Scribd
Upload
7 views

Problem Set5

The document outlines three problems related to detection theory. Problem 1 involves observations that are sums of normal and Bernoulli random variables and testing hypotheses about their mean. Problem 2 involves observations that are sums of normal and signal terms and testing hypotheses about their variance. Problem 3 specifies a signal detection problem and shows the Neyman-Pearson critical region is of a certain form.

Uploaded by

Aditya S Sadhu ee20b003
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
7 views

Problem Set5

The document outlines three problems related to detection theory. Problem 1 involves observations that are sums of normal and Bernoulli random variables and testing hypotheses about their mean. Problem 2 involves observations that are sums of normal and signal terms and testing hypotheses about their variance. Problem 3 specifies a signal detection problem and shows the Neyman-Pearson critical region is of a certain form.

Uploaded by

Aditya S Sadhu ee20b003
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

EE5112: Detection Theory

Problem Set 5

1. (Poor III.F.8) Suppose we have observations Yk = Nk + θSk , k = 1, . . . , n, where N ∼


N (0, I) and where S1 , . . . , Sn are i.i.d. random variables, independent of N , and each
taking on the values +1 and −1 with equal probabilities of 1/2.
(a) Find the likelihood ratio for testing H0 : θ = 0 versus H1 : θ = A, where A is a
known constant.
(b) For the case n = 1, find the Neyman-Pearson rule and corresponding detection
probability for false-alarm probability α ∈ (0, 1), for the hypotheses of (a).
(c) Is there a UMP test of H0 : θ = 0 versus H1 : θ 6= 0 in this model? If so, why and
what is it? If not, why not? Consider the cases n = 1 and n > 1 separately.
2. (Poor III.F.13) Consider the model
Yk = θ1/2 sk Rk + Nk , k = 1, . . . , n,
where s1 , s2 , . . . , sn is a known signal sequence, θ ≥ 0 is a constant, and R1 , R2 , . . . , Rn ,
N1 , N2 , . . . , Nn are i.i.d. N (0, 1) random variables.
(a) Consider the hypothesis pair H0 : θ = 0 versus H1 : θ = A, where A is a known
positive constant. Describe the structure of the Neyman-Pearson detector.
(b) Consider now the hypothesis pair H0 : θ = 0 versus H1 : θ > 0. Under what
conditions on s1 , s2 , . . . , sn does a UMP test exist?
(c) For the hypothesis pair in part (b) with s1 , s2 , . . . , sn general, is there a locally
optimum detector? If so, find it. If not, describe the generalized likelihood ratio test.
3. (Poor III.F.14 (a)) Consider the following hypothesis about a sequence Y1 , Y2 , . . . , Yn , of
real observations:
H0 : Yk = Nk , k = 1, . . . , n,
versus
H1 : Yk = Nk + Θsk , k = 1, 2, . . . , n,
where N1 , N2 , . . . , Nn is a sequence of i.i.d. N (0, σ 2 ) random variables; where s1 , s2 , . . . , sn
is a known signal sequence satisfying sT s = 1; and where Θ is a N (µ, v 2 ) random variable,
independent of N1 , N2 , . . . , Nn . Show that the critical region for Neyman-Pearson testing
between these two hypotheses is of the form
v2 T 2
 
T 0
Γ1 = µs y + 2 |s y| > τ ,
2σ
where τ 0 is an appropriately chosen threshold. [Hint: The covariance matrix of Y equals
σ 2 I + v 2 ssT under hypothesis H1 .]

You might also like