EFFECT FROM THE ACADEMIC YEAR 2013 - 2014

EC 474

SPECTRAL ESTIMATION TECHNIQUES

                                            (ELECTIVE –III)

Instruction                                                                                                                                                                                        4  Periods per week

Duration of University Examination                                                                                                                                                   3  Hours

University Examination                                                                                                                                                                     75 Marks

Sessional                                                                                                                                                                                         25 Marks

UNIT-I

Random variable, Random processes, stationary random processes, statistical average, statistical averages for joint random processes, Discrete-Time Random signals, Time averages for a Discrete Time Random processes, Mean-Ergodic Process, Correlation Ergodic Processes, Power density Spectrum, Representation of a Stationary Random Processes, Rational power spectra, Relation between the filter parameters and autocorrelation.

UNIT-II

Forward and Backward linear prediction-Forward and Backward linear prediction, Relationship of an AR process to linear prediction, Solution of linear equations- The Levinson- Durbin algorithm, Wiener Filters- Wiener filters for Filtering and Prediction, FIR Wiener filter, Orthogonality Principle in linear Mean square Estimation, IIR Weiner Filter, Noncausal Weiner filter.

UNIT-III

Power Spectrum Estimation: Estimation of Spectra from finite duration observation of a signal. Periodogram. DFT in power spectrum estimation. Non-parametric methods – Bartlett’s, Welch’s and Blackman-Tukey methods, Performance Characteristics of Nonparametric Power Spectrum Estimators, Computational requirements and performance characteristics.

UNIT-IV

Parametric methods – Relation between auto correlation sequence and model parameters. Methods for AR model parameters. Yule – Walker method, Burg method, unconstrained Least squares methods. Sequential estimation methods. Selection of AR model order, Moving average (MA) and ARMA models for Power spectrum estimation.

UNIT-V

Eigen Analysis algorithms for Spectrum estimation- Pisarenko’s harmonic decomposition method. Eigen structure methods – Music and ESPIRIT. Order selection criteria. Filter Bank methods- Filter bank realization of the periodogram, Capon’s minimum variance method.

Suggested Reading:

1.          John G. Proakis and Dimitris G. Manolakis, “Digital Signal Processing-Principles, Algorithms and Applications,” 4/e, Pearson/PHI, 2007.

2.          D.G. Manolakis, Ingle and S.M. Kogon, “Statistical and Adaptive Signal Processing,” McGraw Hill, 2000.

3.          John G. Proakis, Rader, et al, “Algorithms for Statistical Signal Processing,” Pearson Education, Asia Publishers, 2002.

4.          Emmanuel Ifeachor and Barrie W. Jervis, “Digital Signal Processing - A Practical Approach,” Pearson, 2004.