Signal Processing

Transcript

MAHATMA GANDHI UNIVERSITY SCHEME AND SYLLABI FOR M. Tech. DEGREE PROGRAMME IN SIGNAL PROCESSING (2013 ADMISSION ONWARDS) SCHEME AND SYLLABI FOR M. Tech. DEGREE PROGRAMME IN SIGNAL PROCESSING SEMESTER – II Hrs / Week Sl. No. Course No. Evaluation Scheme (Marks) Subject Sessional L T P TA CT Sub Total ESE Total Credits (C) 1 MAESP 201 Adaptive & Non Linear Signal Processing 3 1 0 25 25 50 100 150 4 2 MAESP 202* Estimation & Detection Theory 3 1 0 25 25 50 100 150 4 3 MAESP 203 Theory of Transforms 3 1 0 25 25 50 100 150 4 4 MAESP 204 Digital Image and Video Processing 3 1 0 25 25 50 100 150 4 5 MAESP 205 Elective – III 3 0 0 25 25 50 100 150 3 6 MAESP 206 Elective – IV 3 0 0 25 25 50 100 150 3 7 MAESP 207 Signal Processing Lab – II 0 0 3 25 25 50 100 150 2 8 MAESP 208 Seminar -II 0 0 2 50 0 50 0 50 1 Total 18 4 5 225 175 400 700 1100 25 Elective – III (MAESP 205) Elective – IV (MAESP 206) MAESP 205 – 1# Wireless Communication Systems MAESP 205 – 2 Multidimensional Signal Processing MAESP 205 – 3 VLSI Structures for Digital Signal Processing MAESP 206 - 3 Spectrum Analysis MAESP 205 – 4 Information Hiding and Data Encryption MAESP 206 - 4 Pattern Recognition & Analysis MAESP 206 - 1 MAESP 206 – 2** Array Signal Processing Spread Spectrum and CDMA Systems L – Lecture, T – Tutorial, P – Practical TA – Teacher’s Assessment (Assignments, attendance, group discussion, tutorials, seminars, etc.) CT – Class Test (Minimum of two tests to be conducted by the Institute) ESE – End Semester Examination to be conducted by the University Electives: New Electives may be added by the department according to the needs of emerging fields of technology. The name of the elective and its syllabus should be submitted to the University before the course is offered. *Common with MECCE, **- Common with MECCI, # - Common with MECAE 1 MAESP 201 ADAPTIVE & NON LINEAR SIGNAL PROCESSING L T P C 3 1 0 4 Module 1: Review of discrete time Complex Gaussian processes, MA, AR, ARMA processes and their properties, MMSE predictors, LMMSE predictor, orthogonality theorem (concept of innovation processes), Weiner filter, Yule-walker equation, unconstrained Weiner filter (in z domain), recursive Weiner filter (using innovation process). Kalman filter, recursions in Kalman filter. Module 2: Filters with recursions based on the steepest descent and Newton's method, criteria for the convergence, rate of convergence. LMS filter, mean and variance of LMS, the MSE of LMS and misadjusment, Convergence of LMS. Module 3: RLS recursions, assumptions for RLS, convergence of RLS coefficients and MSE. Filter based on innovations, generation of forward and backward innovations, forward and reverse error recursions. Implementation of Weiner, LMS and RLS filters using lattice filters, Linear Prediction, Levinson Durbin algorithm, reverse Levinson Durbin algorithm. Module 4: Non-linear signal processing: Non-linear filters, Non-gaussian models, Generalized Gaussian and stable distributions, Median smoothers, Rank/order filters, Weighted median smoother. References: 1. S. Haykin, “Adaptive Filters Theory”, Prentice-Hall. 2. Dimitris G. Manolakis, Vinay K. Ingle, Stephan M Krgon, “Statistical and Adaptive Signal Processing”, Mc Graw Hill (2000) 3. G. R. Arce , “Non-linear signal processing: A statistical approach”, Wiley 2004. 4. Monson Hayes, “Statistical Signal Processing and Modelling”, Wiley India Pvt. Ltd 2 5. J. Astola, P. Kuosmanen, “Fundamentals of non-linear digital filtering”, CRC Press, 1997. 6. Proakis & Manolakis, “Digital Signal Processing”. PHI, New Delhi 7. S. J. Orfanidis, “Optimum Signal Processing”, Mc-Graw Hill.. 8. Ifeacher,“ Digital Signal Processing,” Addision Wesley 9. Sanjit k. Mitra, “ Digital Signal Processing”,TMH 10. A. V. Oppenheim & Ronald W. Schafer , “Discrete Time Signal processing”, PHI, New Delhi 11. Jones D. Adaptive Filters [Connexions Web site]. May 12, 2005. Available at: http://cnx.rice.edu/content/col10280/1.1/ 3 MAESP 202 ESTIMATION AND DETECTION THEORY L T P C 3 1 0 4 Module 1: Fundamentals of Detection Theory Hypothesis Testing: Bayes’ Detection, MAP Detection, ML Detection, Minimum Probability of Error Criterion, Min-Max Criterion, Neyman-Pearson Criterion, Multiple Hypothesis, Composite Hypothesis Testing: Generalized likelihood ratio test (GLRT), Receiver Operating Characteristic Curves. Module 2: Fundamentals of Estimation Theory Role of Estimation in Signal Processing, Unbiased Estimation, Minimum variance unbiased(MVU) estimators, Finding MVU Estimators, Cramer-Rao Lower Bound, Linear Modeling-Examples, Sufficient Statistics, Use of Sufficient Statistics to find the MVU Estimator Module 3: Estimation Techniques Deterministic Parameter Estimation: Least Squares Estimation-Batch Processing, Recursive Least Squares Estimation, Best Linear Unbiased Estimation, Likelihood and Maximum Likelihood Estimation Module 4: Estimation Techniques (contd) Random Parameter Estimation: Bayesian Philosophy, Selection of a Prior PDF, Bayesian linear model, Minimum Mean Square Error Estimator, Maximum a Posteriori Estimation References: 1. Steven M. Kay, “Statistical Signal Processing: Vol. 1: Estimation Theory, Vol. 2: Detection Theory,” Prentice Hall Inc., 1998. 2. M D Srinath, P K Rajasekaran, R Viswanathan, “Introduction to Statistical Signal Processing with Applications”, Pearson, 1995. 3. H. Vincent Poor, “An Introduction to Signal Detection and Estimation”, 2nd Edition, Springer, 1994. 4. Jerry M. Mendel, “Lessons in Estimation Theory for Signal Processing, Communication and Control," Prentice Hall Inc., 1995 4 5. Ralph D. Hippenstiel, “Detection Theory- Applications and Digital Signal Processing”, CRC Press, 2002. 6. Bernard C. Levy, “Principles of Signal Detection and Parameter Estimation”, Springer, New York, 2008. 7. Harry L. Van Trees, “Detection, Estimation and Modulation Theory, Part 1 and 2," John Wiley & Sons Inc. 1968. 8. Neel A. Macmillan and C. Douglas Creelman, “Detection Theory: A User's Guide (Sec. Edn.)” Lawrence Erlbaum Associates Publishers, USA, 2004. 9. Monson H. Hayes, “Statistical Digital Signal Processing and Modelling," John Wiley & Sons Inc., 1996. 5 MAESP 203 THEORY OF TRANSFORMS L T P C 3 1 0 4 Module 1 Functionals - Norm, Convergence - Cauchy sequence, Completeness of vector spaces; Infinite dimensional vector spaces - Normed linear spaces; Banach Spaces, Inner product spaces, Hilbert spaces; Continuous linear operators. Bounded Linear Operators and Spectral Theory Bounded linear operators in finite dimensional inner product spaces Adjoint of an operator, Norm of an operator; Self-adjoint operators - Spectral analysis of self-adjoint operators; Bessel’s inequality, Parseval’s identity; Reisz Representation Theorem, Compact linear operators Module 2 Generalized functions and the Dirac’s delta; Differential operators - Green’s function and the inverse linear operators. The Making of Integral Transforms The making of Fourier transform, Self-reciprocal functions and operators under Fourier transform - The construction of Fractional Fourier transform Module 3 The making of Laplace transform, Construction of z-transform - Discrete-time Fourier transform and discrete Fourier transform. Lapped Transforms Karhunen-Loeve transform - Lapped orthogonal transforms and biorthogonal transforms – Construction of discrete cosine and sine transforms. Module 4 Reisz basis, Resolution of unity, Definition of frames (introduction only), Geometrical considerations and the general notion of a frame, Frame projector, Example - windowed Fourier frames; Continuous wavelet transform, Introduction to DWT. References: 1. Erwin Kreyszig, “Introductory Functional Analysis with Applications,” John Wiley and Sons, 1989. 2. Athanasios Papoulis, “Fourier Integral and its Applications,” McGraw-Hill International, New York, 1962. 3. Athanasios Papoulis, “Systems and Transforms with Applications in Optics,” 6 McGraw-Hill International, New York, 1968. 4. Lokenath Debnath and Piotr Mikusinski, “Hilbert Spaces with Applications,” 3rd Edition, Academic Press, Indian reprint 2006. 5. A. David Wunsch, “Complex Variables with Applications,” 2nd Edition, AddisonWesley Publishing Company, New York, 1994. 6. George Bachman and Lawrence Narici, “Functional Analysis,” Dover Publications Inc., 2000. 7. Frederick W Byron, Jr and Robert W Fuller, “Mathematics of Classical and Quantum Physics,” Dover Publications Inc., 1992. 8. Anthony N. Michel and Charles J. Herget, “Applied Algebra and Functional Analysis,” Dover Publications Inc., 1993. 9. Gerald Kaiser, “A Friendly Guide to Wavelets,” Birkhauser/Springer International Edition, 1994, Indian reprint 2005. 10. Ingrid Daubechies, “Ten Lectures on Wavelets,” SIAM, 1990. 11. Martin Vetterli & Jelena Kovacevic, Wavelets and Subband Coding, Prentice Hall, 2007. 7 MAESP 204 DIGITAL IMAGE AND VIDEO PROCESSING L T P C 3 1 0 4 Module 1: Image representation: Gray scale and colour Images, image sampling and quantization. Two dimensional orthogonal transforms: DFT, WHT, Haar transform, KLT, DCT. Image enhancement - filters in spatial and frequency domains, histogram-based processing, homomorphic filtering. Edge detection, LOG filters. Module 2: Image Restoration: Degradation Models, PSF, circulant and block - circulant matrices, deconvolution, restoration using inverse filtering and Wiener filtering. Image Segmentation: Pixel classification, Bi-level thresholding, Multi-level thresholding, Adaptive thresholding, Edge detection, Hough transform, Region growing. Module 3: Fundamental concepts of image compression - Compression models – Information theoretic perspective - Fundamental coding theorem - Lossless Compression: Huffman Coding- Arithmetic coding - Bit plane coding - Run length coding. Lossy compression: Transform coding - Image compression standards. Module 4: Video Processing: Representation of Digital Video, Spatio-temporal sampling; Motion Estimation; Video Filtering; Video Compression, Video coding standards- H.264 References : 1. A. K. Jain, “Fundamentals of digital image processing”, Prentice Hall of India, 1989. 2. R. C. Gonzalez, R. E. Woods, “Digital Image Processing”, Pearson Education. 3. Iain E Richardson, “H.264 and MPEG-4 Video Compression”, John Wiley & Sons, September 2003 4. A. M. Tekalp, “Digital Video Processing” , Prentice-Hall 5. A Bovik, “Handbook of Image & Video Processing”, Academic Press, 2000 6. W. K. Pratt, “Digital image processing”, Prentice Hall 8 7. A. Rosenfeld and A. C. Kak, “Digital image processing”, Vols. 1 and 2, Prentice Hall, 1986. 8. H. C. Andrew and B. R. Hunt, “Digital image restoration”, Prentice Hall, 1977 9. R. Jain, R. Kasturi and B.G. Schunck, “Machine Vision”, McGraw-Hill International Edition, 1995 10. K.R.Rao, Zoran.S Bojkovic, Dragorad A Milovanovic, “Multimedia Communication Systems: Techniques ,standards and Networks” , Prentice Hall 9 MAESP/MECAE 205 - 1 WIRELESS COMMUNICATION SYSTEM L T P C 3 0 0 3 Module 1: Fading and Diversity Wireless Channel Models: Path Loss and Shadowing Models, Statistical Fading Models, Narrow Band and Wideband Fading Models. Diversity: Time Diversity, Frequency and Space Diversity, Receive Diversity, Concept of Diversity Branches and Signal Paths, Performance Gains; Combining Methods: Selective Combining, Maximal Ratio Combining, Equal Gain Combining. Module 2: Cellular Communication Cellular Networks; Multiple Access: FDMA, TDMA, Spatial Reuse, Co-Channel Interference Analysis, Hand-off, Erlang Capacity Analysis, Spectral Efficiency and Grade of Service, Improving Capacity: Cell Splitting and Sectorization. Module 3: Spread spectrum and CDMA Motivation- Direct sequence spread spectrum- Frequency Hopping systems- Time Hopping.- Anti-jamming- Pseudo Random (PN) sequence- Maximal length sequencesGold sequences- Generation of PN sequences.- Diversity in DS-SS systems- Rake Receiver- Performance analysis. Spread Spectrum Multiple Access- CDMA SystemsInterference Analysis for Broadcast and Multiple Access Channels- Capacity of cellular CDMA networks- Reverse link power control- Hard and Soft hand off strategies. Module 4: Fading Channel Capacity: Capacity of Wireless Channels- Capacity of flat and frequency selective fading channels- Multiple Input Multiple output (MIMO) systems- Narrow band multiple antenna system model- Parallel Decomposition of MIMO Channels- Capacity of MIMO Channels. Cellular Wireless Communication Standards, Second generation cellular systems: GSM specifications and Air Interface - specifications, IS 95 CDMA- 3G systems: UMTS & CDMA 2000 standards and specifications References: 1. Andrea Goldsmith, “Wireless Communications”, Cambridge University press. 2. Simon Haykin and Michael Moher, “Modern Wireless Communications”, Pearson Education. 10 3. T.S. Rappaport, “Wireless Communication, principles & practice”, PHI, 2001. 4. G.L Stuber, “Principles of Mobile Communications”, 2nd edition, Kluwer Academic Publishers. 5. Kamilo Feher, “Wireless digital communication”, PHI, 1995. 6. R.L Peterson, R.E. Ziemer and David E. Borth, “Introduction to Spread Spectrum Communication”, Pearson Education. 7. A.J.Viterbi, CDMA- “Principles of Spread Spectrum”, Addison Wesley, 1995. 8. D. Tse & P. Viswanath, “Fundamentals of Wireless Communication”, Cambridge University Press, 2005. 11 MAESP 205 -2 MULTIDIMENSIONAL SIGNAL PROCESSING L T P C 3 0 0 3 Module 1: Multidimensional systems Fundamental operations on Multidimensional signals, Linear Shift - Invariant systemscascade and parallel connection of systems- separable systems, stable systems- Frequency responses of 2D LTI Systems- Impulse response- Multidimensional Fourier transforms- z transform, properties of the Fourier and z transform. Module 2: Sampling continuous 2D signals Periodic sampling with rectangular geometry- sampling density, Aliasing effects created by sampling - Periodic sampling with hexagonal geometry. Module 3: Multidimensional Discrete Fourier Transform Multidimensional discrete Fourier transform- Properties of DFT, Circular convolutionCalculation of DFT- DFT for periodically sampled signals - Fast Fourier transform for periodically sampled signals. Module 4: Multidimensional Digital Filter Design Separable Filters- Linear phase filters- FIR Filters- Implementation of FIR filters - design of FIR filters using windows- Two dimensional window functions, IIR Filters References : 1. John Woods, “Multidimensional signal, image, and video processing and coding”, Academic Press, 2006. 2. Dudgeon Dan E., “Multidimensional Digital Signal Processing”, Prentice Hall, Englewood Cliffs, New Jersey 3. P.P. Vaidyanathan. “Multirate systems and filter banks.” Prentice Hall. PTR. 1993. 4. Jae S. Lim, “Two- Dimensional Signal and Image Processing”, Prentice Hall Englewood Cliffs, New Jersey, 1990. 12 MAESP 205 -3 VLSI STRUCTURES FOR DIGITAL SIGNAL PROCESSING L T P C 3 0 0 3 Module I Representation of DSP Algorithms – Block diagrams, Signal flow graph, Data flow graph – Pipelining of FIR digital filters – parallel processing for FIR systems – combined pipelining and parallel processing of FIR filters for low power – Pipelining in IIR filters – parallel processing for IIR filters – combined pipelining and parallel processing of FIR filters. Module II Retiming – definitions and properties, solving system of inequalities, retiming techniques, Unfolding – algorithm for unfolding, properties of unfolding, critical path, unfolding and retiming, applications of unfolding, Folding – folding transformation, register minimization techniques, register minimization in folded architectures, folding of multirate systems Module III Parallel FIR filters – discrete time cosine transform – implementation of DCT based on algorithm – architecture transformations – parallel architectures for rank order filters. Module IV Scaling and round off noise – round off noise in pipelined IIR filters – round off noise in lattice filters – pipelining of lattice IIR digital filters – low power CMOS lattice IIR filters. Reference: 1. Keshab K. Parhi, “VLSI Digital signal processing Systems: Design and Implementation”, John Wiley & Sons, 1999. 2. Uwe meyer-Baes, “DSP with Field programmable gate arrays”, Springer, 2001 3. Sen M Kuo, Woon-Seng S. Gan, “ Digital Signal Processors : Architectures, Implementations and applications”, Prentice Hall, 2004 4. Lars Wanhammar , “DSP integrated circuits”, Academic Press, 1999. 13 MAESP 205 - 4 INFORMATION HIDING AND DATA ENCRYPTION L T P C 3 0 0 3 Module 1: Introduction to Cryptography OSI Security Architecture, Classical Encryption techniques, Cipher Principles, Data Encryption Standard, Block Cipher Design Principles and Modes of Operation, Evaluation criteria for AES, AES Cipher, Triple DES, Placement of Encryption Function , Traffic Confidentiality Module 2: Public Key Cryptography Key Management, Diffie-Hellman key Exchange, Elliptic Curve Architecture and Cryptography, Introduction to Number Theory, Confidentiality using Symmetric Encryption, Public Key Cryptography and RSA. Practical implementation of Cryptography Module 3: Information Hiding: Principle and Objectives of Watermarking and Steganography. Mathematical formulations, Public - Private Key Steganography, Information hiding in noisy data (adaptive and nonadapive )and written texts. Module 4: Steganographic techniques: - substitution and bitplane tools - transform domain toolsSpread Spectrum Techniques- Statistical methods- Distortion and Cover Generation methods. Steganalysis: - of images and audio. Watermarking:- techniques, methods, benchmarks for digital watermarking. Practical implementation of steganograpgy. References : 1. Stefan Katzenbeisser, Fabien A. P. Petitcolas, “Information Hiding Techniques for Steganography and Digital Watermarking”, Artech House Publishers, 2000. 2. Neal Koblitz, “A Course in Number Theory and Cryptography”, 2nd Edition, Springer 3. William Stallings, “Cryptography And Network Security – Principles and Practices”, Prentice Hall of India, Third Edition, 2003. 4. Bruce Schneier, “Applied Cryptography”, John Wiley & Sons Inc, 2001. 14 5. Charles B. Pfleeger, Shari Lawrence Pfleeger, “Security in Computing”, Third Edition, Pearson Education, 2003. 6. H.S. Zuckerman , “An Introduction to the theory of Numbers”, 5th Edition, John Wiley & Sons 7. A.J. Menezes etc al, “Handbook of Applied Cryptography”, CRC Press 8. Branislav Kisacanin, “Mathematical Problems and Proofs, Combinatorics, Number theory and Geometry”. 9. Atul Kahate, “Cryptography and Network Security”, Tata McGraw-Hill, 2003 15 MAESP 206 - 1 ARRAY SIGNAL PROCESSING L T P C 3 0 0 3 Module 1: Spatial Signals : Signals in space and time. Spatial frequency, Direction vs. frequency. Wave fields. Far field and Near field signals. Module 2: Sensor Arrays : Spatial sampling, Nyquist criterion. Sensor arrays. Uniform linear arrays, planar and random arrays. Array transfer (steering) vector. Array steering vector for ULA. Broadband arrays Module 3: Spatial Frequency: Aliasing in spatial frequency domain. Spatial Frequency Transform, Spatial spectrum. Spatial Domain Filtering. Beam Forming. Spatially white signal Module 4: Direction of Arrival Estimation: Non parametric methods - Beam forming and Capon methods. Resolution of Beam forming method. Subspace methods - MUSIC, Minimum Norm and ESPRIT techniques. Spatial Smoothing. References : 1. Dan E. Dugeon and Don H. Johnson, “:Array Signal Processing: Concepts and Techniques”. Prentice Hall, 1993. 2. Petre Stoica and Randolph L. Moses, “ Spectral Analysis of Signals”. Prentice Hall ,2005. 3. Bass J, McPheeters C, Finnigan J, Rodriguez E. “Array Signal Processing” [Connexions Website]. February http://cnx.rice.edu/content/col10255/1.3/ 16 8, 2005. Available at: MAESP 206 - 2 SPREAD SPECTRUM AND CDMA SYSTEMS L T P C 3 0 0 3 Module 1: Fundamentals of Spread Spectrum Introduction to spread spectrum communication, direct sequence spread spectrum, frequency-hop spread spectrum system. Spreading sequences- maximal-length sequences, gold codes, Walsh orthogonal codes- properties and generation of sequences. Synchronization and Tracking: delay lock and tau-dither loops, coarse synchronizationprinciples of serial search and match filter techniques. Module 2: Performance Analysis of SS system Performance of spread spectrum system in jamming environments- Barrage noise jamming, partial band jamming, pulsed noise jamming and single tone jamming. Error probability of DS-CDMA system under AWGN and fading channels, RAKE receiver Module 3: Capacity, Coverage and multiuser detection Basics of spread spectrum multiple access in cellular environments, reverse Link power control, multiple cell pilot tracking, soft and hard handoffs, cell coverage issues with hard and soft handoff, spread spectrum multiple access outage, outage with imperfect power control, Erlang capacity of forward and reverse links. Multi-user Detection -MF detector, decorrelating detector, MMSE detector. Interference Cancellation: successive, Parallel Interference Cancellation, performance analysis of multiuser detectors and interference cancellers. Module 4: CDMA Systems General aspects of CDMA cellular systems, IS-95 standard, Downlink and uplink, Evolution to Third Generation systems, WCDMA and CDMA-2000 standards, Principles of Multicarrier communication, MCCDMA and MC-DS-CDMA. References : 1. Valery P. Ipatov, “Spread Spectrum and CDMA Principles and Applications”, Wiley, 2005 2. R. L. Peterson, R. Ziemer and D. Borth, “Introduction to Spread Spectrum Communications,” Prentice Hall, 1995. 17 3. A. J. Viterbi, “CDMA - Principles of Spread Spectrum Communications,” Addison-Wesley, 1997. 4. S. Verdu, “ Multiuser Detection” , Cambridge University Press- 1998 5. M. K. Simon, J. K. Omura, R. A. Scholts and B. K. Levitt, “ Spread Spectrum Communications Handbook”, McGraw- Hill, Newyork-1994 6. Cooper and McGillem, “Modern Communications and Spread Spectrum” McGraw- Hill, 1985 7. S. Glisic and B. Vucetic, “Spread Spectrum CDMA Systems for Wireless Communications,” Artech House, 1997 18 MAESP 206 - 3 SPECTRUM ANALYSIS L T P C 3 0 0 3 Module1: Power Spectral Density: Energy spectral density of deterministic signals, Power spectral density of random signals, Properties of PSD. Module 2: PSD Estimation - Non-parametric methods Estimation of PSD from finite data, Nonparametric methods : Periodogram properties, bias and variance analysis, BlackmanTuckey method, Window design considerations, time-bandwidth product and resolution variance trade-offs in window design, Refined periodogram methods : Bartlet method, Welch method. Module 3: PSD Estimation - Parametric methods: Parametric method for rational spectra :Covariance structure of ARMA process, AR signals, Yule-Walker method, Least square method, Levinson-Durbin Algorithm, MA signals, Modified Yule-Walker method, Twostage least square method, Burg method for AR parameter estimation. Parametric method for line spectra :- Models of sinusoidal signals in noise, Non-linear least squares method, Higher order Yule-Walker method, MUSIC and Pisarenko methods, Min-norm method, ESPRIT method. Module 4: Filterbank methods: Filterbank interpretation of periodogram, Slepia base-band filters, refined filterbank method for higher resolution spectral analysis, Capon method, Introduction to higher order spectra. References : 1. P. Stoica , R.L. Moses, “Introduction to Spectral Analysis”, Prentice Hall 2. Kay S, M, “Modern Spectral Estimation Theory & Applications”, Prentice Hall 19 MAESP 206 - 4 PATTERN RECOGNITION AND ANALYSIS L T P C 3 0 0 3 Module 1 Introduction - features, feature vectors and classifiers, Supervised versus unsupervised pattern recognition. Classifiers based on Bayes Decision theory- introduction, discriminant functions and decision surfaces, Bayesian classification for normal distributions, Estimation of unknown probability density functions, the nearest neighbour rule. Linear classifiers,- Linear discriminant functions and decision hyper planes, The perceptron algorithm, MSE estimation, Logistic discrimination, Support Vector machines. Module II Non-Linear classifiers- Two layer and three layer perceptrons, Back propagation algorithm, Networks with Weight sharing, Polynomial classifiers, Radial Basis function networks. Module III Non-Linear classifiers- Support Vector machines-nonlinear case, Decision trees, Combining classifiers. Feature selection – ROC, Class separability measures, Optimal feature generation, The Bayesian information criterion. Feature generation – KLT and SVD. Context dependent classification – Markov Chain Model, The Viterbi Algorithm. Module IV Clustering- Cluster analysis, Proximity measures, Clustering Algorithms - Sequential algorithms. Hierarchical algorithms - Agglomerative algorithms, Divisive algorithms. Schemes based on function optimization - Fuzzy clustering algorithms, Probabilistic clustering, K - means algorithm. Clustering algorithms based on graph theory , Competitive learning algorithms, Boundary detection methods, Valley seeking clustering, Kernel clustering methods. Clustering validity - basics References : 1. Sergios Theodoridis, Konstantinos Koutroumbas, “Pattern Recognition”, Academic Press, 2006. 2. Richard O. Duda and Hart P.E, and David G Stork, “Pattern classification” , 2nd Edn., John Wiley & Sons Inc., 2001 20 3. Christopher M Bishop, “Pattern Recognition and Machine Learning”, Springer 2007. 4. Robert Schalkoff, “Pattern Recognition – Statistical, Structural and Neural Approaches”, Wiley India 5. Earl Gose, Richard Johnsonbaugh, and Steve Jost; “Pattern Recognition and Image Analysis”, PHI Pvte. Ltd., NewDelhi-1, 1999. 6. Fu K.S., “Syntactic Pattern recognition and applications”, Prentice Hall, Eaglewood cliffs, N.J., 1982 7. Andrew R. Webb, “Statistical Pattern Recognition”, John Wiley & Sons, 2002. 21 MAESP 207 SIGNAL PROCESSING LAB -II L T P C 0 0 3 2 Tools- Matlab, DSP Kits – TMS320C6XX Multirate Signal Processing – Decimation and Interpolation, Noble Identities, Polyphase Decomposition. Speech processing Understanding various audio and speech file formats and conversion utilities. Reading and writing audio files using Matlab. Implementation of various audio coding algorithms Delta Modulation, Adaptive Delta Modulation, Subband Coding. Image Processing Reading, display, and saving of different image file formats using Matlab Implementation of 2-D transforms- (DFT/ DCT/ Walsh Transform/Wavelets) Image enhancement Operations- Spatial filtering, Edge detection- Canny, Sobel, Prewitt, Roberts Cross. Image segmentation- Region Growing, Thresholding. Adaptive Filter Implementation LMS Algorithm, Wiener Filter. FIR and IIR Filter design using TMS 320 DSK 22 MAESP 208 SEMINAR – II L T P C 0 0 2 1 Each student shall present a seminar on any topic of interest related to the core / elective courses offered in the second semester of the M. Tech. Programme. He / she shall select the topic based on the references from international journals of repute, preferably IEEE journals. They should get the paper approved by the Programme Co-ordinator / Faculty member in charge of the seminar and shall present it in the class. Every student shall participate in the seminar. The students should undertake a detailed study on the topic and submit a report at the end of the semester. Marks will be awarded based on the topic, presentation, participation in the seminar and the report submitted. 23