ECE 53800 - Digital Signal Processing I

Course Details

Lecture Hours: 3 Credits: 3

Areas of Specialization:

  • Communications, Networking, Signal & Image Processing

Counts as:

  • EE Advanced Selective
  • CMPE Special Content Selective

Normally Offered:

Each Fall

Campus/Online:

On-campus and online

Requisites:

ECE 30100 and 30200

Catalog Description:

Theory and algorithms for processing of deterministic and stochastic signals. Topics include discrete signals, systems, and transforms, linear filtering, fast Fourier transform, nonlinear filtering, spectrum estimation, linear prediction, adaptive filtering, and array signal processing.

Course Objectives:

1) Provide the student with a broad, yet strong background in the traditional topics associated with processing of deterministic digital signals, e.g., discrete-time transforms, and linear filtering. 2) Provide student with a strong background in traditional topics associated with processing of stochastic signals, e.g., spectrum estimation and linear prediction. 3) Introduce the student to some of the more recent developments that promise to have a broad impact on digital signal processing, e.g., nonlinear filtering and adaptive filtering.

Required Text(s):

  1. Digital Signal Processing: Principles, Algorithms, and Applications , 4th Edition , J. G. Proakis and D. G. Manolakis , Prentice Hall , 2007 , ISBN No. 0131873741

Recommended Text(s):

None.

Lecture Outline:

Lectures Topic
7 Discrete signals, systems, and transforms; A. Discrete linear system; B. Discrete-time Fourier transform (DTFT); C. 2-sided Z transform; D. Discrete Fourier transform (DFT)
5 Linear Filtering; A. Finite impulse response filters; 1. Windowed designs (Kaiser) ; 2. Equiripple design; B. Infinite impulse response filters ; 1. Bilinear Z transform; 2. Computer-aided techniques
3 Fast Fourier transform (FFT) algorithms; A. Decimation in time ; B. Decimation in frequency ; C. Chirp Z-Transform ; D. Sectioned convolution
3 Nonparametric methods of power spectrum estimation; A. Estimation of the autocorrelation sequency for random signals; B. Smoothing the periodogram: the Blackman-Turkey method
6 Model-based power spectrum estimation; A. Autoregressive (AR)spectral estimation; B. Lattice filter: Burg's method; C. Signal subspace methods; D. Applications
9 Adaptive signal processing; A. Applications; B. Least mean square (LMS) adaptive algorithm; C. Recursive least squares (RLS) lattice filters; D. Adaptive beamforming
9 Nonlinear filtering; A. Rank Order filters; B. Deterministic and statistical analysis of median filters; C. Threshold decomposition -- stock filters; D. Applications
3 Exams

Assessment Method:

Exams & projects. (3/2022)