Aug 18, 2022  
2020-2021 Undergraduate Academic Catalog 
2020-2021 Undergraduate Academic Catalog [ARCHIVED CATALOG]

Add to Portfolio (opens a new window)

BE 4340 - Advanced Topics in Biomedical Digital Signal Processing

3 lecture hours 0 lab hours 3 credits
Course Description
The objective of this course is to introduce the students to the advanced topics and methodologies of digital signal processing and to have students apply these methodologies to the analysis of biological signals such as ECG, EEG, local field potentials, and phonocardiogram signals. Topics covered include Welch Periodogram power spectral estimation, cross-spectral estimation and coherence, introduction to time-frequency analysis, and short-segment Fast Fourier Transform. (prereq: BE 4800 or EE 3220 or equivalent with permission of instructor)
Course Learning Outcomes
Upon successful completion of this course, the student will be able to:
  • Possess advanced skills in digital signal processing necessary to function as a successful biomedical engineer, whose role involves quantitative analysis of biological signals
  • Implement using computer tools specific advanced signal processing methodologies that are commonly used for extraction of clinically relevant information from biological signals
  • Understand the challenges of digital signal processing when applied to the analysis of various biological signals

Prerequisites by Topic
  • Continuous and discrete signals and systems concepts
  • Continuous and discrete Fourier Transform and Series
  • Sampling, aliasing, and spectral replication
  • Fast Fourier Transform
  • A/D conversion and quantization concepts
  • Digital filters (FIR and IIR) and digital filter design fundamentals
  • Z-Transforms and Unit Circle concepts
  • Statistical analysis
  • Proficiency in MATLAB programming

Course Topics
  • Review of fundamental digital signal processing concepts such as sampling, aliasing, and Discrete Fourier Transform and Series (2 classes)
  • Overview of random (single) processes, stationarity, ergodicity and autocorrelation concepts (4 classes)
  • Standard methodologies of power spectral estimation such as Welch Periodogram and Blackman Tukey (4 classes)
  • Introduction of joint random processes, covariance and cross-correlation measures (4 classes)
  • Cross-spectral estimation and coherence (4 classes)
  • Introduction to time-frequency analysis. Short-Segment Fast Fourier Transform (4 classes)
  • Bi-weekly quizzes (2 classes total)

Laboratory Topics
  • No laboratory. However, the students will be responsible for take-home computer projects involving the implementation of various signal processing methodologies and their application to the analysis of specific biological signals. The students will be required to submit a formal report

Dr. Olga Imas

Add to Portfolio (opens a new window)