Statistical Signal Processing
- Department of Radioelectronics
The course provides fundamentals in three main domains of the statistical signal processing: 1) estimation theory, 2) detection theory, 3) optimal and adaptive filtering. The statistical signal processing is a core theory with many applications ranging from digital communications, audio and video processing, radar and radio navigation, measurement and experiment evaluation, etc.
- Syllabus of lectures:
1a. MVU estimator, Cramer-Rao lower bound, composite hypothesis, performance criteria
1b. Sufficient statistics
1c. Maximum Likelihood estimator, EM algorithm
1d. Bayesian estimators (MMSE, MAP)
2a. Hypothesis testing (binary, multiple, composite)
2b. Deterministic signals
2c. Random signals
3. Optimal and adaptive Filtration
3a. Signal modeling (ARMA, Padé approximation, ...)
3b. Toeplitz equation, Levinson-Durbin recursion
3c. MMSE filters, Wiener filter.
3d. Kalman filter.
3e. Least Squares, RLS
3f. Steepest descent and stochastic gradient algorithms.
3g. Spectrum estimation
- Syllabus of tutorials:
- Study Objective:
- Study materials:
1. Steven Kay: Fundamentals of Statistical Signal Processing - Estimation theory
2. Steven Kay: Fundamentals of Statistical Signal Processing - Detection theory
3. Monson Hayes: Statistical digital signal processing and modeling
4. Ali Sayed: Fundamentals of Adaptive Filtering
5. S. M. Kay: Fundamentals of statistical signal processing-detection theory, Prentice-Hall 1998
- Further information:
- No time-table has been prepared for this course
- The course is a part of the following study plans: