Digital Signal Processing and Statistical Inference
This course introduces discrete-time signal and system models in deterministic and stochastic settings and develops signal processing and statistical inference methodologies for real-time sensing and control applications. The course is intended for upper-level undergraduate and beginning graduate engineering students in engineering departments.The course covers both deterministic and stochastic techniques. Specific topics include time and frequency domain representation of signals and systems, state-space representation, feedback, stability, linear and nonlinear filtering, signal and state estimation and tracking, hypothesis testing, and signal detection. Applications in communications and control system design are integrated into the course material.
Topics
- Be able to obtain impulse response from frequency and state-space models and vice versa. Be able to analyze system stability, reachability, and observability given a linear time-invariant state space model.
- Be able to design and implement state and observer-based feedback systems that stabilize an unstable system.
- Be able to understand stationary and wide-sense stationary models of discrete-time signal and the notion of power spectrum density of a wide-sense stationary process.
- Be able to solve signal estimation and detection problems under parametric and state-space models, including implementing Wiener and Kalman filtering techniques for estimation, and using matched filtering in signal detection.
Official roster learning outcomes
Prerequisites & course information
MATH 1920; MATH 2940; ECE 3100 or ENGRD 2700 or ECE 2720 or MATH 4710; ECE 3250.
Sources
- Course catalogue ↗2026-2027