Basic Information
- Course Code: EE 638
- Course Name: Estimation and Identification
- Course Offered In: 2022-23
- Semester Season: Autumn
- Instructors: Prof. Debraj Chakraborty
- Prerequisites: Familiarity with Probability theory, Stochastic processes, Linear Systems and Linear Algebra
- Difficulty (1 being easy and 5 being tough): 4
Course Content
The content is changed nearly every year. In the previous offering, these were covered:
-
Parameter Estimation: a. Minimum Variance Unbiased Estimation b. Cramer-Rao Lower Bound c. Sufficient Statistics d. Best Linear Unbiased Estimators e. Maximum Likelihood Estimation
-
Signal Estimation: f. Least Squares Estimation: Deterministic and Stochastic g. Finite and Infinite Horizon Weiner-Hopf Filter: The Innovations Process, Canonical Spectral Factorization h. The Kalman Filter: State Space Models, Various forms, EKF i. Case Studies: Sensor Fusion using Kalman Filter, Orientation Estimation from IMU data
-
Identification: Nonparametric and parametric methods for LTI systems, Introduction to Subspace methods
Feedback on Lectures
The lectures were good as the professor provided frequent examples to improve the understanding of the content. There were no tutorial sessions as such. The lectures were interactive in nature and professor used chalk-board and provided handwritten notes. The professor was very approachable for the content. Majority of the course focused on the estimation part and identification was covered very briefly; however, the content of the course is variable in nature.
Feedback on Evaluations
Assignments (20%), Mid-Sem (30%), End-Sem (40%), Term Project (10%).
Assignments were lengthy and required thought. They were problems given in the exercises of the references book. The difficulty of assignments was more than that of the exams. Exams focused on basic concepts and were based on assignments. The project involved implementing an extended Kalman filter for estimation orientation on a given sensor dataset. Only the project had programming, rest all assignments were mathematical in nature. The project had to be done individually.
Study Material and Resources
- Steven M. Kay, Fundamentals of Statistical Signal Processing: Estimation, Prentice-Hall, 1993.
- G. Casella and R. Berger, Statistical Inference, Duxbury Thomson Learning, 2002.
- T. Kailath, A.H. Sayed and B. Hassibi, Linear Estimation, Prentice- Hall, 2000.
- L. Ljung, System Identification-Theory for the User, Prentice-Hall, 1999.
- P.V. Overschee, B.D. Moore, Subspace Identification for Linear Systems, Kluwer Academic, 1996
You can find my notes on: https://dokania-tanmay.github.io/assets/pdf/ee638.pdf or https://dokania-tanmay.github.io/
Follow-up Courses
If one wants to explore estimation further, a good course is SC651: Estimation on Lie groups.
Final Takeaway
Good course, heavy in nature. Missing lectures will not be a good idea as the content is too much to cover in a night.