ECE-S 490
[ ] Prof. Athina Petropulu,
Rm. 7-221, Tel. x2358, e-mail: athina@cbis.ece.drexel.edu
TEACHING ASSISTANT
[ ]
Binnig Chen, Rm. 7-707, Tel. x2066, e-mail: chen@iason.ece.drexel.edu
CLASS SCHEDULE:
[ ] Lectures: MO,FRI 2:00-2:50 RM Curtis 457, Recitation: WED 2-2:50: RM 7-403
[ ] Lab: WED 3-5: RM 7-403
TEXT
[ ] Class Notes
[ ]
Digital Signal Signal Processing: Principles, Algorithms, and Applications,
third edition, J.G. Proakis and D.G. Manolakis, Prentice Hall, 1996.
GRADING POLICY
[] Projects: 80%
Quizzes: 20 %
Projects will be assigned according to the schedule given in Course Description. They will be based on material covered in class and lab/recitation sessions, and could be implemented in the Multimedia Signal Processing Laboratory (Rm 7-403).
There will be a bi-weekly quiz to be held every other Friday at 2:40 - 2:50. No make-ups will be given. Missed quiz, without prior formal excuse, will count as zero.
COURSE DESCRIPTION
[]The representation of a stationary random process by its spectrum can be an efficient and revealing description of the process. Spectral analysis is used to study periodicities in the data, and is quite powerful in signal processing tasks such as data modeling, forecasting, system identification and signal detection.
The course covers state-of-the-art, as well as conventional
power spectrum estimation techniques. Projects drawn from real-world
applications are used to reenforce the theoretical concepts.
The objective is to provide the students
with theoretical as well as working knowledge necessary to evaluate and
use existing DSP techniques and tools as well as competence to develop
novel discrete-time signal processing methods.
TENTATIVE COURSE OUTLINE
Basics on Random Processes (Week 1)
Stationary processes; Autocorrelation; Power spectrum; Estimation of
autocorrelation and power spectrum from finite length data; Frequency Resolution.
Spectral Analysis (Weeks 2-3 )
Periodogram, Welch spectrum, Blackman-Tukey.
Project #1: Determination of the type of an aircraft based of radar returns (to be assigned during the third Week)
Harmonic Decomposition methods (Weeks 4-6 ) Prony's method; Identification of exponentials in noise; Eigenanalysis based frequency estimators (MUSIC, Pisarenko).
Project #2: Estimation of power-system disturbances from real voltage measurements (to be assigned during the 6-th Week)
Parametric Modeling of Time Series (Weeks 7-10 )
Linear Prediction; Levinson
Recursion; Spectral Estimation (MMSE, MLE,...); Wold Decomposition
Project #3: Reconstruction of a speech phrase with missing segments (to be assigned during the 8-th Week).


