Digital Signal Processing
2nd Cycle Studies - Mestrado
Probability and Statistics; Signal Processing
Expository lessons with a detailed presentation of the concepts and solving of practical examples of application with demonstrations.
Laboratorial classes where the student will perform computational simulations of the presented topics.
The continuous evaluation consists of theoretical-practical tests (60%) and laboratorial assignments and practical laboratory work (40%).
It is intended that students acquire a strong training in statistical and deterministic processing of discrete-time signals. The skills obtained will allow developing modern applications and signaling processing algorithms in different areas of knowledge.
1. Multi-rate signal processing
Decimation, expansion, band-limited interpolation. Bank of filters.
2. Discrete-time stochastic processes
Definitions. Estimation of autocorrelation. Harmonic and white noise processes. Regular stochastic processes: AR, MA and ARMA.
3. Power spectrum estimation
Periodogram. Classical smoothing methods. Maximum Entropy method. Structural Methods.
4. Optimal filtering
Wiener filtering. Wiener-Hopf equations. Practical cases: echo canceling; noise canceling; linear prediction, noise reduction in time and in frequency.
Fernando Manuel dos Santos Perdigão
Laboratory work or Field work: 40.0%
S. Mitra, Digital Signal Processing: A Computer-Based Approach, McGraw-Hill, 3rd. ed., 2005.
Monson H. Hayes, Statistical Digital Signal Processing and Modeling, John Wiley, 1996.
Oppenheim and Schafer, Discrete-Time Signal Processing, 3rd edition, Pearson, 2010.