devn00b
/
EQ2EMu


			
				
					
						
						
							123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336
							<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=US-ASCII">
<title>Signal Statistics</title>
<link rel="stylesheet" href="../math.css" type="text/css">
<meta name="generator" content="DocBook XSL Stylesheets V1.79.1">
<link rel="home" href="../index.html" title="Math Toolkit 2.11.0">
<link rel="up" href="../statistics.html" title="Chapter&#160;6.&#160;Statistics">
<link rel="prev" href="bivariate_statistics.html" title="Bivariate Statistics">
<link rel="next" href="anderson_darling.html" title="The Anderson-Darling Test">
</head>
<body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF">
<table cellpadding="2" width="100%"><tr>
<td valign="top"><img alt="Boost C++ Libraries" width="277" height="86" src="../../../../../boost.png"></td>
<td align="center"><a href="../../../../../index.html">Home</a></td>
<td align="center"><a href="../../../../../libs/libraries.htm">Libraries</a></td>
<td align="center"><a href="http://www.boost.org/users/people.html">People</a></td>
<td align="center"><a href="http://www.boost.org/users/faq.html">FAQ</a></td>
<td align="center"><a href="../../../../../more/index.htm">More</a></td>
</tr></table>
<hr>
<div class="spirit-nav">
<a accesskey="p" href="bivariate_statistics.html"><img src="../../../../../doc/src/images/prev.png" alt="Prev"></a><a accesskey="u" href="../statistics.html"><img src="../../../../../doc/src/images/up.png" alt="Up"></a><a accesskey="h" href="../index.html"><img src="../../../../../doc/src/images/home.png" alt="Home"></a><a accesskey="n" href="anderson_darling.html"><img src="../../../../../doc/src/images/next.png" alt="Next"></a>
</div>
<div class="section">
<div class="titlepage"><div><div><h2 class="title" style="clear: both">
<a name="math_toolkit.signal_statistics"></a><a class="link" href="signal_statistics.html" title="Signal Statistics">Signal Statistics</a>
</h2></div></div></div>
<h4>
<a name="math_toolkit.signal_statistics.h0"></a>
      <span class="phrase"><a name="math_toolkit.signal_statistics.synopsis"></a></span><a class="link" href="signal_statistics.html#math_toolkit.signal_statistics.synopsis">Synopsis</a>
    </h4>
<pre class="programlisting"><span class="preprocessor">#include</span> <span class="special">&lt;</span><span class="identifier">boost</span><span class="special">/</span><span class="identifier">math</span><span class="special">/</span><span class="identifier">statistics</span><span class="special">/</span><span class="identifier">signal_statistics</span><span class="special">.</span><span class="identifier">hpp</span><span class="special">&gt;</span>

<span class="keyword">namespace</span> <span class="identifier">boost</span><span class="special">::</span><span class="identifier">math</span><span class="special">::</span><span class="identifier">statistics</span> <span class="special">{</span>

    <span class="keyword">template</span><span class="special">&lt;</span><span class="keyword">class</span> <span class="identifier">Container</span><span class="special">&gt;</span>
    <span class="keyword">auto</span> <span class="identifier">absolute_gini_coefficient</span><span class="special">(</span><span class="identifier">Container</span> <span class="special">&amp;</span> <span class="identifier">c</span><span class="special">);</span>

    <span class="keyword">template</span><span class="special">&lt;</span><span class="keyword">class</span> <span class="identifier">ForwardIterator</span><span class="special">&gt;</span>
    <span class="keyword">auto</span> <span class="identifier">absolute_gini_coefficient</span><span class="special">(</span><span class="identifier">ForwardIterator</span> <span class="identifier">first</span><span class="special">,</span> <span class="identifier">ForwardIterator</span> <span class="identifier">last</span><span class="special">);</span>

    <span class="keyword">template</span><span class="special">&lt;</span><span class="keyword">class</span> <span class="identifier">Container</span><span class="special">&gt;</span>
    <span class="keyword">auto</span> <span class="identifier">sample_absolute_gini_coefficient</span><span class="special">(</span><span class="identifier">Container</span> <span class="special">&amp;</span> <span class="identifier">c</span><span class="special">);</span>

    <span class="keyword">template</span><span class="special">&lt;</span><span class="keyword">class</span> <span class="identifier">ForwardIterator</span><span class="special">&gt;</span>
    <span class="keyword">auto</span> <span class="identifier">sample_absolute_gini_coefficient</span><span class="special">(</span><span class="identifier">ForwardIterator</span> <span class="identifier">first</span><span class="special">,</span> <span class="identifier">ForwardIterator</span> <span class="identifier">last</span><span class="special">);</span>

    <span class="keyword">template</span><span class="special">&lt;</span><span class="keyword">class</span> <span class="identifier">Container</span><span class="special">&gt;</span>
    <span class="keyword">auto</span> <span class="identifier">hoyer_sparsity</span><span class="special">(</span><span class="identifier">Container</span> <span class="keyword">const</span> <span class="special">&amp;</span> <span class="identifier">c</span><span class="special">);</span>

    <span class="keyword">template</span><span class="special">&lt;</span><span class="keyword">class</span> <span class="identifier">ForwardIterator</span><span class="special">&gt;</span>
    <span class="keyword">auto</span> <span class="identifier">hoyer_sparsity</span><span class="special">(</span><span class="identifier">ForwardIterator</span> <span class="identifier">first</span><span class="special">,</span> <span class="identifier">ForwardIterator</span> <span class="identifier">last</span><span class="special">);</span>

    <span class="keyword">template</span><span class="special">&lt;</span><span class="keyword">class</span> <span class="identifier">Container</span><span class="special">&gt;</span>
    <span class="keyword">auto</span> <span class="identifier">oracle_snr</span><span class="special">(</span><span class="identifier">Container</span> <span class="keyword">const</span> <span class="special">&amp;</span> <span class="identifier">signal</span><span class="special">,</span> <span class="identifier">Container</span> <span class="keyword">const</span> <span class="special">&amp;</span> <span class="identifier">noisy_signal</span><span class="special">);</span>

    <span class="keyword">template</span><span class="special">&lt;</span><span class="keyword">class</span> <span class="identifier">Container</span><span class="special">&gt;</span>
    <span class="keyword">auto</span> <span class="identifier">oracle_snr_db</span><span class="special">(</span><span class="identifier">Container</span> <span class="keyword">const</span> <span class="special">&amp;</span> <span class="identifier">signal</span><span class="special">,</span> <span class="identifier">Container</span> <span class="keyword">const</span> <span class="special">&amp;</span> <span class="identifier">noisy_signal</span><span class="special">);</span>

    <span class="keyword">template</span><span class="special">&lt;</span><span class="keyword">class</span> <span class="identifier">ForwardIterator</span><span class="special">&gt;</span>
    <span class="keyword">auto</span> <span class="identifier">m2m4_snr_estimator</span><span class="special">(</span><span class="identifier">ForwardIterator</span> <span class="identifier">first</span><span class="special">,</span> <span class="identifier">ForwardIterator</span> <span class="identifier">last</span><span class="special">,</span> <span class="keyword">decltype</span><span class="special">(*</span><span class="identifier">first</span><span class="special">)</span> <span class="identifier">estimated_signal_kurtosis</span><span class="special">=</span><span class="number">1</span><span class="special">,</span> <span class="keyword">decltype</span><span class="special">(*</span><span class="identifier">first</span><span class="special">)</span> <span class="identifier">estimated_noise_kurtosis</span><span class="special">=</span><span class="number">3</span><span class="special">);</span>

    <span class="keyword">template</span><span class="special">&lt;</span><span class="keyword">class</span> <span class="identifier">Container</span><span class="special">&gt;</span>
    <span class="keyword">auto</span> <span class="identifier">m2m4_snr_estimator</span><span class="special">(</span><span class="identifier">Container</span> <span class="keyword">const</span> <span class="special">&amp;</span> <span class="identifier">noisy_signal</span><span class="special">,</span> <span class="keyword">typename</span> <span class="identifier">Container</span><span class="special">::</span><span class="identifier">value_type</span> <span class="identifier">estimated_signal_kurtosis</span><span class="special">=</span><span class="number">1</span><span class="special">,</span> <span class="keyword">typename</span> <span class="identifier">Container</span><span class="special">::</span><span class="identifier">value_type</span> <span class="identifier">estimate_noise_kurtosis</span><span class="special">=</span><span class="number">3</span><span class="special">);</span>

    <span class="keyword">template</span><span class="special">&lt;</span><span class="keyword">class</span> <span class="identifier">ForwardIterator</span><span class="special">&gt;</span>
    <span class="keyword">auto</span> <span class="identifier">m2m4_snr_estimator_db</span><span class="special">(</span><span class="identifier">ForwardIterator</span> <span class="identifier">first</span><span class="special">,</span> <span class="identifier">ForwardIterator</span> <span class="identifier">last</span><span class="special">,</span> <span class="keyword">decltype</span><span class="special">(*</span><span class="identifier">first</span><span class="special">)</span> <span class="identifier">estimated_signal_kurtosis</span><span class="special">=</span><span class="number">1</span><span class="special">,</span> <span class="keyword">decltype</span><span class="special">(*</span><span class="identifier">first</span><span class="special">)</span> <span class="identifier">estimated_noise_kurtosis</span><span class="special">=</span><span class="number">3</span><span class="special">);</span>

    <span class="keyword">template</span><span class="special">&lt;</span><span class="keyword">class</span> <span class="identifier">Container</span><span class="special">&gt;</span>
    <span class="keyword">auto</span> <span class="identifier">m2m4_snr_estimator_db</span><span class="special">(</span><span class="identifier">Container</span> <span class="keyword">const</span> <span class="special">&amp;</span> <span class="identifier">noisy_signal</span><span class="special">,</span><span class="keyword">typename</span> <span class="identifier">Container</span><span class="special">::</span><span class="identifier">value_type</span> <span class="identifier">estimated_signal_kurtosis</span><span class="special">=</span><span class="number">1</span><span class="special">,</span> <span class="keyword">typename</span> <span class="identifier">Container</span><span class="special">::</span><span class="identifier">value_type</span> <span class="identifier">estimate_noise_kurtosis</span><span class="special">=</span><span class="number">3</span><span class="special">);</span>

<span class="special">}</span>
</pre>
<h4>
<a name="math_toolkit.signal_statistics.h1"></a>
      <span class="phrase"><a name="math_toolkit.signal_statistics.description"></a></span><a class="link" href="signal_statistics.html#math_toolkit.signal_statistics.description">Description</a>
    </h4>
<p>
      The file <code class="computeroutput"><span class="identifier">boost</span><span class="special">/</span><span class="identifier">math</span><span class="special">/</span><span class="identifier">statistics</span><span class="special">/</span><span class="identifier">signal_statistics</span><span class="special">.</span><span class="identifier">hpp</span></code> is a
      set of facilities for computing quantities commonly used in signal analysis.
    </p>
<p>
      Our examples use <code class="computeroutput"><span class="identifier">std</span><span class="special">::</span><span class="identifier">vector</span><span class="special">&lt;</span><span class="keyword">double</span><span class="special">&gt;</span></code> to
      hold the data, but this not required. In general, you can store your data in
      an Eigen array, and Armadillo vector, <code class="computeroutput"><span class="identifier">std</span><span class="special">::</span><span class="identifier">array</span></code>,
      and for many of the routines, a <code class="computeroutput"><span class="identifier">std</span><span class="special">::</span><span class="identifier">forward_list</span></code>.
      These routines are usable in float, double, long double, and Boost.Multiprecision
      precision, as well as their complex extensions whenever the computation is
      well-defined.
    </p>
<h4>
<a name="math_toolkit.signal_statistics.h2"></a>
      <span class="phrase"><a name="math_toolkit.signal_statistics.absolute_gini_coefficient"></a></span><a class="link" href="signal_statistics.html#math_toolkit.signal_statistics.absolute_gini_coefficient">Absolute
      Gini Coefficient</a>
    </h4>
<p>
      The Gini coefficient, first used to measure wealth inequality, is also one
      of the best measures of the sparsity of an expansion in a basis. A sparse expansion
      has most of its norm concentrated in just a few coefficients, making the connection
      with wealth inequality obvious. See <a href="https://arxiv.org/pdf/0811.4706.pdf" target="_top">Hurley
      and Rickard</a> for details. However, for measuring sparsity, the phase
      of the numbers is irrelevant, so we provide the <code class="computeroutput"><span class="identifier">absolute_gini_coefficient</span></code>:
    </p>
<pre class="programlisting"><span class="keyword">using</span> <span class="identifier">boost</span><span class="special">::</span><span class="identifier">math</span><span class="special">::</span><span class="identifier">statistics</span><span class="special">::</span><span class="identifier">sample_absolute_gini_coefficient</span><span class="special">;</span>
<span class="keyword">using</span> <span class="identifier">boost</span><span class="special">::</span><span class="identifier">math</span><span class="special">::</span><span class="identifier">statistics</span><span class="special">::</span><span class="identifier">absolute_gini_coefficient</span><span class="special">;</span>
<span class="identifier">std</span><span class="special">::</span><span class="identifier">vector</span><span class="special">&lt;</span><span class="identifier">std</span><span class="special">::</span><span class="identifier">complex</span><span class="special">&lt;</span><span class="keyword">double</span><span class="special">&gt;&gt;</span> <span class="identifier">v</span><span class="special">{{</span><span class="number">0</span><span class="special">,</span><span class="number">1</span><span class="special">},</span> <span class="special">{</span><span class="number">0</span><span class="special">,</span><span class="number">0</span><span class="special">},</span> <span class="special">{</span><span class="number">0</span><span class="special">,</span><span class="number">0</span><span class="special">},</span> <span class="special">{</span><span class="number">0</span><span class="special">,</span><span class="number">0</span><span class="special">}};</span>
<span class="keyword">double</span> <span class="identifier">abs_gini</span> <span class="special">=</span> <span class="identifier">sample_absolute_gini_coefficient</span><span class="special">(</span><span class="identifier">v</span><span class="special">);</span>
<span class="comment">// now abs_gini = 1; maximally unequal</span>

<span class="identifier">std</span><span class="special">::</span><span class="identifier">vector</span><span class="special">&lt;</span><span class="identifier">std</span><span class="special">::</span><span class="identifier">complex</span><span class="special">&lt;</span><span class="keyword">double</span><span class="special">&gt;&gt;</span> <span class="identifier">w</span><span class="special">{{</span><span class="number">0</span><span class="special">,</span><span class="number">1</span><span class="special">},</span> <span class="special">{</span><span class="number">1</span><span class="special">,</span><span class="number">0</span><span class="special">},</span> <span class="special">{</span><span class="number">0</span><span class="special">,-</span><span class="number">1</span><span class="special">},</span> <span class="special">{-</span><span class="number">1</span><span class="special">,</span><span class="number">0</span><span class="special">}};</span>
<span class="identifier">abs_gini</span> <span class="special">=</span> <span class="identifier">absolute_gini_coefficient</span><span class="special">(</span><span class="identifier">w</span><span class="special">);</span>
<span class="comment">// now abs_gini = 0; every element of the vector has equal magnitude</span>

<span class="identifier">std</span><span class="special">::</span><span class="identifier">vector</span><span class="special">&lt;</span><span class="keyword">double</span><span class="special">&gt;</span> <span class="identifier">u</span><span class="special">{-</span><span class="number">1</span><span class="special">,</span> <span class="number">1</span><span class="special">,</span> <span class="special">-</span><span class="number">1</span><span class="special">};</span>
<span class="identifier">abs_gini</span> <span class="special">=</span> <span class="identifier">absolute_gini_coefficient</span><span class="special">(</span><span class="identifier">u</span><span class="special">);</span>
<span class="comment">// now abs_gini = 0</span>
<span class="comment">// Alternative call useful for computing over subset of the input:</span>
<span class="identifier">abs_gini</span> <span class="special">=</span> <span class="identifier">absolute_gini_coefficient</span><span class="special">(</span><span class="identifier">u</span><span class="special">.</span><span class="identifier">begin</span><span class="special">(),</span> <span class="identifier">u</span><span class="special">.</span><span class="identifier">begin</span><span class="special">()</span> <span class="special">+</span> <span class="number">1</span><span class="special">);</span>
</pre>
<p>
      The sample Gini coefficient returns unity for a vector which has only one nonzero
      coefficient. The population Gini coefficient of a vector with one non-zero
      element is dependent on the length of the input.
    </p>
<p>
      The sample Gini coefficient lacks one desirable property of the population
      Gini coefficient, namely that "cloning" a vector has the same Gini
      coefficient; though cloning holds to very high accuracy with the sample Gini
      coefficient and can easily be recovered by a rescaling.
    </p>
<p>
      If sorting the input data is too much expense for a sparsity measure (is it
      going to be perfect anyway?), consider calculating the Hoyer sparsity instead.
    </p>
<h4>
<a name="math_toolkit.signal_statistics.h3"></a>
      <span class="phrase"><a name="math_toolkit.signal_statistics.hoyer_sparsity"></a></span><a class="link" href="signal_statistics.html#math_toolkit.signal_statistics.hoyer_sparsity">Hoyer
      Sparsity</a>
    </h4>
<p>
      The Hoyer sparsity measures a normalized ratio of the &#8467;<sup>1</sup> and &#8467;<sup>2</sup> norms.
      As the name suggests, it is used to measure the sparsity of an expansion in
      some basis.
    </p>
<p>
      The Hoyer sparsity computes (&#8730;<span class="emphasis"><em>N</em></span> - &#8467;<sup>1</sup>(v)/&#8467;<sup>2</sup>(v))/(&#8730;N
      -1). For details, see <a href="http://www.jmlr.org/papers/volume5/hoyer04a/hoyer04a.pdf" target="_top">Hoyer</a>
      as well as <a href="https://arxiv.org/pdf/0811.4706.pdf" target="_top">Hurley and Rickard</a>.
    </p>
<p>
      A few special cases will serve to clarify the intended use: If <span class="emphasis"><em>v</em></span>
      has only one nonzero coefficient, the Hoyer sparsity attains its maxima of
      1. If the coefficients of <span class="emphasis"><em>v</em></span> all have the same magnitude,
      then the Hoyer sparsity attains its minima of zero. If the elements of <span class="emphasis"><em>v</em></span>
      are uniformly distributed on an interval [0, <span class="emphasis"><em>b</em></span>], then
      the Hoyer sparsity is approximately 0.133.
    </p>
<p>
      Usage:
    </p>
<pre class="programlisting"><span class="identifier">std</span><span class="special">::</span><span class="identifier">vector</span><span class="special">&lt;</span><span class="identifier">Real</span><span class="special">&gt;</span> <span class="identifier">v</span><span class="special">{</span><span class="number">1</span><span class="special">,</span><span class="number">0</span><span class="special">,</span><span class="number">0</span><span class="special">};</span>
<span class="identifier">Real</span> <span class="identifier">hs</span> <span class="special">=</span> <span class="identifier">boost</span><span class="special">::</span><span class="identifier">math</span><span class="special">::</span><span class="identifier">statistics</span><span class="special">::</span><span class="identifier">hoyer_sparsity</span><span class="special">(</span><span class="identifier">v</span><span class="special">);</span>
<span class="comment">// hs = 1</span>
<span class="identifier">std</span><span class="special">::</span><span class="identifier">vector</span><span class="special">&lt;</span><span class="identifier">Real</span><span class="special">&gt;</span> <span class="identifier">v</span><span class="special">{</span><span class="number">1</span><span class="special">,-</span><span class="number">1</span><span class="special">,</span><span class="number">1</span><span class="special">};</span>
<span class="identifier">Real</span> <span class="identifier">hs</span> <span class="special">=</span> <span class="identifier">boost</span><span class="special">::</span><span class="identifier">math</span><span class="special">::</span><span class="identifier">statistics</span><span class="special">::</span><span class="identifier">hoyer_sparsity</span><span class="special">(</span><span class="identifier">v</span><span class="special">.</span><span class="identifier">begin</span><span class="special">(),</span> <span class="identifier">v</span><span class="special">.</span><span class="identifier">end</span><span class="special">());</span>
<span class="comment">// hs = 0</span>
</pre>
<p>
      The container must be forward iterable and the contents are not modified. Accepts
      real, complex, and integer inputs. If the input is an integral type, the output
      is a double precision float.
    </p>
<h4>
<a name="math_toolkit.signal_statistics.h4"></a>
      <span class="phrase"><a name="math_toolkit.signal_statistics.oracle_signal_to_noise_ratio"></a></span><a class="link" href="signal_statistics.html#math_toolkit.signal_statistics.oracle_signal_to_noise_ratio">Oracle
      Signal-to-noise ratio</a>
    </h4>
<p>
      The function <code class="computeroutput"><span class="identifier">oracle_snr</span></code> computes
      the ratio &#8214; <span class="emphasis"><em>s</em></span> &#8214;<sub>2</sub><sup>2</sup> / &#8214; <span class="emphasis"><em>s</em></span>
      - <span class="emphasis"><em>x</em></span> &#8214;<sub>2</sub><sup>2</sup>, where <span class="emphasis"><em>s</em></span> is signal
      and <span class="emphasis"><em>x</em></span> is a noisy signal. The function <code class="computeroutput"><span class="identifier">oracle_snr_db</span></code>
      computes 10<code class="computeroutput"><span class="identifier">log</span></code><sub>10</sub>(&#8214;
      <span class="emphasis"><em>s</em></span> &#8214;<sup>2</sup> / &#8214; <span class="emphasis"><em>s</em></span> - <span class="emphasis"><em>x</em></span>
      &#8214;<sup>2</sup>). The functions are so named because in general, one does not know
      how to decompose a real signal <span class="emphasis"><em>x</em></span> into <span class="emphasis"><em>s</em></span>
      + <span class="emphasis"><em>w</em></span> and as such <span class="emphasis"><em>s</em></span> is regarded as
      oracle information. Hence this function is mainly useful for unit testing other
      SNR estimators.
    </p>
<p>
      Usage:
    </p>
<pre class="programlisting"><span class="identifier">std</span><span class="special">::</span><span class="identifier">vector</span><span class="special">&lt;</span><span class="keyword">double</span><span class="special">&gt;</span> <span class="identifier">signal</span><span class="special">(</span><span class="number">500</span><span class="special">,</span> <span class="number">3.2</span><span class="special">);</span>
<span class="identifier">std</span><span class="special">::</span><span class="identifier">vector</span><span class="special">&lt;</span><span class="keyword">double</span><span class="special">&gt;</span> <span class="identifier">noisy_signal</span><span class="special">(</span><span class="number">500</span><span class="special">);</span>
<span class="comment">// fill 'noisy_signal' signal + noise</span>
<span class="keyword">double</span> <span class="identifier">snr_db</span> <span class="special">=</span> <span class="identifier">boost</span><span class="special">::</span><span class="identifier">math</span><span class="special">::</span><span class="identifier">statistics</span><span class="special">::</span><span class="identifier">oracle_snr_db</span><span class="special">(</span><span class="identifier">signal</span><span class="special">,</span> <span class="identifier">noisy_signal</span><span class="special">);</span>
<span class="keyword">double</span> <span class="identifier">snr</span> <span class="special">=</span> <span class="identifier">boost</span><span class="special">::</span><span class="identifier">math</span><span class="special">::</span><span class="identifier">statistics</span><span class="special">::</span><span class="identifier">oracle_snr</span><span class="special">(</span><span class="identifier">signal</span><span class="special">,</span> <span class="identifier">noisy_signal</span><span class="special">);</span>
</pre>
<p>
      The input can be real, complex, or integral. Integral inputs produce double
      precision floating point outputs. The input data is not modified and must satisfy
      the requirements of a <code class="computeroutput"><span class="identifier">RandomAccessContainer</span></code>.
    </p>
<h4>
<a name="math_toolkit.signal_statistics.h5"></a>
      <span class="phrase"><a name="math_toolkit.signal_statistics.m_sub_2_m_sub_4_snr_estimation"></a></span><a class="link" href="signal_statistics.html#math_toolkit.signal_statistics.m_sub_2_m_sub_4_snr_estimation"><span class="emphasis"><em>M</em></span><sub>2</sub><span class="emphasis"><em>M</em></span><sub>4</sub> SNR
      Estimation</a>
    </h4>
<p>
      Estimates the SNR of a noisy signal via the <span class="emphasis"><em>M</em></span><sub>2</sub><span class="emphasis"><em>M</em></span><sub>4</sub> method.
      See <a href="https://doi.org/10.1109/26.871393" target="_top">Pauluzzi and N.C. Beaulieu</a>
      and <a href="https://doi.org/10.1109/ISIT.1994.394869" target="_top">Matzner and Englberger</a>
      for details.
    </p>
<pre class="programlisting"><span class="identifier">std</span><span class="special">::</span><span class="identifier">vector</span><span class="special">&lt;</span><span class="keyword">double</span><span class="special">&gt;</span> <span class="identifier">noisy_signal</span><span class="special">(</span><span class="number">512</span><span class="special">);</span>
<span class="comment">// fill noisy_signal with data contaminated by Gaussian white noise:</span>
<span class="keyword">double</span> <span class="identifier">est_snr_db</span> <span class="special">=</span> <span class="identifier">boost</span><span class="special">::</span><span class="identifier">math</span><span class="special">::</span><span class="identifier">statistics</span><span class="special">::</span><span class="identifier">m2m4_snr_estimator_db</span><span class="special">(</span><span class="identifier">noisy_signal</span><span class="special">);</span>
</pre>
<p>
      The <span class="emphasis"><em>M</em></span><sub>2</sub><span class="emphasis"><em>M</em></span><sub>4</sub> SNR estimator is an "in-service"
      estimator, meaning that the estimate is made using the noisy, data-bearing
      signal, and does not require a background estimate. This estimator has been
      found to be work best between roughly -3 and 15db, tending to overestimate
      the noise below -3db, and underestimate the noise above 15db. See <a href="https://www.mdpi.com/2078-2489/8/3/75/pdf" target="_top">Xue
      et al</a> for details.
    </p>
<p>
      The <span class="emphasis"><em>M</em></span><sub>2</sub><span class="emphasis"><em>M</em></span><sub>4</sub> SNR estimator, by default,
      assumes that the kurtosis of the signal is 1 and the kurtosis of the noise
      is 3, the latter corresponding to Gaussian noise. These parameters, however,
      can be overridden:
    </p>
<pre class="programlisting"><span class="identifier">std</span><span class="special">::</span><span class="identifier">vector</span><span class="special">&lt;</span><span class="keyword">double</span><span class="special">&gt;</span> <span class="identifier">noisy_signal</span><span class="special">(</span><span class="number">512</span><span class="special">);</span>
<span class="comment">// fill noisy_signal with the data:</span>
<span class="keyword">double</span> <span class="identifier">signal_kurtosis</span> <span class="special">=</span> <span class="number">1.5</span><span class="special">;</span>
<span class="comment">// Noise is assumed to follow Laplace distribution, which has kurtosis of 6:</span>
<span class="keyword">double</span> <span class="identifier">noise_kurtosis</span> <span class="special">=</span> <span class="number">6</span><span class="special">;</span>
<span class="keyword">double</span> <span class="identifier">est_snr</span> <span class="special">=</span> <span class="identifier">boost</span><span class="special">::</span><span class="identifier">math</span><span class="special">::</span><span class="identifier">statistics</span><span class="special">::</span><span class="identifier">m2m4_snr_estimator_db</span><span class="special">(</span><span class="identifier">noisy_signal</span><span class="special">,</span> <span class="identifier">signal_kurtosis</span><span class="special">,</span> <span class="identifier">noise_kurtosis</span><span class="special">);</span>
</pre>
<p>
      Now, technically the method is a "blind SNR estimator", meaning that
      the no <span class="emphasis"><em>a-priori</em></span> information about the signal is required
      to use the method. However, the performance of the method is <span class="emphasis"><em>vastly</em></span>
      better if you can come up with a better estimate of the signal and noise kurtosis.
      How can we do this? Suppose we know that the SNR is much greater than 1. Then
      we can estimate the signal kurtosis simply by using the noisy signal kurtosis.
      If the SNR is much less than one, this method breaks down as the noisy signal
      kurtosis will tend to the noise kurtosis-though in this limit we have an excellent
      estimator of the noise kurtosis! In addition, if you have a model of what your
      signal should look like, you can precompute the signal kurtosis. For example,
      sinusoids have a kurtosis of 1.5. See <a href="http://www.jcomputers.us/vol8/jcp0808-21.pdf" target="_top">here</a>
      for a study which uses estimates of this sort to improve the performance of
      the <span class="emphasis"><em>M</em></span><sub>2</sub><span class="emphasis"><em>M</em></span><sub>4</sub> estimator.
    </p>
<p>
      <span class="emphasis"><em>Nota bene</em></span>: The traditional definition of SNR is <span class="emphasis"><em>not</em></span>
      mean invariant. By this we mean that if a constant is added to every sample
      of a signal, the SNR is changed. For example, adding DC bias to a signal changes
      its SNR. For most use cases, this is really not what you intend; for example
      a signal consisting of zeros plus Gaussian noise has an SNR of zero, whereas
      a signal with a constant DC bias and random Gaussian noise might have a very
      large SNR.
    </p>
<p>
      The <span class="emphasis"><em>M</em></span><sub>2</sub><span class="emphasis"><em>M</em></span><sub>4</sub> SNR estimator is computed
      from mean-invariant quantities, and hence it should really be compared to the
      mean-invariant SNR.
    </p>
<p>
      <span class="emphasis"><em>Nota bene</em></span>: This computation requires the solution of a
      system of quadratic equations involving the noise kurtosis, the signal kurtosis,
      and the second and fourth moments of the data. There is no guarantee that a
      solution of this system exists for all value of these parameters, in fact nonexistence
      can easily be demonstrated for certain data. If there is no solution to the
      system, then failure is communicated by returning NaNs. This happens distressingly
      often; if a user is aware of any blind SNR estimators which do not suffer from
      this drawback, please open a github ticket and let us know.
    </p>
<p>
      The author has not managed to fully characterize the conditions under which
      a real solution with <span class="emphasis"><em>S &gt; 0</em></span> and <span class="emphasis"><em>N &gt;0</em></span>
      exists. However, a very intuitive example demonstrates why nonexistence can
      occur. Suppose the signal and noise kurtosis are equal. Then the method has
      no way to distinguish between the signal and the noise, and the solution is
      non-unique.
    </p>
<h4>
<a name="math_toolkit.signal_statistics.h6"></a>
      <span class="phrase"><a name="math_toolkit.signal_statistics.references"></a></span><a class="link" href="signal_statistics.html#math_toolkit.signal_statistics.references">References</a>
    </h4>
<div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
<li class="listitem">
          Mallat, Stephane. <span class="emphasis"><em>A wavelet tour of signal processing: the sparse
          way.</em></span> Academic press, 2008.
        </li>
<li class="listitem">
          Hurley, Niall, and Scott Rickard. <span class="emphasis"><em>Comparing measures of sparsity.</em></span>
          IEEE Transactions on Information Theory 55.10 (2009): 4723-4741.
        </li>
<li class="listitem">
          Jensen, Arne, and Anders la Cour-Harbo. <span class="emphasis"><em>Ripples in mathematics:
          the discrete wavelet transform.</em></span> Springer Science &amp; Business
          Media, 2001.
        </li>
<li class="listitem">
          D. R. Pauluzzi and N. C. Beaulieu, <span class="emphasis"><em>A comparison of SNR estimation
          techniques for the AWGN channel,</em></span> IEEE Trans. Communications,
          Vol. 48, No. 10, pp. 1681-1691, 2000.
        </li>
<li class="listitem">
          Hoyer, Patrik O. <span class="emphasis"><em>Non-negative matrix factorization with sparseness
          constraints.</em></span>, Journal of machine learning research 5.Nov (2004):
          1457-1469.
        </li>
</ul></div>
</div>
<table xmlns:rev="http://www.cs.rpi.edu/~gregod/boost/tools/doc/revision" width="100%"><tr>
<td align="left"></td>
<td align="right"><div class="copyright-footer">Copyright &#169; 2006-2019 Nikhar
      Agrawal, Anton Bikineev, Paul A. Bristow, Marco Guazzone, Christopher Kormanyos,
      Hubert Holin, Bruno Lalande, John Maddock, Jeremy Murphy, Matthew Pulver, Johan
      R&#229;de, Gautam Sewani, Benjamin Sobotta, Nicholas Thompson, Thijs van den Berg,
      Daryle Walker and Xiaogang Zhang<p>
        Distributed under the Boost Software License, Version 1.0. (See accompanying
        file LICENSE_1_0.txt or copy at <a href="http://www.boost.org/LICENSE_1_0.txt" target="_top">http://www.boost.org/LICENSE_1_0.txt</a>)
      </p>
</div></td>
</tr></table>
<hr>
<div class="spirit-nav">
<a accesskey="p" href="bivariate_statistics.html"><img src="../../../../../doc/src/images/prev.png" alt="Prev"></a><a accesskey="u" href="../statistics.html"><img src="../../../../../doc/src/images/up.png" alt="Up"></a><a accesskey="h" href="../index.html"><img src="../../../../../doc/src/images/home.png" alt="Home"></a><a accesskey="n" href="anderson_darling.html"><img src="../../../../../doc/src/images/next.png" alt="Next"></a>
</div>
</body>
</html>