Study of estimation, detection, and identification methods. Detection: Binary and M-ary hypothesis testing. Neyman-Pearson detection. Invariance. Matched filter, CFAR matched filter and variants. Bayes detection. Likelihood ratios. Estimation: Maximum-likelihood estimation. Bayes estimation. Sufficiency and invariance. Cramer-Rao bounds. Estimation with the linear statistical model. Minimum mean square error. Recursive estimation. Kalman-Bucy filter. Identification: ML identification of ARMA models, signal subspaces, parameters in sinusoidal models and Machine learning methods. Topics may vary.
Prerequisites: 18-751 and senior or graduate standing
This course is currently being offered.