This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: ECE 806, Detection and Estimation Theory Apr. 6, 2009 OSU, Spring 2009 Due: Apr. 13, 2009 Problem Set 1 This problem set has two parts. The first part is designed to give you a feel for the hypothesis testing problem through simulations. The second set of problems are more classical, mathematical type problems. They will enable you to go through the details of designing optimal Bayesian decision rules for some interesting pdfs. Peculiar pdfs are considered so that you will have exposure to some of the tricky aspects of decision rule design (non-contiguous decision regions for example). Problem 1 - MATLAB Exercise Your task is to generate a binary decision rule for some sonar data that has been collected. You are trying to classify underwater objects from the Titanic. You are going to design the initial system which must simply distinguish between metal objects ( H 1 ) and wooden objects ( H ). At your disposal are several data files: observations for H 1 only, observations for H only and a test file containing observations from both...
View Full Document
- Spring '09
- Bayesian probability, Prior probability, decision rules, homework web page, uniform costs