extracton In moduaton dentfcaton probem, fndng the proper features s very mportant. For exampe QAM moduaton schemes contan nformaton n both amptude and phase (that are regarded as compex sgnas), thus fndng the proper feature that coud be abe to dentfy them (especay n case of hgher order and/or non-square) s dffcut. Based on our researches, the hgher order moments and hgher order cumuants up to eghth acheve the most hghy performances to dscrmnatng of dgta moduatons such as consdered set n ths paper. hese features have many advantages e.g. they provde a good way to descrbe the shape of the probabty densty functon. Probabty dstrbuton moments are a generazaton of concept of the expected vaue [3]. he auto-moment of the random varabe may be defned as foows: p q q M pq = E[ s ( s ) ] () where p caed moment order. Assume a zero-mean dscrete based-band sgna sequence of the form s k = ak + bk. Usng the (5), the expressons for dfferent order may be easy derved. th he symbosm for p order of cumuant s defned as: C pq = Cum[ 3 s,..., s, s 443,..., s ] () ( p q) terms ( q) terms here are reatons between moments and cumuants. he n th order cumuant s a functon of the moments of orders up to ncudng n. Moments s expressed n terms of cumuants as: M [ s ] { } { },.., sn = Cum s... um s v (3) v v where the summaton ndex s over a parttons v = v,..., v ) for the set of ndexes (,,..., n ), and q ( q s the number of eements n a gven partton. Cumuants may be aso be derved n terms of moments. he n order cumuant of a dscrete sgna th s(n) s gven by: q Cum [ s,.., s ] = ( ) ( )! [ ].. [ ] q E n s E s (4) v v v q In ths paper the consdered moduatons set s {ASK4, ASK8, PSK, PSK4, PSK8, Star-QAM8, V9, QAM6, QAM64, and QAM8}. We have computed the features of consdered dgta moduaton set. 3 Cassfer As sad we have used a herarchca SVM-based structure as cassfer. Support Vector Machne (SVM) s a supervsed machne earnng technque that can be apped as robust too for bnary and mutcass cassfcatons [5]. In case of SVMs, structura rsk mnmzaton (SRM) prncpe s used mnmzng an upper bound on the expected rsk whereas n ANNs, emprca rsk mnmzaton (ERM) s used mnmzng the error on the tranng data. he dfference n RM eads to better generazaton performance for SVMs than ANNs. he SVM performs cassfcaton tasks by constructng optma separatng hyperpanes (OSH). he dea of bnary SVM can be expressed as foows [6]. 3. Bnary SVM Suppose the tranng set, d ( x, y ), =,,...,, x R, y {, + } can be separated by the hyperpane w x + b = 0, where w r s

3 Proceedngs of the 6th WSEAS Internatona Conference on Apped Informatcs and Communcatons, Eounda, Greece, August 8-0, 006 (pp ) weght vector and b s bas. If ths hyperpane maxmze the margn, then the foowng nequaty s vad for a nput data: y ( w x,for a x =,,..., (5) he margn of the hyper-pane s / w r, thus, the probem s: maxmzng the margn by mnmzng of w subect to constrants (5), that s a convex quadratc programmng (QP) probem. hs probem has a goba optmum, and Lagrange mutpers are used to sove t: LP = w α [ y ( w x ] (6) = where a, =,..., are the Lagrange mutpers ( α 0).he souton to ths QP probem s gven by mnmzng L P wth respect to w and b. After dfferentatng L P wth respect to w and b and settng the dervatves equa to 0, yeds: w = α x (7) = y It can obtan the dua varabes Lagrangan by mposng the Karush-Kuhn-ucker (KK) condtons: L d = = α w = α αα y y x x = = = o fnd the OSH, t must maxmze (8) L d under the constrants of α = 0,andα 0. Note that the = y Lagrange mutpers are ony non-zero ( α f 0 ) when y ( w x =. hose tranng ponts for whch the equaty n (5) hods are caed support vectors (SV) that can satsfyα f 0. he optma weghts are gven by (7) and the bas s gven by: b = y w x (9) for any support vector x. he optma decson functon (ODF) s then gven by: f ( x) = sgn( y α x x + b ) (0) = where α s are optma Lagrange mutpers. For nput data wth a hgh nose eve, SVM uses soft margns can be expressed as foows wth the ntroducton of the non-negatve sack varabesξ, =,..., : y ( w x ξ for =,,..., () o obtan the OSH, t shoud be mnmzng the k Φ = w + C ξ subect to constrants (), = where C s the penaty parameter, whch contros the tradeoff between the compexty of the decson functon and the number of tranng exampes mscassfed,.e., contros the tradeoff between margn maxmzaton and error mnmzaton. hs probem can be soved wth smar method. In the nonneary separabe cases, the SVM map the tranng ponts, nonneary, to a hghdmensona feature space usng kerne functon K( x r r, x ), where near separaton may be possbe. he most famous kerne functons are near, poynoma, rada bass functon (RBF), and sgmod. For exampe RBF w be shown by: K( x, y) = exp( x y / σ ) () where σ s the wdth of the RBF kerne. After a kerne functon s seected, the QP probem s: Ld = α αα y y K( x, x ) (3) the = = = α s derved by: α = arg max L 0 α C; =,,..., ; α d = α y = 0 (4) After tranng, the foowng, the decson functon, becomes: f ( x) = sgn( y α K( x, x ) + b ) (5) = he performance of SVM can be controed through a few free parameters ke the term C and the kerne parameter whch caed are hyperparameters. 3. Mutcass SVM As sad SVM was orgnay deveoped for bnary cassfcaton probems, but t can be used for mutcass cassfcaton. here are severa approaches avaabe to extend bnary SVMs to mut-cass probems that fa nto two categores: the oneaganst-others method and one-aganst-one method [7]. In ths research, we have proposed a nove, smpe and effectve souton for mutcass SVMbased that s herarchca. In ths scheme at frst a of nputs fed to one SVM. hs SVM separates them n two groups. Each of these groups fed to another SVMs and process w be contnued t a of patterns are cassfed. he detas of ths structure ntroduced n secton 5.

4 Proceedngs of the 6th WSEAS Internatona Conference on Apped Informatcs and Communcatons, Eounda, Greece, August 8-0, 006 (pp ) 4 Mode seecton and GA Drvng the optma vaues of SVMs parameters are mportant to acheve hgh generazaton abty and performance. GAs wth ther characterstcs of hgh effcency and goba optmzaton are wdey apped n many areas. In ths paper, accordng the determned approprate ftness functon for GA operaton, an effectve strategy for parameters seecton for SVM s proposed by usng the GA. GA s a stochastc optmzaton agorthm whch adopts Darwn s theory of survva of the fttest. o appy GAs to the SVM parameters seecton probem, one has to consder the foowng ssues: the encodng scheme, the methodoogy to produce the nta popuaton, the ftness functon and the genetc operators such as reproducton, crossover and mutaton. 4. Encodng Scheme Here rea-encoded scheme s seected as the representaton of the parameters n ths paper. he research space of these parameters s C [5 : 5 : 50], σ [0.: 0.: 3.0] 4. Produce the Inta Popuaton Because the rea-coded scheme s used, the souton space concdes wth the chromosome space. Consderng the bgger popuaton w enarge the GA runnng tme and dsperse the conformaton of the popuaton, the sze of popuaton, pop_sze s choose 6 n orderng to avodng the convergence of the popuaton becomes dffcut. For producng the nta popuaton, the nta vaues of the desgned parameters are dstrbuted n the souton space as even as possbe. 4.3 Ftness Functon Accordng to the aforementoned anayss, the average performance of the SVM cassfer s depended on E { R / γ ) and not smpy on the arge margn γ. he Radus-margn bound [8] s proposed as the ftness functon R = (6) γ where γ denotes the margn, s the sze of the tranng sampes, R s the radus of the smaest sphere contanng the tranng data, R = Genetc Operators Genetc operator ncudes the foowng three basc operators such as seecton, crossover and mutaton. Here, a heurstc search strategy s adopted to reaze the genetc optmzaton for automatc parameters seecton. he adopted GA operators are brefy presented as foows Seecton operator Seecton operators s composed of a copy seecton operaton and a survve seecton operaton. Here the method of survva of the fttest was used to seect the next generaton ndvdua. Gven the ftness functon ft ( a ) of the ndvdua a, the probabty of a seected as the next generaton one s as foow: ft( a ) P( a ) = pop _ sze (7) pop _ sze ft( a ) = 4.4. Crossover operator he means of crossover mpement s cosey ntegrated to the encodng scheme. Due to the reaencodng scheme s utzed; the crossover operator n ths paper can be defned as [9]: P = ap + ( a) P (8) where P' s the offsprng after crossover operaton,p and P are two parents to be mpemented the crossover operaton, and a s a constant whch beongs to (0,).Here a = Mutaton operator How the bgger vaue of the mutaton operator s chose to mantan the dversty of the popuaton n the eary GA operaton and avod the precocty? he adaptve mutaton probabty s adopted n ths paper to sove the above two probems as foows: exp( b t / ) P m = (9) pop _ sze L where t s the generaton of the genetc teraton, pop _ sze s the sze of the popuaton, L s the ength of the ndvdua, b=.5 s a preset parameter he stoppng crtera In genera GA agorthm, termnate the program when the best ftness has not changed more than a very sma vaue,.e. 0-6 over the ast generatons. But we choose the average ftness rather than best ftness as the stoppng crtera. 5 Smuaton study hs secton present some evauaton resuts of proposed dentfer for moduaton set that s defned n secton. For smpfyng the ndcaton, we substtute the moduatons ASK4, ASK8, PSK, PSK4, PSK8, Star-QAM8, V9, QAM6, QAM64,

5 Proceedngs of the 6th WSEAS Internatona Conference on Apped Informatcs and Communcatons, Eounda, Greece, August 8-0, 006 (pp ) QAM8 wth P, P, P 3, P 4, P 5, P 6, P 7, P 8, P 9, P 0 respectvey. A sgna are dgtay moduated n MALAB smuator. he smuated sgnas were aso band-mted and Gaussan nose was added accordng to SNRs, 3, 0, 3, 6, 9,, and 8 db. Each moduaton type has 00 reazatons of 048 sampes. Fgure shows the herarcha SVM-based cassfer. abe shows the chosen features for each SVM that dscrmnate sgna types wth them. At the frst, we evauate the performance of system wthout GA. Based on some smuatons, the vaue σ = s seected for a SVMs. he effect of nose on cassfer performance s aso studed wth dfferent SNRs. Random ony 0% of data s used for tranng. Overa performance (OP) s shown n abe. As we see, the performance s generay good even wth ow SNR vaues. Now, we appy the GA or mode seecton of SVMs. We have used the GA method for each of SVM separatey. One of the advantages of herarchca mutcass-svm based n comparson others mut-cass SVM-based methods (.e. one-agansta and one-aganst-one) s that optmzaton of each SVM coud be done separatey. In the other methods even though each SVM s tuned very we for the bnary probem, there s no guarantee that they work we together for the entre probem and the parameters of the kernes affect the structure of the feature space and the cassfcaton accuracy. abe 3, shows the overa performance of proposed dentfer. As we see, mode seecton exacty mproves the performances of system n a SNRs; especay n ower SNRs. SVM P 3 SVM SVM4 SVM6 SVM3 SVM5 P 6 P P P 4 P 5 abe 3: OP wth appyng of GA for mode seecton at dfferent SNRs. SNR (db) OP (%) SVM7 P 0 P 7 SVM8 SVM9 Fgure. Herarchca SVM-Based Cassfer abe : Chosen features for each SVM P 8 P 9 SVM s Number Chosen features SVM C 83 SVM M 4 SVM 3 M 4 SVM 4 C 80 SVM 5 C 80, M 6 SVM 6 C 63 SVM 7 C 8 SVM 8 C 63 SVM 9 C 80, M 84 abe : OP wthout SVMs mode seecton at dfferent SNRs. SNR (db) OP (%) For comparson our method wth other methods as sad there are a few researches on moduaton dentfcaton that consder the number of moduatons about ten or a tte and most of methods that have been proposed have mtaton to consder addtona moduaton. hs s many because of ther features and/or cassfers. In [0], the author reported a generazaton rate of 90% and 93% of accuracy of data sets wth SNR of 5 5 db for consdered moduaton set. However, the performances for ower SNRs are reported to be ess than 80% for a fuy connected network, and about 85% for a herarchca network. In [], the authors deveoped and compared one decson tree and one neura network cassfer. Haf of the smuated sgnas used for tranng had a SNR vaue of 5 db and the other haf 0 db. It used nstantaneous features n addton spectra features. he neura network cassfer conssted of three MLPs. he resuts showed 88% success rate at 5 db SNR for consdered moduatons set. In [], the researches show the average dentfcaton rate s 83%, and t reaches over 90% for SNR vaue of over 0 db. However, f SNR s ess than 0 db, the performance drops to ess than 70%. he dentfer proposed n ths work shows a steady performance wth dfferent SNR vaues and has a hgh performance at very ow SNRs. Usng an opt-

6 Proceedngs of the 6th WSEAS Internatona Conference on Apped Informatcs and Communcatons, Eounda, Greece, August 8-0, 006 (pp ) mzaton method, the performance of dentfer ncrease. he structure of proposed method s very smpe. he proposed method has a hgh generazaton abty and accuracy. 6 Concusons Automatc sgnas type cassfcaton (ASC) has seen ncreasng demand n both mtary and cvan. Most of prevous technques can dentfy ony a few knds of sgna types and usuay need hgh SNR to acheve a mnmum acceptabe performance and don t ncude hgher order and new types of sgnas. he work presented here proposes nove hgh performance method for dentfcaton of dgta sgna types. In ths method we have used a herarchca mutcass cassfer based on support vector machne (SVM). he nputs of ths cassfer are hgher order moments and cumuants (up to eght). he SVM cassfer, use the feature vector and maps the nput vectors non-nearty nto hgh dmensona feature space and constructs the optmum separatng hyper-pane n the space to reaze sgna recognton. hs method s robust and avods over-fttng and oca mnmum. Optmzaton usng GA, mproves the performance of system especay n ower SNRs. References: [] C. Le Martret, and D. Botea, A genera max mum kehood cassfer for moduaton cassfcaton, Proc. ICASSP, Vo. 4, 998, pp [] J. A. Ss, Maxmum-kehood moduaton cassfcaton for PSK/QAM, Proc. MILCOM, 999, pp [3] P. Panagotou, and A. Poydoros, Lkehood rato tests for moduaton cassfcatons, Proc. MILCOM, 000, pp [4] K.C. Ho, W. Prokopw and Y.. Chan, Moduaton dentfcaton of dgta sgnas by waveet transform, Proc. IEE,Radar, Sonar Navg., Vo.47, No.4, 000,pp [5] B. G. Mobas, Dgta moduaton cassfcaton usng consteaton shape, Sgna Processng, Vo. 80, 000, pp [6] J. Lopatka, and P. Macre, Automatc moduaton cassfcaton usng statstca moments and a fuzzy cassfer, Proc. ICSP, 000, pp.- 7. [7] A. Swam, and B. M. Sader, Herarchca dgta moduaton cassfcaton usng cumuants, IEEE rans. Comm., Vo. 48, No. 3, 000, pp [8] S. A. Seyedn, and, A. Ebrahmzadeh, Automatc sgna type dentfcaton by neurawaveet based method, Proc. SICED, 005, pp [9] A. Ebrahmzadeh, S. A. Seyedn, A new method for dentfyng of sgna type, Proc. CIS, 005, [0] C. L. P. Seher, Automatc moduaton recognton wth a herarchca neura network, Proc. MILCOM, 993, pp. 5. [] A.K. Nand, E.E. Azzouz, Agorthms for automatc moduaton recognton of communcaton sgnas, IEEE rans. Commun., Vo. 46, No. 4, 998, pp [] L. Mngquan, X. Xanc, L. Lemng, Cycc spectra features based moduaton recognton, Proc. Comm. ech., Vo., 998, pp [3] A. Ebrahmzadeh, and S. A. Seyedn, Identfcaton of sgna type usng svm n fadng envronment, Proc. ICEE, 006. [4]A. Ebrahmzadeh, and S. A. Seyedn, An Integent echnque for Identfcaton of Dgta Moduatons, accepted to pubsh n WSEAS rans Comp., 006. [5] C. Cortes, and V. Vapnc, Support Vector Network, Machne Learnng, Vo. 0, 995, pp.- 5. [6] C. Burges A utora on Support Vector Machnes for Pattern Recognton, Data Mnng and Knowedge Dscovery, Vo., 998, pp -67. [7] B. SchOkopf, C. Burges, and V. Vapnk, Extractng support data for a gven task, ICKDDM,995, pp [8] O. Chapee, V. Vapnk, O. Bousquet, and S. Mukheree, Choosng mutpe parameters for support vector machnes, Machne Learnng, 00, Vo. 46, No., pp [9] Y. X. Su, B. Y. Duan, C. H. Zheng, Genetc desgn of knematcay optma fne tunng Stewart patform for arge spherca rado teescope, Mechatroncs, 00, Vo., No.7, pp

Example: Suppose we want to build a classifier that recognizes WebPages of graduate students.

Example: Suppose we want to build a classifier that recognizes WebPages of graduate students. Exampe: Suppose we want to bud a cassfer that recognzes WebPages of graduate students. How can we fnd tranng data? We can browse the web and coect a sampe of WebPages of graduate students of varous unverstes.

More information

Automatic Digital Modulation Identification in Dispersive Channels

Automatic Digital Modulation Identification in Dispersive Channels roceedngs of e 5 WSEAS Internatona Conference on eecommuncatons and Informatc Istanbu, urey, ay 7-9, 006 (pp409-44) Automatc Dgta oduaton Identfcaton n Dspersve Channes AAOLLAH EBRAHIZADEH, SEYED ALIREZA

More information

NONLINEAR SYSTEM IDENTIFICATION BASE ON FW-LSSVM

NONLINEAR SYSTEM IDENTIFICATION BASE ON FW-LSSVM Journa of heoretca and Apped Informaton echnoogy th February 3. Vo. 48 No. 5-3 JAI & LLS. A rghts reserved. ISSN: 99-8645 www.jatt.org E-ISSN: 87-395 NONLINEAR SYSEM IDENIFICAION BASE ON FW-LSSVM, XIANFANG

More information

Research on Complex Networks Control Based on Fuzzy Integral Sliding Theory

Research on Complex Networks Control Based on Fuzzy Integral Sliding Theory Advanced Scence and Technoogy Letters Vo.83 (ISA 205), pp.60-65 http://dx.do.org/0.4257/ast.205.83.2 Research on Compex etworks Contro Based on Fuzzy Integra Sdng Theory Dongsheng Yang, Bngqng L, 2, He

More information

Optimum Selection Combining for M-QAM on Fading Channels

Optimum Selection Combining for M-QAM on Fading Channels Optmum Seecton Combnng for M-QAM on Fadng Channes M. Surendra Raju, Ramesh Annavajjaa and A. Chockangam Insca Semconductors Inda Pvt. Ltd, Bangaore-56000, Inda Department of ECE, Unversty of Caforna, San

More information

Image Classification Using EM And JE algorithms

Image Classification Using EM And JE algorithms Machne earnng project report Fa, 2 Xaojn Sh, jennfer@soe Image Cassfcaton Usng EM And JE agorthms Xaojn Sh Department of Computer Engneerng, Unversty of Caforna, Santa Cruz, CA, 9564 jennfer@soe.ucsc.edu

More information

Sparse Training Procedure for Kernel Neuron *

Sparse Training Procedure for Kernel Neuron * Sparse ranng Procedure for Kerne Neuron * Janhua XU, Xuegong ZHANG and Yanda LI Schoo of Mathematca and Computer Scence, Nanng Norma Unversty, Nanng 0097, Jangsu Provnce, Chna xuanhua@ema.nnu.edu.cn Department

More information

MARKOV CHAIN AND HIDDEN MARKOV MODEL

MARKOV CHAIN AND HIDDEN MARKOV MODEL MARKOV CHAIN AND HIDDEN MARKOV MODEL JIAN ZHANG JIANZHAN@STAT.PURDUE.EDU Markov chan and hdden Markov mode are probaby the smpest modes whch can be used to mode sequenta data,.e. data sampes whch are not

More information

Neural network-based athletics performance prediction optimization model applied research

Neural network-based athletics performance prediction optimization model applied research Avaabe onne www.jocpr.com Journa of Chemca and Pharmaceutca Research, 04, 6(6):8-5 Research Artce ISSN : 0975-784 CODEN(USA) : JCPRC5 Neura networ-based athetcs performance predcton optmzaton mode apped

More information

Nested case-control and case-cohort studies

Nested case-control and case-cohort studies Outne: Nested case-contro and case-cohort studes Ørnuf Borgan Department of Mathematcs Unversty of Oso NORBIS course Unversty of Oso 4-8 December 217 1 Radaton and breast cancer data Nested case contro

More information

IDENTIFICATION OF NONLINEAR SYSTEM VIA SVR OPTIMIZED BY PARTICLE SWARM ALGORITHM

IDENTIFICATION OF NONLINEAR SYSTEM VIA SVR OPTIMIZED BY PARTICLE SWARM ALGORITHM Journa of Theoretca and Apped Informaton Technoogy th February 3. Vo. 48 No. 5-3 JATIT & LLS. A rghts reserved. ISSN: 99-8645 www.att.org E-ISSN: 87-395 IDENTIFICATION OF NONLINEAR SYSTEM VIA SVR OPTIMIZED

More information

Support Vector Machines. Vibhav Gogate The University of Texas at dallas

Support Vector Machines. Vibhav Gogate The University of Texas at dallas Support Vector Machnes Vbhav Gogate he Unversty of exas at dallas What We have Learned So Far? 1. Decson rees. Naïve Bayes 3. Lnear Regresson 4. Logstc Regresson 5. Perceptron 6. Neural networks 7. K-Nearest

More information

The University of Auckland, School of Engineering SCHOOL OF ENGINEERING REPORT 616 SUPPORT VECTOR MACHINES BASICS. written by.

The University of Auckland, School of Engineering SCHOOL OF ENGINEERING REPORT 616 SUPPORT VECTOR MACHINES BASICS. written by. The Unversty of Auckand, Schoo of Engneerng SCHOOL OF ENGINEERING REPORT 66 SUPPORT VECTOR MACHINES BASICS wrtten by Vojsav Kecman Schoo of Engneerng The Unversty of Auckand Apr, 004 Vojsav Kecman Copyrght,

More information

Short-Term Load Forecasting for Electric Power Systems Using the PSO-SVR and FCM Clustering Techniques

Short-Term Load Forecasting for Electric Power Systems Using the PSO-SVR and FCM Clustering Techniques Energes 20, 4, 73-84; do:0.3390/en40073 Artce OPEN ACCESS energes ISSN 996-073 www.mdp.com/journa/energes Short-Term Load Forecastng for Eectrc Power Systems Usng the PSO-SVR and FCM Custerng Technques

More information

Application of support vector machine in health monitoring of plate structures

Application of support vector machine in health monitoring of plate structures Appcaton of support vector machne n heath montorng of pate structures *Satsh Satpa 1), Yogesh Khandare ), Sauvk Banerjee 3) and Anrban Guha 4) 1), ), 4) Department of Mechanca Engneerng, Indan Insttute

More information

Associative Memories

Associative Memories Assocatve Memores We consder now modes for unsupervsed earnng probems, caed auto-assocaton probems. Assocaton s the task of mappng patterns to patterns. In an assocatve memory the stmuus of an ncompete

More information

WAVELET-BASED IMAGE COMPRESSION USING SUPPORT VECTOR MACHINE LEARNING AND ENCODING TECHNIQUES

WAVELET-BASED IMAGE COMPRESSION USING SUPPORT VECTOR MACHINE LEARNING AND ENCODING TECHNIQUES WAVELE-BASED IMAGE COMPRESSION USING SUPPOR VECOR MACHINE LEARNING AND ENCODING ECHNIQUES Rakb Ahmed Gppsand Schoo of Computng and Informaton echnoogy Monash Unversty, Gppsand Campus Austraa. Rakb.Ahmed@nfotech.monash.edu.au

More information

Multispectral Remote Sensing Image Classification Algorithm Based on Rough Set Theory

Multispectral Remote Sensing Image Classification Algorithm Based on Rough Set Theory Proceedngs of the 2009 IEEE Internatona Conference on Systems Man and Cybernetcs San Antono TX USA - October 2009 Mutspectra Remote Sensng Image Cassfcaton Agorthm Based on Rough Set Theory Yng Wang Xaoyun

More information

On the Equality of Kernel AdaTron and Sequential Minimal Optimization in Classification and Regression Tasks and Alike Algorithms for Kernel

On the Equality of Kernel AdaTron and Sequential Minimal Optimization in Classification and Regression Tasks and Alike Algorithms for Kernel Proceedngs of th European Symposum on Artfca Neura Networks, pp. 25-222, ESANN 2003, Bruges, Begum, 2003 On the Equaty of Kerne AdaTron and Sequenta Mnma Optmzaton n Cassfcaton and Regresson Tasks and

More information

A General Column Generation Algorithm Applied to System Reliability Optimization Problems

A General Column Generation Algorithm Applied to System Reliability Optimization Problems A Genera Coumn Generaton Agorthm Apped to System Reabty Optmzaton Probems Lea Za, Davd W. Cot, Department of Industra and Systems Engneerng, Rutgers Unversty, Pscataway, J 08854, USA Abstract A genera

More information

COXREG. Estimation (1)

COXREG. Estimation (1) COXREG Cox (972) frst suggested the modes n whch factors reated to fetme have a mutpcatve effect on the hazard functon. These modes are caed proportona hazards (PH) modes. Under the proportona hazards

More information

Optimization of JK Flip Flop Layout with Minimal Average Power of Consumption based on ACOR, Fuzzy-ACOR, GA, and Fuzzy-GA

Optimization of JK Flip Flop Layout with Minimal Average Power of Consumption based on ACOR, Fuzzy-ACOR, GA, and Fuzzy-GA Journa of mathematcs and computer Scence 4 (05) - 5 Optmzaton of JK Fp Fop Layout wth Mnma Average Power of Consumpton based on ACOR, Fuzzy-ACOR, GA, and Fuzzy-GA Farshd Kevanan *,, A Yekta *,, Nasser

More information

Application of Particle Swarm Optimization to Economic Dispatch Problem: Advantages and Disadvantages

Application of Particle Swarm Optimization to Economic Dispatch Problem: Advantages and Disadvantages Appcaton of Partce Swarm Optmzaton to Economc Dspatch Probem: Advantages and Dsadvantages Kwang Y. Lee, Feow, IEEE, and Jong-Bae Par, Member, IEEE Abstract--Ths paper summarzes the state-of-art partce

More information

Design and Optimization of Fuzzy Controller for Inverse Pendulum System Using Genetic Algorithm

Design and Optimization of Fuzzy Controller for Inverse Pendulum System Using Genetic Algorithm Desgn and Optmzaton of Fuzzy Controller for Inverse Pendulum System Usng Genetc Algorthm H. Mehraban A. Ashoor Unversty of Tehran Unversty of Tehran h.mehraban@ece.ut.ac.r a.ashoor@ece.ut.ac.r Abstract:

More information

On the Power Function of the Likelihood Ratio Test for MANOVA

On the Power Function of the Likelihood Ratio Test for MANOVA Journa of Mutvarate Anayss 8, 416 41 (00) do:10.1006/jmva.001.036 On the Power Functon of the Lkehood Rato Test for MANOVA Dua Kumar Bhaumk Unversty of South Aabama and Unversty of Inos at Chcago and Sanat

More information

Kernel Methods and SVMs Extension

Kernel Methods and SVMs Extension Kernel Methods and SVMs Extenson The purpose of ths document s to revew materal covered n Machne Learnng 1 Supervsed Learnng regardng support vector machnes (SVMs). Ths document also provdes a general

More information

The Application of BP Neural Network principal component analysis in the Forecasting the Road Traffic Accident

The Application of BP Neural Network principal component analysis in the Forecasting the Road Traffic Accident ICTCT Extra Workshop, Bejng Proceedngs The Appcaton of BP Neura Network prncpa component anayss n Forecastng Road Traffc Accdent He Mng, GuoXucheng &LuGuangmng Transportaton Coege of Souast Unversty 07

More information

Downlink Power Allocation for CoMP-NOMA in Multi-Cell Networks

Downlink Power Allocation for CoMP-NOMA in Multi-Cell Networks Downn Power Aocaton for CoMP-NOMA n Mut-Ce Networs Md Shpon A, Eram Hossan, Arafat A-Dwe, and Dong In Km arxv:80.0498v [eess.sp] 6 Dec 207 Abstract Ths wor consders the probem of dynamc power aocaton n

More information

Predicting Model of Traffic Volume Based on Grey-Markov

Predicting Model of Traffic Volume Based on Grey-Markov Vo. No. Modern Apped Scence Predctng Mode of Traffc Voume Based on Grey-Marov Ynpeng Zhang Zhengzhou Muncpa Engneerng Desgn & Research Insttute Zhengzhou 5005 Chna Abstract Grey-marov forecastng mode of

More information

An Improved multiple fractal algorithm

An Improved multiple fractal algorithm Advanced Scence and Technology Letters Vol.31 (MulGraB 213), pp.184-188 http://dx.do.org/1.1427/astl.213.31.41 An Improved multple fractal algorthm Yun Ln, Xaochu Xu, Jnfeng Pang College of Informaton

More information

Polite Water-filling for Weighted Sum-rate Maximization in MIMO B-MAC Networks under. Multiple Linear Constraints

Polite Water-filling for Weighted Sum-rate Maximization in MIMO B-MAC Networks under. Multiple Linear Constraints 2011 IEEE Internatona Symposum on Informaton Theory Proceedngs Pote Water-fng for Weghted Sum-rate Maxmzaton n MIMO B-MAC Networks under Mutpe near Constrants An u 1, Youjan u 2, Vncent K. N. au 3, Hage

More information

9 Adaptive Soft K-Nearest-Neighbour Classifiers with Large Margin

9 Adaptive Soft K-Nearest-Neighbour Classifiers with Large Margin 9 Adaptve Soft -Nearest-Neghbour Cassfers wth Large argn Abstract- A nove cassfer s ntroduced to overcome the mtatons of the -NN cassfcaton systems. It estmates the posteror cass probabtes usng a oca Parzen

More information

A new P system with hybrid MDE- k -means algorithm for data. clustering. 1 Introduction

A new P system with hybrid MDE- k -means algorithm for data. clustering. 1 Introduction Wesun, Lasheng Xang, Xyu Lu A new P system wth hybrd MDE- agorthm for data custerng WEISUN, LAISHENG XIANG, XIYU LIU Schoo of Management Scence and Engneerng Shandong Norma Unversty Jnan, Shandong CHINA

More information

Linear Classification, SVMs and Nearest Neighbors

Linear Classification, SVMs and Nearest Neighbors 1 CSE 473 Lecture 25 (Chapter 18) Lnear Classfcaton, SVMs and Nearest Neghbors CSE AI faculty + Chrs Bshop, Dan Klen, Stuart Russell, Andrew Moore Motvaton: Face Detecton How do we buld a classfer to dstngush

More information

Chapter 6 Hidden Markov Models. Chaochun Wei Spring 2018

Chapter 6 Hidden Markov Models. Chaochun Wei Spring 2018 896 920 987 2006 Chapter 6 Hdden Markov Modes Chaochun We Sprng 208 Contents Readng materas Introducton to Hdden Markov Mode Markov chans Hdden Markov Modes Parameter estmaton for HMMs 2 Readng Rabner,

More information

Adaptive and Iterative Least Squares Support Vector Regression Based on Quadratic Renyi Entropy

Adaptive and Iterative Least Squares Support Vector Regression Based on Quadratic Renyi Entropy daptve and Iteratve Least Squares Support Vector Regresson Based on Quadratc Ren Entrop Jngqng Jang, Chu Song, Haan Zhao, Chunguo u,3 and Yanchun Lang Coege of Mathematcs and Computer Scence, Inner Mongoa

More information

The Entire Solution Path for Support Vector Machine in Positive and Unlabeled Classification 1

The Entire Solution Path for Support Vector Machine in Positive and Unlabeled Classification 1 Abstract The Entre Souton Path for Support Vector Machne n Postve and Unabeed Cassfcaton 1 Yao Lmn, Tang Je, and L Juanz Department of Computer Scence, Tsnghua Unversty 1-308, FIT, Tsnghua Unversty, Bejng,

More information

2E Pattern Recognition Solutions to Introduction to Pattern Recognition, Chapter 2: Bayesian pattern classification

2E Pattern Recognition Solutions to Introduction to Pattern Recognition, Chapter 2: Bayesian pattern classification E395 - Pattern Recognton Solutons to Introducton to Pattern Recognton, Chapter : Bayesan pattern classfcaton Preface Ths document s a soluton manual for selected exercses from Introducton to Pattern Recognton

More information

The Study of Teaching-learning-based Optimization Algorithm

The Study of Teaching-learning-based Optimization Algorithm Advanced Scence and Technology Letters Vol. (AST 06), pp.05- http://dx.do.org/0.57/astl.06. The Study of Teachng-learnng-based Optmzaton Algorthm u Sun, Yan fu, Lele Kong, Haolang Q,, Helongang Insttute

More information

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur Module 3 LOSSY IMAGE COMPRESSION SYSTEMS Verson ECE IIT, Kharagpur Lesson 6 Theory of Quantzaton Verson ECE IIT, Kharagpur Instructonal Objectves At the end of ths lesson, the students should be able to:

More information

Demodulation of PPM signal based on sequential Monte Carlo model

Demodulation of PPM signal based on sequential Monte Carlo model Internatona Journa of Computer Scence and Eectroncs Engneerng (IJCSEE) Voume 1, Issue 1 (213) ISSN 232 428 (Onne) Demoduaton of M sgna based on seuenta Monte Caro mode Lun Huang and G. E. Atkn Abstract

More information

A DIMENSION-REDUCTION METHOD FOR STOCHASTIC ANALYSIS SECOND-MOMENT ANALYSIS

A DIMENSION-REDUCTION METHOD FOR STOCHASTIC ANALYSIS SECOND-MOMENT ANALYSIS A DIMESIO-REDUCTIO METHOD FOR STOCHASTIC AALYSIS SECOD-MOMET AALYSIS S. Rahman Department of Mechanca Engneerng and Center for Computer-Aded Desgn The Unversty of Iowa Iowa Cty, IA 52245 June 2003 OUTLIE

More information

REAL-TIME IMPACT FORCE IDENTIFICATION OF CFRP LAMINATED PLATES USING SOUND WAVES

REAL-TIME IMPACT FORCE IDENTIFICATION OF CFRP LAMINATED PLATES USING SOUND WAVES 8 TH INTERNATIONAL CONERENCE ON COMPOSITE MATERIALS REAL-TIME IMPACT ORCE IDENTIICATION O CRP LAMINATED PLATES USING SOUND WAVES S. Atobe *, H. Kobayash, N. Hu 3 and H. ukunaga Department of Aerospace

More information

Supplementary Material: Learning Structured Weight Uncertainty in Bayesian Neural Networks

Supplementary Material: Learning Structured Weight Uncertainty in Bayesian Neural Networks Shengyang Sun, Changyou Chen, Lawrence Carn Suppementary Matera: Learnng Structured Weght Uncertanty n Bayesan Neura Networks Shengyang Sun Changyou Chen Lawrence Carn Tsnghua Unversty Duke Unversty Duke

More information

CS 3710: Visual Recognition Classification and Detection. Adriana Kovashka Department of Computer Science January 13, 2015

CS 3710: Visual Recognition Classification and Detection. Adriana Kovashka Department of Computer Science January 13, 2015 CS 3710: Vsual Recognton Classfcaton and Detecton Adrana Kovashka Department of Computer Scence January 13, 2015 Plan for Today Vsual recognton bascs part 2: Classfcaton and detecton Adrana s research

More information

Interference Alignment and Degrees of Freedom Region of Cellular Sigma Channel

Interference Alignment and Degrees of Freedom Region of Cellular Sigma Channel 2011 IEEE Internatona Symposum on Informaton Theory Proceedngs Interference Agnment and Degrees of Freedom Regon of Ceuar Sgma Channe Huaru Yn 1 Le Ke 2 Zhengdao Wang 2 1 WINLAB Dept of EEIS Unv. of Sc.

More information

Greyworld White Balancing with Low Computation Cost for On- Board Video Capturing

Greyworld White Balancing with Low Computation Cost for On- Board Video Capturing reyword Whte aancng wth Low Computaton Cost for On- oard Vdeo Capturng Peng Wu Yuxn Zoe) Lu Hewett-Packard Laboratores Hewett-Packard Co. Pao Ato CA 94304 USA Abstract Whte baancng s a process commony

More information

3. Stress-strain relationships of a composite layer

3. Stress-strain relationships of a composite layer OM PO I O U P U N I V I Y O F W N ompostes ourse 8-9 Unversty of wente ng. &ech... tress-stran reatonshps of a composte ayer - Laurent Warnet & emo Aerman.. tress-stran reatonshps of a composte ayer Introducton

More information

Support Vector Machine Technique for Wind Speed Prediction

Support Vector Machine Technique for Wind Speed Prediction Internatona Proceedngs of Chemca, Boogca and Envronmenta Engneerng, Vo. 93 (016) DOI: 10.7763/IPCBEE. 016. V93. Support Vector Machne Technque for Wnd Speed Predcton Yusuf S. Turkan 1 and Hacer Yumurtacı

More information

Natural Language Processing and Information Retrieval

Natural Language Processing and Information Retrieval Natural Language Processng and Informaton Retreval Support Vector Machnes Alessandro Moschtt Department of nformaton and communcaton technology Unversty of Trento Emal: moschtt@ds.untn.t Summary Support

More information

A finite difference method for heat equation in the unbounded domain

A finite difference method for heat equation in the unbounded domain Internatona Conerence on Advanced ectronc Scence and Technoogy (AST 6) A nte derence method or heat equaton n the unbounded doman a Quan Zheng and Xn Zhao Coege o Scence North Chna nversty o Technoogy

More information

[WAVES] 1. Waves and wave forces. Definition of waves

[WAVES] 1. Waves and wave forces. Definition of waves 1. Waves and forces Defnton of s In the smuatons on ong-crested s are consdered. The drecton of these s (μ) s defned as sketched beow n the goba co-ordnate sstem: North West East South The eevaton can

More information

Lower Bounding Procedures for the Single Allocation Hub Location Problem

Lower Bounding Procedures for the Single Allocation Hub Location Problem Lower Boundng Procedures for the Snge Aocaton Hub Locaton Probem Borzou Rostam 1,2 Chrstoph Buchhem 1,4 Fautät für Mathemat, TU Dortmund, Germany J. Faban Meer 1,3 Uwe Causen 1 Insttute of Transport Logstcs,

More information

GENERATION OF GOLD-SEQUENCES WITH APPLICATIONS TO SPREAD SPECTRUM SYSTEMS

GENERATION OF GOLD-SEQUENCES WITH APPLICATIONS TO SPREAD SPECTRUM SYSTEMS GENERATION OF GOLD-SEQUENCES WITH APPLICATIONS TO SPREAD SPECTRUM SYSTEMS F. Rodríguez Henríquez (1), Member, IEEE, N. Cruz Cortés (1), Member, IEEE, J.M. Rocha-Pérez (2) Member, IEEE. F. Amaro Sánchez

More information

A parametric Linear Programming Model Describing Bandwidth Sharing Policies for ABR Traffic

A parametric Linear Programming Model Describing Bandwidth Sharing Policies for ABR Traffic parametrc Lnear Programmng Mode Descrbng Bandwdth Sharng Poces for BR Traffc I. Moschoos, M. Logothets and G. Kokknaks Wre ommuncatons Laboratory, Dept. of Eectrca & omputer Engneerng, Unversty of Patras,

More information

ECE559VV Project Report

ECE559VV Project Report ECE559VV Project Report (Supplementary Notes Loc Xuan Bu I. MAX SUM-RATE SCHEDULING: THE UPLINK CASE We have seen (n the presentaton that, for downlnk (broadcast channels, the strategy maxmzng the sum-rate

More information

Which Separator? Spring 1

Which Separator? Spring 1 Whch Separator? 6.034 - Sprng 1 Whch Separator? Mamze the margn to closest ponts 6.034 - Sprng Whch Separator? Mamze the margn to closest ponts 6.034 - Sprng 3 Margn of a pont " # y (w $ + b) proportonal

More information

MULTIVARIABLE FUZZY CONTROL WITH ITS APPLICATIONS IN MULTI EVAPORATOR REFRIGERATION SYSTEMS

MULTIVARIABLE FUZZY CONTROL WITH ITS APPLICATIONS IN MULTI EVAPORATOR REFRIGERATION SYSTEMS MULTIVARIABLE FUZZY CONTROL WITH I APPLICATIONS IN MULTI EVAPORATOR REFRIGERATION SYSTEMS LIAO QIANFANG Schoo of Eectrca and Eectronc Engneerng A thess submtted to the Nanyang Technoogca Unversty n parta

More information

International Journal "Information Theories & Applications" Vol.13

International Journal Information Theories & Applications Vol.13 290 Concuson Wthn the framework of the Bayesan earnng theory, we anayze a cassfer generazaton abty for the recognton on fnte set of events. It was shown that the obtane resuts can be appe for cassfcaton

More information

Decentralized Adaptive Control for a Class of Large-Scale Nonlinear Systems with Unknown Interactions

Decentralized Adaptive Control for a Class of Large-Scale Nonlinear Systems with Unknown Interactions Decentrazed Adaptve Contro for a Cass of Large-Scae onnear Systems wth Unknown Interactons Bahram Karm 1, Fatemeh Jahangr, Mohammad B. Menhaj 3, Iman Saboor 4 1. Center of Advanced Computatona Integence,

More information

Optimal and Suboptimal Linear Receivers for Time-Hopping Impulse Radio Systems

Optimal and Suboptimal Linear Receivers for Time-Hopping Impulse Radio Systems MITSUBISHI ELECTRIC RESEARCH LABORATORIES http://www.mer.com Optma and Suboptma Lnear Recevers for Tme-Hoppng Impuse Rado Systems Snan Gezc Hsash Kbayash H. Vncent Poor Andreas F. Mosch TR2004-052 May

More information

Quantitative Evaluation Method of Each Generation Margin for Power System Planning

Quantitative Evaluation Method of Each Generation Margin for Power System Planning 1 Quanttatve Evauaton Method of Each Generaton Margn for Poer System Pannng SU Su Department of Eectrca Engneerng, Tohoku Unversty, Japan moden25@ececetohokuacjp Abstract Increasng effcency of poer pants

More information

D hh ν. Four-body charm semileptonic decay. Jim Wiss University of Illinois

D hh ν. Four-body charm semileptonic decay. Jim Wiss University of Illinois Four-body charm semeptonc decay Jm Wss Unversty of Inos D hh ν 1 1. ector domnance. Expected decay ntensty 3. SU(3) apped to D s φν 4. Anaytc forms for form factors 5. Non-parametrc form factors 6. Future

More information

ADVANCED MACHINE LEARNING ADVANCED MACHINE LEARNING

ADVANCED MACHINE LEARNING ADVANCED MACHINE LEARNING 1 ADVANCED ACHINE LEARNING ADVANCED ACHINE LEARNING Non-lnear regresson technques 2 ADVANCED ACHINE LEARNING Regresson: Prncple N ap N-dm. nput x to a contnuous output y. Learn a functon of the type: N

More information

L-Edge Chromatic Number Of A Graph

L-Edge Chromatic Number Of A Graph IJISET - Internatona Journa of Innovatve Scence Engneerng & Technoogy Vo. 3 Issue 3 March 06. ISSN 348 7968 L-Edge Chromatc Number Of A Graph Dr.R.B.Gnana Joth Assocate Professor of Mathematcs V.V.Vannaperuma

More information

Adaptive LRBP Using Learning Automata for Neural Networks

Adaptive LRBP Using Learning Automata for Neural Networks Adaptve LRBP Usng Learnng Automata for eura etworks *B. MASHOUFI, *MOHAMMAD B. MEHAJ (#, *SAYED A. MOTAMEDI and **MOHAMMAD R. MEYBODI *Eectrca Engneerng Department **Computer Engneerng Department Amrkabr

More information

Boostrapaggregating (Bagging)

Boostrapaggregating (Bagging) Boostrapaggregatng (Baggng) An ensemble meta-algorthm desgned to mprove the stablty and accuracy of machne learnng algorthms Can be used n both regresson and classfcaton Reduces varance and helps to avod

More information

Lecture 10 Support Vector Machines II

Lecture 10 Support Vector Machines II Lecture 10 Support Vector Machnes II 22 February 2016 Taylor B. Arnold Yale Statstcs STAT 365/665 1/28 Notes: Problem 3 s posted and due ths upcomng Frday There was an early bug n the fake-test data; fxed

More information

1 Convex Optimization

1 Convex Optimization Convex Optmzaton We wll consder convex optmzaton problems. Namely, mnmzaton problems where the objectve s convex (we assume no constrants for now). Such problems often arse n machne learnng. For example,

More information

Lecture Notes on Linear Regression

Lecture Notes on Linear Regression Lecture Notes on Lnear Regresson Feng L fl@sdueducn Shandong Unversty, Chna Lnear Regresson Problem In regresson problem, we am at predct a contnuous target value gven an nput feature vector We assume

More information

Research Article H Estimates for Discrete-Time Markovian Jump Linear Systems

Research Article H Estimates for Discrete-Time Markovian Jump Linear Systems Mathematca Probems n Engneerng Voume 213 Artce ID 945342 7 pages http://dxdoorg/11155/213/945342 Research Artce H Estmates for Dscrete-Tme Markovan Jump Lnear Systems Marco H Terra 1 Gdson Jesus 2 and

More information

Distributed Moving Horizon State Estimation of Nonlinear Systems. Jing Zhang

Distributed Moving Horizon State Estimation of Nonlinear Systems. Jing Zhang Dstrbuted Movng Horzon State Estmaton of Nonnear Systems by Jng Zhang A thess submtted n parta fufment of the requrements for the degree of Master of Scence n Chemca Engneerng Department of Chemca and

More information

Development of whole CORe Thermal Hydraulic analysis code CORTH Pan JunJie, Tang QiFen, Chai XiaoMing, Lu Wei, Liu Dong

Development of whole CORe Thermal Hydraulic analysis code CORTH Pan JunJie, Tang QiFen, Chai XiaoMing, Lu Wei, Liu Dong Deveopment of whoe CORe Therma Hydrauc anayss code CORTH Pan JunJe, Tang QFen, Cha XaoMng, Lu We, Lu Dong cence and technoogy on reactor system desgn technoogy, Nucear Power Insttute of Chna, Chengdu,

More information

Discriminating Fuzzy Preference Relations Based on Heuristic Possibilistic Clustering

Discriminating Fuzzy Preference Relations Based on Heuristic Possibilistic Clustering Mutcrtera Orderng and ankng: Parta Orders, Ambgutes and Apped Issues Jan W. Owsńsk and aner Brüggemann, Edtors Dscrmnatng Fuzzy Preerence eatons Based on Heurstc Possbstc Custerng Dmtr A. Vattchenn Unted

More information

Pulse Coded Modulation

Pulse Coded Modulation Pulse Coded Modulaton PCM (Pulse Coded Modulaton) s a voce codng technque defned by the ITU-T G.711 standard and t s used n dgtal telephony to encode the voce sgnal. The frst step n the analog to dgtal

More information

Correspondence. Performance Evaluation for MAP State Estimate Fusion I. INTRODUCTION

Correspondence. Performance Evaluation for MAP State Estimate Fusion I. INTRODUCTION Correspondence Performance Evauaton for MAP State Estmate Fuson Ths paper presents a quanttatve performance evauaton method for the maxmum a posteror (MAP) state estmate fuson agorthm. Under dea condtons

More information

Support Vector Machines for Classification and Regression

Support Vector Machines for Classification and Regression ISIS Technca Report Support Vector Machnes for Cassfcaton and Regresson Steve Gunn 0 November 997 Contents Introducton 3 2 Support Vector Cassfcaton 4 2. The Optma Separatng Hyperpane...5 2.. Lneary Separabe

More information

Multigradient for Neural Networks for Equalizers 1

Multigradient for Neural Networks for Equalizers 1 Multgradent for Neural Netorks for Equalzers 1 Chulhee ee, Jnook Go and Heeyoung Km Department of Electrcal and Electronc Engneerng Yonse Unversty 134 Shnchon-Dong, Seodaemun-Ku, Seoul 1-749, Korea ABSTRACT

More information

Pop-Click Noise Detection Using Inter-Frame Correlation for Improved Portable Auditory Sensing

Pop-Click Noise Detection Using Inter-Frame Correlation for Improved Portable Auditory Sensing Advanced Scence and Technology Letters, pp.164-168 http://dx.do.org/10.14257/astl.2013 Pop-Clc Nose Detecton Usng Inter-Frame Correlaton for Improved Portable Audtory Sensng Dong Yun Lee, Kwang Myung Jeon,

More information

NUMERICAL DIFFERENTIATION

NUMERICAL DIFFERENTIATION NUMERICAL DIFFERENTIATION 1 Introducton Dfferentaton s a method to compute the rate at whch a dependent output y changes wth respect to the change n the ndependent nput x. Ths rate of change s called the

More information

Cyclic Codes BCH Codes

Cyclic Codes BCH Codes Cycc Codes BCH Codes Gaos Feds GF m A Gaos fed of m eements can be obtaned usng the symbos 0,, á, and the eements beng 0,, á, á, á 3 m,... so that fed F* s cosed under mutpcaton wth m eements. The operator

More information

Average Decision Threshold of CA CFAR and excision CFAR Detectors in the Presence of Strong Pulse Jamming 1

Average Decision Threshold of CA CFAR and excision CFAR Detectors in the Presence of Strong Pulse Jamming 1 Average Decson hreshold of CA CFAR and excson CFAR Detectors n the Presence of Strong Pulse Jammng Ivan G. Garvanov and Chrsto A. Kabachev Insttute of Informaton echnologes Bulgaran Academy of Scences

More information

Delay tomography for large scale networks

Delay tomography for large scale networks Deay tomography for arge scae networks MENG-FU SHIH ALFRED O. HERO III Communcatons and Sgna Processng Laboratory Eectrca Engneerng and Computer Scence Department Unversty of Mchgan, 30 Bea. Ave., Ann

More information

On Uplink-Downlink Sum-MSE Duality of Multi-hop MIMO Relay Channel

On Uplink-Downlink Sum-MSE Duality of Multi-hop MIMO Relay Channel On Upn-Downn Sum-MSE Duat of Mut-hop MIMO Rea Channe A Cagata Cr, Muhammad R. A. handaer, Yue Rong and Yngbo ua Department of Eectrca Engneerng, Unverst of Caforna Rversde, Rversde, CA, 95 Centre for Wreess

More information

Uncertainty in measurements of power and energy on power networks

Uncertainty in measurements of power and energy on power networks Uncertanty n measurements of power and energy on power networks E. Manov, N. Kolev Department of Measurement and Instrumentaton, Techncal Unversty Sofa, bul. Klment Ohrdsk No8, bl., 000 Sofa, Bulgara Tel./fax:

More information

Achieving Optimal Throughput Utility and Low Delay with CSMA-like Algorithms: A Virtual Multi-Channel Approach

Achieving Optimal Throughput Utility and Low Delay with CSMA-like Algorithms: A Virtual Multi-Channel Approach Achevng Optma Throughput Utty and Low Deay wth SMA-ke Agorthms: A Vrtua Mut-hanne Approach Po-Ka Huang, Student Member, IEEE, and Xaojun Ln, Senor Member, IEEE Abstract SMA agorthms have recenty receved

More information

Characterizing Probability-based Uniform Sampling for Surrogate Modeling

Characterizing Probability-based Uniform Sampling for Surrogate Modeling th Word Congress on Structura and Mutdscpnary Optmzaton May 9-4, 3, Orando, Forda, USA Characterzng Probabty-based Unform Sampng for Surrogate Modeng Junqang Zhang, Souma Chowdhury, Ache Messac 3 Syracuse

More information

A MIN-MAX REGRET ROBUST OPTIMIZATION APPROACH FOR LARGE SCALE FULL FACTORIAL SCENARIO DESIGN OF DATA UNCERTAINTY

A MIN-MAX REGRET ROBUST OPTIMIZATION APPROACH FOR LARGE SCALE FULL FACTORIAL SCENARIO DESIGN OF DATA UNCERTAINTY A MIN-MAX REGRET ROBST OPTIMIZATION APPROACH FOR ARGE SCAE F FACTORIA SCENARIO DESIGN OF DATA NCERTAINTY Travat Assavapokee Department of Industra Engneerng, nversty of Houston, Houston, Texas 7704-4008,

More information

Diplomarbeit. Support Vector Machines in der digitalen Mustererkennung

Diplomarbeit. Support Vector Machines in der digitalen Mustererkennung Fachberech Informatk Dpomarbet Support Vector Machnes n der dgtaen Mustererkennung Ausgeführt be der Frma Semens VDO n Regensburg vorgeegt von: Chrstan Mkos St.-Wofgangstrasse 9305 Regensburg Betreuer:

More information

DISTRIBUTED PROCESSING OVER ADAPTIVE NETWORKS. Cassio G. Lopes and Ali H. Sayed

DISTRIBUTED PROCESSING OVER ADAPTIVE NETWORKS. Cassio G. Lopes and Ali H. Sayed DISTRIBUTED PROCESSIG OVER ADAPTIVE ETWORKS Casso G Lopes and A H Sayed Department of Eectrca Engneerng Unversty of Caforna Los Angees, CA, 995 Ema: {casso, sayed@eeucaedu ABSTRACT Dstrbuted adaptve agorthms

More information

Supervised Learning. Neural Networks and Back-Propagation Learning. Credit Assignment Problem. Feedforward Network. Adaptive System.

Supervised Learning. Neural Networks and Back-Propagation Learning. Credit Assignment Problem. Feedforward Network. Adaptive System. Part 7: Neura Networ & earnng /2/05 Superved earnng Neura Networ and Bac-Propagaton earnng Produce dered output for tranng nput Generaze reaonaby & appropratey to other nput Good exampe: pattern recognton

More information

Optimal Guaranteed Cost Control of Linear Uncertain Systems with Input Constraints

Optimal Guaranteed Cost Control of Linear Uncertain Systems with Input Constraints Internatona Journa Optma of Contro, Guaranteed Automaton, Cost Contro and Systems, of Lnear vo Uncertan 3, no Systems 3, pp 397-4, wth Input September Constrants 5 397 Optma Guaranteed Cost Contro of Lnear

More information

Reactive Power Allocation Using Support Vector Machine

Reactive Power Allocation Using Support Vector Machine Reactve Power Aocaton Usng Support Vector Machne M.W. Mustafa, S.N. Khad, A. Kharuddn Facuty of Eectrca Engneerng, Unverst Teknoog Maaysa Johor 830, Maaysa and H. Shareef Facuty of Eectrca Engneerng and

More information

For now, let us focus on a specific model of neurons. These are simplified from reality but can achieve remarkable results.

For now, let us focus on a specific model of neurons. These are simplified from reality but can achieve remarkable results. Neural Networks : Dervaton compled by Alvn Wan from Professor Jtendra Malk s lecture Ths type of computaton s called deep learnng and s the most popular method for many problems, such as computer vson

More information

GENERATIVE AND DISCRIMINATIVE CLASSIFIERS: NAIVE BAYES AND LOGISTIC REGRESSION. Machine Learning

GENERATIVE AND DISCRIMINATIVE CLASSIFIERS: NAIVE BAYES AND LOGISTIC REGRESSION. Machine Learning CHAPTER 3 GENERATIVE AND DISCRIMINATIVE CLASSIFIERS: NAIVE BAYES AND LOGISTIC REGRESSION Machne Learnng Copyrght c 205. Tom M. Mtche. A rghts reserved. *DRAFT OF September 23, 207* *PLEASE DO NOT DISTRIBUTE

More information

Multilayer Perceptrons and Backpropagation. Perceptrons. Recap: Perceptrons. Informatics 1 CG: Lecture 6. Mirella Lapata

Multilayer Perceptrons and Backpropagation. Perceptrons. Recap: Perceptrons. Informatics 1 CG: Lecture 6. Mirella Lapata Multlayer Perceptrons and Informatcs CG: Lecture 6 Mrella Lapata School of Informatcs Unversty of Ednburgh mlap@nf.ed.ac.uk Readng: Kevn Gurney s Introducton to Neural Networks, Chapters 5 6.5 January,

More information

VQ widely used in coding speech, image, and video

VQ widely used in coding speech, image, and video at Scalar quantzers are specal cases of vector quantzers (VQ): they are constraned to look at one sample at a tme (memoryless) VQ does not have such constrant better RD perfomance expected Source codng

More information

Generalized Linear Methods

Generalized Linear Methods Generalzed Lnear Methods 1 Introducton In the Ensemble Methods the general dea s that usng a combnaton of several weak learner one could make a better learner. More formally, assume that we have a set

More information

Part II. Support Vector Machines

Part II. Support Vector Machines Part II Support Vector Machnes 35 Chapter 5 Lnear Cassfcaton 5. Lnear Cassfers on Lnear Separabe Data As a frst step n understandng and constructng Support Vector Machnes e stud the case of near separabe

More information

Problem Set 9 Solutions

Problem Set 9 Solutions Desgn and Analyss of Algorthms May 4, 2015 Massachusetts Insttute of Technology 6.046J/18.410J Profs. Erk Demane, Srn Devadas, and Nancy Lynch Problem Set 9 Solutons Problem Set 9 Solutons Ths problem

More information