This is a graduate-level course at IIT-Kharagpur (with a strong emphasis on problem-solving) on the fundamental concepts of statistical estimation and hypothesis testing encountered in numerous signal-processing applications. There is no specific textbook that will be followed throughout the course, but most of the materials covered can be found in the following key references:
An introduction to signal detection and estimation (2nd edition), H. Vincent Poor.
Fundamentals of statistical signal processing, Steven M. Kay. This book has two volumes: one on estimation theory and the other on detection theory.
Many of the problems in the problem sets below are also taken from the exercises and solved examples in the textbooks above.
Fall 2024
Grading: Quiz: 25%, mid-sem: 25%, scribing and class performance: 10%, final exam: 40%
Problem sets
Fall 2023
Grading: Quiz and project: 20%, mid-sem: 30%, final exam: 50%
Syllabus: Module-1: Review of probability; Module-2: Bayesian estimation: MMSE, MMAE, and MAP estimators, conjugate prior, linear MMSE estimation, linear observation model and Gauss-Markov theorem, Wiener-Kolmogorov filtering, Lenvionson algorithm, Kalman filters; Module-3: Non-random parameter estimation: UMVUE, sufficient statistic, factorization theorem, completeness, Rao-Blackwell theorem, information bound, ML estimators and their theoretical properties (consistency and asymptotic normality), best linear unbiased estimators (BLUEs); Module-4: Bayes hypothesis testing, minimax test, Neyman Pearson lemma, composite hypothesis tests.
Problem sets