diff options
Diffstat (limited to 'help/en_US/pmusic.xml')
| -rw-r--r-- | help/en_US/pmusic.xml | 166 |
1 files changed, 38 insertions, 128 deletions
diff --git a/help/en_US/pmusic.xml b/help/en_US/pmusic.xml index b445409..107a1d2 100644 --- a/help/en_US/pmusic.xml +++ b/help/en_US/pmusic.xml @@ -17,7 +17,7 @@ <refnamediv> <refname>pmusic</refname> - <refpurpose>Computes Psuedospectrum using MUSIC algorithm</refpurpose> + <refpurpose>Psuedospectrum using MUSIC algorithm</refpurpose> </refnamediv> @@ -33,144 +33,54 @@ [S,f] = pmusic(x,p,nfft,fs,nwin,noverlap) [...] = pmusic(...,freqrange) [...,v,e] = pmusic(...) - - - - - - </synopsis> -</refsynopsisdiv> - - -<refsection> - <title>Parameters</title> - <variablelist> - <varlistentry><term>x:</term> - <listitem><para> int|double - vector|matrix</para> -<para>Input signal. In case of a matrix, each row of x represents a + + Parameters: + x - int|double - vector|matrix + Input signal. In case of a matrix, each row of x represents a seperate observation of the signal. If 'corr' flag is specified, then x is the correlation matrix. If w is not specified in the input, it is determined by the algorithm. If x is real valued, then range of w is [0, pi]. - Otherwise, the range of w is [0, 2pi)</para> -</listitem></varlistentry> - </variablelist> - -<variablelist> - <varlistentry><term>p:</term> - <listitem><para> int|double - scalar|vector</para> -<para>p(1) is the dimension of the signal subspace + Otherwise, the range of w is [0, 2pi) + p - int|double - scalar|vector + p(1) is the dimension of the signal subspace p(2), if specified, represents a threshold that is multiplied by - the smallest estimated eigenvalue of the signal's correlation matrix.</para> - -</listitem></varlistentry> - </variablelist> - -<variablelist> -<varlistentry><term>w:</term> - <listitem><para> int|double - vector</para> -<para>w is the vector of normalized frequencies over which the - pseuspectrogram is to be computed.</para> - -</listitem></varlistentry> - -</variablelist> - - -<variablelist> -<varlistentry><term>nfft:</term> - <listitem><para> int - scalar (Default = 256)</para> -<para>Length of the fft used to compute pseudospectrum. The length of S + the smallest estimated eigenvalue of the signal's correlation matrix. + w - int|double - vector + w is the vector of normalized frequencies over which the + pseuspectrogram is to be computed. + nfft - int - scalar (Default = 256) + Length of the fft used to compute pseudospectrum. The length of S (and hence w/f) depends on the type of values in x and nfft. If x is real, length of s is (nfft/2 + 1) {Range of w = [0, pi]} if nfft is even and (nfft+1)/2 {Range of w = [0, pi)} otherwise. - If x is complex, length of s is nfft.</para> - -</listitem></varlistentry> - -</variablelist> - -<variablelist> -<varlistentry><term>fs:</term> - <listitem><para> int|double - scalar (Default = 1)</para> -<para>Sampling rate. Used to convert the normalized frequencies (w) to - actual values (f) and vice-versa.</para> - -</listitem></varlistentry> - -</variablelist> - -<variablelist> -<varlistentry><term>nwin:</term> - <listitem><para> int|double - scalar (int only)|vector (Default = 2*p(1))</para> -<para> If nwin is scalar, it is the length of the rectangular window. + If x is complex, length of s is nfft. + fs - int|double - scalar (Default = 1) + Sampling rate. Used to convert the normalized frequencies (w) to + actual values (f) and vice-versa. + nwin - int|double - scalar (int only)|vector (Default = 2*p(1)) + If nwin is scalar, it is the length of the rectangular window. Otherwise, the vector input is considered as the window coefficients. - Not used if 'corr' flag present.</para> - -</listitem></varlistentry> - -</variablelist> -<variablelist> -<varlistentry><term>noverlap:</term> - <listitem><para> int - scalar (Default = nwin-1)</para> -<para> number of points by which successive windows overlap. noverlap not - used if x is a matrix</para> - -</listitem></varlistentry> - -</variablelist> -<variablelist> -<varlistentry><term>freqrange:</term> - <listitem><para> string</para> -<para>The range of frequencies over which the pseudospetrogram is + Not used if 'corr' flag present. + If x is a vector, windowing not done in nwin in scalar. If x is a + matrix, + noverlap - int - scalar (Default = nwin-1) + number of points by which successive windows overlap. noverlap not + used if x is a matrix + freqrange - string + The range of frequencies over which the pseudospetrogram is computed. Three possible values - 'onesided', 'twosided', 'centered' 'corr' flag Presence indicates that the primary input x is actually a correlation matrix - </para> - -</listitem></varlistentry> - -</variablelist> -</refsection> - -<refsection> - <title>Description</title> -<para>[S,w] = pmusic(x,p) implements the MUSIC (Multiple Signal Classification) algorithm and returns S, the pseudospectrum estimate of the input signal x, and a vector w of normalized frequencies (in rad/sample) at which the pseudospectrum is evaluated. The pseudospectrum is calculated using estimates of the eigenvectors of a correlation matrix associated with the input data x, where x is specified as either:</para> - - <para> A row or column vector representing one observation of the signal - - </para> -<para>A rectangular array for which each row of x represents a separate observation of the signal (for example, each row is one output of an array of sensors, as in array processing), such that x'*x is an estimate of the correlation matrix</para> -</refsection> - - -<refsection> - <title>Examples</title> - <programlisting role="example"><![CDATA[ - n = 0:199; - x = cos(0.257*%pi*n) + sin(0.2*%pi*n) ; - [S,w]=pmusic(x,2,16,1) - ]]></programlisting> -</refsection> - - -<refsection> - <title>See also</title> - <simplelist type="inline"> - <member><link linkend="pburg">pburg| peig | periodogram | pmtm | prony | pwelch | rooteig | rootmusic</link></member> - </simplelist> -</refsection> - -<refsection> - <title>Authors</title> - <simplelist type="vert"> - <member>Ayush Baid</member> - <member>References</member> - <member>[1] Petre Stoica and Randolph Moses, Introduction To Spectral</member> - <member>Analysis, Prentice-Hall, 1997, pg. 15</member> - <member>[2] S. J. Orfanidis, Optimum Signal Processing. An Introduction.</member> - <member>2nd Ed., Macmillan, 1988.</member> - </simplelist> -</refsection> + + Examples: + + Ex1: + n = 0:199; + x = cos(0.257*%pi*n) + sin(0.2*%pi*n) + 0.01*rand(size(n,"r"),"normal"); + [S,w]=pmusic(x,[%inf,1.1],[],8000,2) ;//where x: [1x200 constant] p:-infinite signal space and threshold value is 1.1 window length:-7 Fs:-8000Hz........fftlength:-256 + plot(w,S);.........to see the plot of psuedospectrum estimate of x vs frequencies w + </synopsis> +</refsynopsisdiv> </refentry> |