Random Signal and Data Analysis. 3 credits, 3 contact hours

Catalog description

Engineers and scientists are expected to understand and mitigate inherent uncertainties, noise, incomplete information and unpredictable fluctuations that are present in a wide range of systems, applications and data. This course builds a rigorous yet practical foundation for modeling and analyzing such systems, signals and data, to extract important information from noise-corrupted data and signals, design robust systems and improve reliability in the presence of uncertainty. Covered topics include probabilistic computer simulation of random phenomena, fundamentals of single and multiple random variables and data vectors and their key characteristics, transformations of random variables, fundamentals of random signals and time-series data, correlation and spectral analysis, multiple random signals and multi-sensor data, systems responses to random inputs, prediction of random signals and data, and optimal Wiener noise filtering and data smoothing. Examples of engineering applications are discussed as well. Computer implementation and simulation are key components of the course and are integrated into the course materials. Hands-on in-class computer sessions allow students to understand the materials by turning abstract concepts into tangible experiences. By bridging the gap between theory and application, interactive computer sessions enable students to see how theoretical concepts and methods work in practice, to solve real-world problems. The sessions include downloading or collecting real-world data and signals in class to model and analyze.