Scribd
Upload
77 views2 pages

EE 602-Assignment 3: 1 II II 1 I I TH

1. Prove that the diagonal elements of the inverse of the square root of a positive definite matrix I(θ) are always greater than or equal to the reciprocal of the corresponding diagonal elements of the original matrix. Equality is achieved when applying the Cauchy-Schwarz inequality. 2. Determine the Cramer-Rao lower bound (CRLB) for the slope B and intercept A parameters of a linear regression model with observations modeled as x[n] = A + Bn + w[n], where w[n] is white Gaussian noise. 3. Find the CRLB for the correlation coefficient ρ of independent bivariate Gaussian samples distributed as x[n] ~ N(0

Uploaded by

Kushagra Singhal
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
77 views2 pages

EE 602-Assignment 3: 1 II II 1 I I TH

1. Prove that the diagonal elements of the inverse of the square root of a positive definite matrix I(θ) are always greater than or equal to the reciprocal of the corresponding diagonal elements of the original matrix. Equality is achieved when applying the Cauchy-Schwarz inequality. 2. Determine the Cramer-Rao lower bound (CRLB) for the slope B and intercept A parameters of a linear regression model with observations modeled as x[n] = A + Bn + w[n], where w[n] is white Gaussian noise. 3. Find the CRLB for the correlation coefficient ρ of independent bivariate Gaussian samples distributed as x[n] ~ N(0

Uploaded by

Kushagra Singhal
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

EE 602-Assignment 3

1. Prove that

[ 1 ] I () ii

1 . [I ()]ii

Under what conditions will the new bound be achieved? Hint: Ap T ply Cauchy-Schwarz inequality to ei I () I1 ()ei , where ei is the vectors of all zeros except for a 1 as the ith element. The square root of a positive denite matrix A is dened to be the matrix with same eigenvectors as A but whose eigenvalues are the square roots of A. 2. Consider the problem of line tting or given the observations, x[n] = A + Bn + w[n], 0 n N 1,

where w[n] is WGN. Determine the CRLB for the slope B and the [ ]T intercept A. The parameter vector in this case is = A B . 3. Independent bivariate Gaussian samples {x [0] , x [1] , ........., x [N 1]} are observed. Each observation is a 2 1 vector which is distributed as x [n] N (0, C ) and [ ] 1 C= 1 Find the CRLB for the correlation coecient . 4. Let X1 , X2 , ...., Xm be a random sample, where each Xi has an exponential density with parameter , 1 fX (x) = ex/ (a) Determine ML Estimate of . 1 , x 0.

M L ]. (b) Determine E [ (c) Find the Fisher Information matrix, Is the estimate minimum variance ? Ecient? 5. Show that for a Rp+1 the minimum of a T Qa subject to the constraint that ap+1 = 1 is obtained when, a= 1 e
T 1 p+1 Q

Q1 e
p+1

p+1 ,

where ep+1 is the unit vector with 1 in element P + 1.

You might also like