
25162: Spectrum Estimation
Course Name: Spectrum Estimation
Course Number: 25162
Prerequisite(s): 25155 (Digital Signals Processing 1)
Co-requisite(s): 25181 (Random Process)
Units: 3
Level: Postgraduate
Last Revision: Fall 2012
Description:
Syllabus:
References:
Course Number: 25162
Prerequisite(s): 25155 (Digital Signals Processing 1)
Co-requisite(s): 25181 (Random Process)
Units: 3
Level: Postgraduate
Last Revision: Fall 2012
Description:
This course examines theoretical principles of estimation theory, traditional spectral estimation methods, modern spectral estimation methods, AR, MR, and ARMA models in spectral estimation, gradient algorithm, Pisarenko method, Prony method, Maximum Likelihood method, subspace-based methods, applications of spectral estimation, and so forth.
Syllabus:
- Principles of Estimation, Optimal Estimation Criterions
- Autocorrelation Sequence Estimation, Biased Estimation, Unbiased Estimation
- Traditional Spectral Estimation Methods Based on Fourier Transform, Periodogram, The Limitations of Traditional Spectral Estimation Methods
- Modern Spectral Estimation: Modeling and Parameter Identification Approach
- AR, MR, and ARMA Models in Spectral Estimation, Optimal and Suboptimal Methods
- Gradient Algorithm and step-by-step strategy for the Estimation of Parameters
- Model Order and Criterions for Selection of Model Order
- Pisarenko Method
- Prony’s Method (Continuous and Discrete Spectrum)
- Maximum Likelihood Method
- Sub-Space Methods:
- Multiple Signal Classification (MUSIC, SPIRIT)
- Pisarenko as a special case of Sub-Space
- Applications of Spectral Estimation
- A Comparison of Spectral Estimation Methods
References:
- S. M. Kay, Modern Spectral Estimation, Prentice Hall, 1988
- P. Stoica, R. Mouse, Introduction to Spectral Analysis, Prentice Hall, 1997
- S. M. Kay, S.L. Marple, Spectrum Analysis, A Modern Perspective, Proc. of IEEE, 1981
Last Update: 2024-07-08