02 001 the ml decisions are indicated by the

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: binomial coefficients cancel, so Λ(k ) = 2k 3 1 5−k 3 15 2 2 3 k =2 5 ≈ 2k . 7.6 Therefore, the ML decision rule is to declare H1 whenever Λ(X ) ≥ 1, or equivalently, X ≥ 3. For the ML rule, 1 2 pfalse alarm = 10 5 +5 5 1 2 + 1 2 5 15 2 14 22 +5 + 10 3 3 3 3 = (0.2)pfalse alarm + (0.8)pmiss ≈ 0.26. pmiss = pe = 0.5 1 3 3 = 51 ≈ 0.201 243 The MAP decision rule is to declare H1 whenever Λ(X ) ≥ 0.25, or equivalently, X ≥ 1. That is, the MAP rule declares H0 only if X = 0. For the MAP rule, pfalse alarm = 1 − 1 2 5 ≈ 0.97 15 1 = ≈ 0.041 3 243 = (0.2)pfalse alarm + (0.8)pmiss ≈ 0.227. pmiss = pe As expected, pe for the MAP rule (designed with the correct prior probabilities in mind) is smaller than pe for the ML rule. 2.11. BINARY HYPOTHESIS TESTING WITH DISCRETE-TYPE OBSERVATIONS 59 Example 2.11.2 (Detection problem with Poisson distributed observations) A certain deep space transmitter uses on-off modulation of a laser to send a bit, with value either zero or one. If the bit is zero, the number of photons, X, arriving at the re...
View Full Document

Ask a homework question - tutors are online