Stochastic Processes, Detection, and Estimation
The goal of this course was to develop analytical tools for the modeling and analysis of random phenomena and the application of these tools to a range of problems arising in engineering, manufacturing, and operations research. Emphasis was placed on hypothesis testing, confidence intervals, and nonparametric methods.
​
Course topics included:
-
MMSE/MAP/MLE Estimation
-
Cramer-Rao Bound
-
Vector Estimation
-
Bayes Linear Least Squares Estimation
-
Binary and M-Ary Hypothesis Testing
-
Asymptotic Analysis of Likelihood Ratio Tests
-
Composite Hypothesis Testing
-
Random Processes (RPs)
-
Stationarity/Moment Analysis of RPs
-
IID Processes
-
Markov Processes
-
Discrete Time Markov Chains
-
Bernoulli/Poisson/Wiener Processes
-
LSI Systems/WSS Processes
-
Power Spectral Density
-
Cross-Spectral Density & Optimal Estimation
-
(Non)Causal Wiener Filter
-
Discrete Wiener Filters
Stochastically Modeling MBE
Beyond examinations and problem sets, a final paper was also required for this course. I elected to write mine on stochastically modeling molecular beam epitaxy (MBE). A short video from my simulations can be seen below. It shows the simulated growth of one layer of a crystal lattice using MBE, where the black dots represent individual atoms. Islands can be seen nucleating.
Below this, the full paper is available for reading.