Digital Signal Processing

Ingegneria Informatica Digital Signal Processing

0622700005
DIPARTIMENTO DI INGEGNERIA DELL'INFORMAZIONE ED ELETTRICA E MATEMATICA APPLICATA
EQF7
COMPUTER ENGINEERING
2018/2019



YEAR OF COURSE 1
YEAR OF DIDACTIC SYSTEM 2017
SECONDO SEMESTRE
CFUHOURSACTIVITY
432LESSONS
540LAB
Objectives
THE COURSE AIMS AT ILLUSTRATING THE PRINCIPAL MATHEMATICAL METHODS FOR THE ANALYSIS AND PROCESSING OF DIGITAL SIGNALS AND INCLUDES NUMERICAL AND LABORATORY EXERCISES.

KNOWLEDGE AND UNDERSTANDING
THE OBJECTIVE IS TO PRESENT THE FUNDAMENTALS OF
• SIGNAL REPRESENTATION,
• ANALYSIS AND DESIGN OF SYSTEMS DEFINED BY DIFFERENCE EQUATIONS,
• DIGITAL SIGNAL PROCESSING OF BOTH DETERMINISTIC AND RANDOM SIGNALS.


APPLYING KNOWLEDGE AND UNDERSTANDING
LABORATORY EXERCISES CONSIST IN THE APPLICATION OF SOFTWARE PACKAGES FOR ACQUIRING THE ABILITY TO IMPLEMENT PRACTICAL ALGORITHMS AIMED TO
• THE ANALYSIS AND COMPRESSION A SIGNAL VIA TRANSFORMS,
• THE DESIGN OF FIR AND IIR DIGITAL FILTERS,
• STATISTICAL AND ADAPTIVE SIGNAL FILTERING.

Prerequisites
FOR THE SUCCESSFUL ACHIEVEMENT OF THE COURSE GOALS, UNDERGRADUATE LEVEL MATHEMATICS, WITH PARTICULAR REFERENCE TO MATRIX CALCULUS AND PROBABILITY THEORY AND BASIC KNOWLEDGE OF SIGNAL PROCESSING ARE REQUIRED.
Contents
PRESENTATION OF THE MOTIVATION AND POTENTIAL OF NUMERICAL PROCESSING. (HOURS LECTURE/ PRACTICE / LABORATORY 2/-/-)
ANALOG AND DIGITAL SIGNALS. INTERPOLATION. QUANTIZATION. (HOURS 10/-/16). ONE-DIMENSIONAL AND TWO-DIMENSIONAL SIGNAL REPRESENTATION.
APPLICATIONS (HOURS 8/-/4).
SYSTEMS ANALYSIS IN THE FREQUENCY AND ZETA DOMAINS. SYSTEMS DEFINED BY DIFFERENCE EQUATIONS. FREQUENCY RESPONSE, AMPLITUDE AND PHASE RELATIONSHIP, ALL-PASS SYSTEMS, MINIMUM PHASE SYSTEMS. FILTER STRUCTURES. (HOURS 3/-/4)
DESIGN OF IIR FILTERS BY ANALOG FILTERS TRANSFORMATIONS. ANALOG AND DIGITAL FREQUENCY TRANSFORMATIONS. FIR FILTERS DESIGN WITH THE WINDOW METHOD. IIR AND FIR FILTERS COMPARISON. (HOURS 3/-/7)
THE STATISTICAL FILTERING. BAYESIAN APPROACH. WIENER FILTERS. LEAST-SQUARES. APPLICATIONS (HOURS 3/-/5)
ADAPTIVE FILTERING. GRADIENT METHOD AND STOCHASTIC GRADIENT. APPLICATIONS. (HOURS 3/-/4)
TOTAL HOURS (32/-/40)
Teaching Methods
LECTURES AND LABORATORY PRACTICE.
IN LABORATORY HOURS SOME APPLICATION EXAMPLES OF THE PRESENTED METHODS ARE ASSIGNED AND INTERACTIVELY COMPLETED IN THE MATLAB ENVIRONMENT.
Verification of learning
THE FINAL EXAM IS AIMED AT EVALUATING: THE KNOWLEDGE AND UNDERSTANDING OF THE CONCEPTS PRESENTED IN THE COURSE, THE ABILITY TO APPLY THAT KNOWLEDGE TO SOLVE PROBLEMS OF SIGNAL ANALYSIS AND DESIGN OF DIGITAL PROCESSORS FOR DETERMINISTIC AND RANDOM SIGNALS, INDEPENDENCE OF JUDGMENT, COMMUNICATION SKILLS AND THE LEARNING ABILITY.
THE EXAM CONSISTS OF A WRITTEN TEST, WHOSE PURPOSE IS TO ASSESS THE ABILITY TO APPLY KNOWLEDGE AND THE INDEPENDENCE OF JUDGMENT, AND AN ORAL INTERVIEW, FOR ASSESSING THE ACQUIRED KNOWLEDGE, THE UNDERSTANDING ABILITY, THE LEARNING SKILL, AND THE ORAL PRESENTATION.
THE WRITTEN TEST CONSISTS IN SOLVING, THROUGH A MATLAB PROGRAM, TYPICAL PROBLEMS PRESENTED IN THE COURSE INCLUDING: 1) EVALUATION OF DIGITAL SYSTEMS RESPONSE BY USING DFT, 2) IMAGE ANALYSIS AND PROCESSING IN THE TIME DOMAIN AND FREQUENCY, 3) DESIGN OF FIR AND IIR SYSTEMS WHICH MEET THE ASSIGNED SPECIFICATIONS; 4) STATISTICAL AND ADAPTIVE FILTER DESIGN FOR CLASSIFICATION, PREDICTION, AND DENOISING.
THE WRITTEN TEST WILL BE EVALUATED BY THE FOLLOWING SCALE: EXCELLENT (OTTIMO), GOOD (BUONO), FAIR (DISCRETE), SUFFICIENT (SUFFICIENT), BARELY SUFFICIENT (APPENA SUFFICIENTE), REJECTED (INSUFFICIENTE). REJECTED TEST MUST BE REPEATED. THE TEST WILL BE ASSESSED AS SUFFICIENT IF THE STUDENT PRESENTS THE SOLUTION OF AT LEAST HALF OF THE PROBLEMS PROPOSED WITHOUT CONCEPTUAL ERRORS. THE TEXT WILL BE EVALUATED AS EXCELLENT IF ALL THE PROBLEMS HAVE BEEN CORRECTLY SOLVED AND PRESENTED WITH A CLEAR EXPOSITION OF THE USED PROCESSES.
DURING THE LABORATORY HOURS, IN ADDITION TO THE NORMAL EXERCISES, SOME TESTS WILL BE (OPTIONALLY) PRESENTED TO THE STUDENTS WITH THE AIM TO ASSESS THE PRACTICAL SKILLS ACQUIRED THEREOF. THEY ARE EQUIVALENT TO THE WRITTEN TEST AND ARE THEREFORE SUBJECT TO A FINAL OVERALL EVALUATION WITH THE SAME SCALE.
THE ORAL EXAMINATION WILL COVER ALL THE TOPICS OF THE COURSE AND ASSESSMENT WILL TAKE INTO ACCOUNT THE KNOWLEDGE DEMONSTRATED BY THE STUDENT AND THE DEGREE OF ITS DEPTH, PROVEN ABILITY TO LEARN, THE QUALITY OF THE PRESENTATION; ORAL EXAMINATION EVENTUALLY INCLUDES THE DISCUSSION OF THE WRITTEN TEST.
IN THE FINAL EVALUATION, EXPRESSED IN THIRTIETHS, THE EVALUATION OF THE WRITTEN TEST AND THE RELATED ORAL DISCUSSION WILL ACCOUNT FOR 50%, WHILE THE INTERVIEW FOR 50%. THE CUM LAUDE MAY BE GIVEN TO STUDENTS WHO DEMONSTRATE THAT THEY CAN AUTONOMOUSLY APPLY THE ACQUIRED KNOWLEDGE TO DIVERSE PROBLEMS.
Texts
A.V. OPPENHEIM, R.W. SCHAFER, DISCRETE-TIME SIGNAL PROCESSING, 2ND ED. PRENTICE HALL, 1999.
V.K. INGLE, J.G. PROAKIS, DIGITAL SIGNAL PROCESSING USING MATLAB, BROOKS-COLE, 2000.
S. HAYKIN, ADAPTIVE FILTER THEORY, 4TH EDITION, PRENTICE HALL.
More Information
TEACHING WILL BE PROVIDED BY THE LECTURER. LABORATORY EXERCISES WILL BE CARRIED OUT INTERACTIVELY. TEACHING IN ITALIAN.
  BETA VERSION Data source ESSE3 [Ultima Sincronizzazione: 2019-10-21]