ECE 418 - Statistical Digital Signal Processing

Description: Credit 3U/4G Stochastic signal models, LMS identification, identification of signals from noise, Wiener filtering, blind separation of mixed signal, discrete Wavelet Transforms, compression and denoising, ceptral analysis.

Prerequisite: ECE 317 and ECE 341


1. Introduction to statistical SP

2. Basic concepts in stochastic processes

3. The Yule-Walker Equation

4. AR, MA ARMA signal models & their inter-relations

5. Stability and invertibility, Stochastic convergence
and efficiency

6. Ergodicity

7. The linear Signal Model

8. LMS identification of signal models

9. Nonstationary models

10. Order determination of models

11. The discrete Wiener filter and signal model

12. Blind identification of signal in measurement

13. The Box-Jenkins controller as a filter

14. Time Frequency Methods: Wavelet Transform (WT)

15. Multi-Resolution analysis

16. Discrete WT

17. Inverse WT

18. Debuchies WT in Matrix form

19. Denoising and compression by WT

20. 2D WT

21. Cepstral analysis

22. Cepstral filtering of noise

Copyright 2016 The Board of Trustees
of the University of Illinois.
Contact the webmaster