From 9ca7882cee16ad48b18df989e8300c697010e55a Mon Sep 17 00:00:00 2001
From: Sunil Shetye
Date: Wed, 25 Jul 2018 16:27:51 +0530
Subject: code changes by Sonu Sharma during FOSSEE Fellowship 2018

---
 help/en_US/lpc.xml | 67 ++++++++++++++++++++++++++++++++++++++----------------
 1 file changed, 47 insertions(+), 20 deletions(-)

(limited to 'help/en_US/lpc.xml')

diff --git a/help/en_US/lpc.xml b/help/en_US/lpc.xml
index 1ebe531..079923e 100644
--- a/help/en_US/lpc.xml
+++ b/help/en_US/lpc.xml
@@ -18,6 +18,7 @@
   <refnamediv>
     <refname>lpc</refname>
     <refpurpose>Linear prediction filter coefficients</refpurpose>
+    <para> </para>
   </refnamediv>
 
 
@@ -26,9 +27,8 @@
    <synopsis>
    [a,g] = lpc(x)
    [a,g] = lpc(x,p)
-   
-   
    </synopsis>
+    <para> </para>
 </refsynopsisdiv>
 
 <refsection>
@@ -36,44 +36,71 @@
    <para>
 [a,g] = lpc(x,p)
 Determines the coefficients of a pth order forward linear predictor
-filter by minimizing the squared error. If p is unspecified, a
-default value of length(x)-1 is used.
+filter by minimizing the squared error. <para>If p is unspecified, a
+default value of length(x)-1 is used.</para>
    </para>
-   <para>
-</para>
+   <para> </para>
 </refsection>
 
 <refsection>
    <title>Parameters</title>
    <variablelist>
    <varlistentry><term>x:</term>
-      <listitem><para> double</para></listitem></varlistentry>
+      <listitem><para> double</para><para>input signal, if it is a matrix each column is computed independently</para></listitem></varlistentry>
    <varlistentry><term>p:</term>
-      <listitem><para> int, natural number, scalar</para></listitem></varlistentry>
+      <listitem><para> int, natural number, scalar</para><para>order of linear predictor filter, value must be scalar, positive and must be less than or equal to length of input signal </para></listitem></varlistentry>
    <varlistentry><term>a:</term>
-      <listitem><para> double</para></listitem></varlistentry>
+      <listitem><para> double</para><para>coefficient of forward linear predictor, coefficient for each signal input is returned as a row vector</para></listitem></varlistentry>
    <varlistentry><term>g:</term>
-      <listitem><para> double</para></listitem></varlistentry>
+      <listitem><para> double</para><para>Column vector of averaged square prediction error</para></listitem></varlistentry>
    </variablelist>
 </refsection>
 
+<refsection>
+   <title>Description</title>
+   <para> This function determines coefficients of a forward linear predictor by minimizing prediction error in least squares sense. It is used in Digital Filter Design </para>
+    <para> </para>
+</refsection>
+
 <refsection>
    <title>Examples</title>
-   <programlisting role="example"><![CDATA[
-1)
-noise = randn(20000,1);
-x = filter(1,[1 1/5 1/3 1/4],noise);
-x = x(15904:20000);
-[a,g] = lpc(x,3);
 
+   <programlisting role="example"><![CDATA[
+noise = rand(50000,1,"normal"); //Gaussian White Noise
+x = filter(1,[1 1/2 1/3 1/4],noise);
+x = x(45904:50000);
+[a,g]= lpc(x,3)
+est_x = filter([0 -a(2:$)],1,x);
+e = x-est_x;
+[acs,lags] = xcorr(e,'coeff');
+plot(1:97,x(4001:4097),1:97,est_x(4001:4097),'--');
+a = gca();
+a.grid = [1,1];
+title 'Original Signal vs. LPC Estimate';
+xlabel 'Sample number', ylabel 'Amplitude';
+legend('Original signal','LPC estimate');
+   ]]></programlisting>
 
-References
-[1] Hayes, Monson H. Statistical digital signal processing and modeling.
-John Wiley & Sons, 2009, pg. 220
+ <scilab:image>
+noise = rand(50000,1,"normal"); //Gaussian White Noise
+x = filter(1,[1 1/2 1/3 1/4],noise);
+x = x(45904:50000);
+[a,g]= lpc(x,3)
+est_x = filter([0 -a(2:$)],1,x);
+e = x-est_x;
+[acs,lags] = xcorr(e,'coeff');
+plot(1:97,x(4001:4097),1:97,est_x(4001:4097),'--');
+a = gca();
+a.grid = [1,1];
+title 'Original Signal vs. LPC Estimate';
+xlabel 'Sample number', ylabel 'Amplitude';
+legend('Original signal','LPC estimate');
+        </scilab:image>
+   <para> </para>
 
-   ]]></programlisting>
 </refsection>
 
+
 <refsection>
    <title>See also</title>
    <simplelist type="inline">
-- 
cgit