spectral estimation modern

Upload: hayder-mazin

Post on 07-Apr-2018

231 views

Category:

Documents


0 download

TRANSCRIPT

  • 8/3/2019 Spectral Estimation Modern

    1/43

    Lecture 5

    Spectrum Estimation

  • 8/3/2019 Spectral Estimation Modern

    2/43

    Periodogram Based

    .Periodogram

    .Modified Per.

    .Bartlett

    .Welch

    Classical Methods(Fourier Transform

    Based)

    ACF Based

    .Blackmen Tukey

    Filterbank Approach.Minimum Variance

    Nonclassical Methods

    (Non Fourier Based)

    Family of Non Parametric Methods

  • 8/3/2019 Spectral Estimation Modern

    3/43

  • 8/3/2019 Spectral Estimation Modern

    4/43

  • 8/3/2019 Spectral Estimation Modern

    5/43

    In practice we have one set of finite duration data.This approach has 2 practical problems

    1. Cant do the expected value2. Cant do the limit

    The periodogram is a method that ignores them both

    Based on the definition of PSD:

    In practice we compute it using the discrete Fourier transform (DFT)(possibly with zero padding) which computes the discrete time

    Fourier transform (DTFT) at discrete frequency points

    Periodogram Definition (recalled)

    2

    1( ) lim ( )

    2 1

    M

    jwnxM

    n M

    S w E x n eM

    21

    0

    1 ( ) ( )N

    jwnPER

    n

    S w x n eN

  • 8/3/2019 Spectral Estimation Modern

    6/43

    Periodogram-Viewed as a filterbank

    Although the Periodogram is Always implemented using

    the DFT, a filterbank interpretation is also very helpful.

    Defining the impulse response of an FIR filter as

    1, 0 >1 and ww1, then WR(w-w1)0

  • 8/3/2019 Spectral Estimation Modern

    30/43

    function Px = minvar(x,p)%MINVAR Spectrum estimation using the minimum variance method.% The spectrum of a process x is estimated using the minimum% variance method (sometimes called the maximum likelihood method).% x : Input sequence% p : Order of the minimum variance estimate - for short% sequences, p is typically about length(x)/3% The spectrum estimate is returned in Px using a dB scale.%x = x(:);

    R = covar(x,p);[v,d]=eig(R);U = diag(inv(abs(d)+eps));V = abs(fft(v,1024)).^2;Px = 10*log10(p)-10*log10(V*U);end;

    The Minimum Variance Spectral Estimation:MATLAB code

  • 8/3/2019 Spectral Estimation Modern

    31/43

    function R=covar(x,p)%COVAR Generates a covariance matrix/

    %USAGE R=covar(x,p)%% Generates a p x p covariance matrix for the sequence x.

    %---------------------------------------------------------------

    x = x(:);m = length(x);x = x - ones(m,1)*(sum(x)/m);R = convm(x,p)'*convm(x,p)/(m-1);end;

  • 8/3/2019 Spectral Estimation Modern

    32/43

  • 8/3/2019 Spectral Estimation Modern

    33/43

  • 8/3/2019 Spectral Estimation Modern

    34/43

  • 8/3/2019 Spectral Estimation Modern

    35/43

  • 8/3/2019 Spectral Estimation Modern

    36/43

  • 8/3/2019 Spectral Estimation Modern

    37/43

  • 8/3/2019 Spectral Estimation Modern

    38/43

  • 8/3/2019 Spectral Estimation Modern

    39/43

  • 8/3/2019 Spectral Estimation Modern

    40/43

  • 8/3/2019 Spectral Estimation Modern

    41/43

  • 8/3/2019 Spectral Estimation Modern

    42/43

  • 8/3/2019 Spectral Estimation Modern

    43/43