0% found this document useful (0 votes)
29 views2 pages

Both Unknown. Find Maximum Likelihood (ML) : Detection and Estimation, Fall 2015 Problem Set 1 (Due October 18, 2015)

This document contains 7 problems related to detection and estimation. Problem 1 asks to find maximum likelihood estimates for the mean and variance of a Gaussian distribution and examine if they are unbiased and consistent. Problem 2 asks to find a sufficient statistic for estimating an unknown parameter θ based on observations that are a linear function of θ plus noise. Problem 3 asks to find the maximum likelihood estimate of an unknown parameter θ given an observation from a uniform distribution. Problem 4 asks to show that there exist no unbiased estimators for an unknown parameter x given an observation from a specified distribution. The remaining problems 5-7 ask to find maximum likelihood estimates for unknown parameters based on various observation models involving noise.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
29 views2 pages

Both Unknown. Find Maximum Likelihood (ML) : Detection and Estimation, Fall 2015 Problem Set 1 (Due October 18, 2015)

This document contains 7 problems related to detection and estimation. Problem 1 asks to find maximum likelihood estimates for the mean and variance of a Gaussian distribution and examine if they are unbiased and consistent. Problem 2 asks to find a sufficient statistic for estimating an unknown parameter θ based on observations that are a linear function of θ plus noise. Problem 3 asks to find the maximum likelihood estimate of an unknown parameter θ given an observation from a uniform distribution. Problem 4 asks to show that there exist no unbiased estimators for an unknown parameter x given an observation from a specified distribution. The remaining problems 5-7 ask to find maximum likelihood estimates for unknown parameters based on various observation models involving noise.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

Detection and Estimation, Fall 2015

Problem Set 1 (Due October 18, 2015)

1. Consider N i.i.d samples from a Gaussian distribution of mean


and variance

2 ,

both unknown. Find maximum likelihood (ML)

estimates for the mean and variance. Examine whether the estimators
are unbiased and consistent. If the estimator for the variance is biased,
suggest an unbiased estimator.

2. Consider an estimation problem in which the observations

{ Y n , n=1,2, } are real-valued random variables given by


Y n=x +V n
x and V n

where

are i.i.d.

N ( 0,1 ) . Find a sufficient statistic for

3. Consider an observation given by


Y =+ Z

where

is an unknown fixed parameter to be estimated and Z is

uniformly distributed over the interval


of

given

q q
,
2 2

. Find the ML estimate

Y=y .

4. Let

f ( yx ) = x ,if 0 y 1/ x
0, otherwise
for x > 0. Show that there exist no unbiased estimators

^x ( y )

for x.

Note that because


only x > 0 are possible values, an unbiased estimator need only be
unbiased for x > 0.
2
5. Consider a point x on a unit circle in R . Given N i.i.d observations

from

R2 of the form

y k =x + z k

where

1 k N , and

matrix

[ ]
1
1

with

zk R

is zero-mean Gaussian with covariance

||<1. Find the ML estimator for x.

6. Consider N i.i.d samples X k , where 1 k N , from a Bernoulli


distribution with parameter p, i.e.,
X k =0

with probability

X k =1

with probability p and

1 p . Find the ML estimate of p. Find the

Cramer-Rao lower bound on the variance of the estimator. Is the


estimator efficient?

7. Consider N samples from a noisy sinusoidal signal with known


frequency

and amplitude A

y k = A cos ( k T s + ) + z k
where

1 k N ,

2 , and

Ts

zk

is zero-mean Gaussian with variance equal to

is the sampling time such that

period of the sinusoidal signal. Parameter


the period. Angle
ML estimate for

Ts

NTs

is equal to one

is much smaller than

is an unknown phase to be estimated. Find the

. Sketch a block diagram of estimator operation.

You might also like