1<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
2<html>
3
4<head>
5<META http-equiv="Content-Type" content="text/html; charset=windows-1252">
6<META name="GENERATOR" content="IBM WebSphere Studio Homepage Builder Version 8.0.0.0 for Windows">
7<META http-equiv="Content-Style-Type" content="text/css">
8<title>MATH_IFFTReal</title>
9<LINK rel="stylesheet" href="../../css/nitro.css" type="text/css">
10</head>
11
12<body>
13
14<h1 align="left">MATH_IFFTReal <img src="../../image/NTR.gif"align="middle"><img src="../../image/TWL.gif" align="middle"></H1>
15<H2>Syntax</H2>
16
17<dl>
18  <dd>
19  <PRE><CODE>#include &lt;nitro/math/fft.h&gt;
20
21void MATH_IFFTReal( fx32* data, u32 nShift, const fx16* sinTable, const fx16* sinTable2 );
22</CODE></PRE>
23  </dd>
24</dl><h2>Arguments</h2>
25
26<table border="1" width="100%">
27    <TR>
28      <TD width="176"><em><strong><font face="Courier New">data</font></strong></em></TD>
29      <TD width="670">Pointer to the data to transform. Data after the transform is overwritten.</TD>
30    </TR>
31    <TR>
32      <TD width="176"><em><strong><font face="Courier New">nShift</font></strong></em></TD>
33      <TD width="670">The value that is the base-2 logarithm of the number of real numbers that is expected to be output.</TD>
34    </TR>
35    <TR>
36      <TD width="176"><em><strong><font face="Courier New">sinTable</font></strong></em></TD>
37      <TD width="670">Pointer to the table of sine values.</TD>
38    </TR>
39    <TR>
40      <TD width="176"><em><strong><font face="Courier New">sinTable2</font></strong></em></TD>
41      <TD width="670">Pointer to the sine value table that omits every other value in sinTable.</TD>
42    </TR>
43  </table>
44<h2>Return Values</h2>
45<p>None.</p>
46<H2>Description</H2>
47<P>
48Uses a fast Fourier transform algorithm to perform the inverse transform of a discrete Fourier transform. Converts an array of complex numbers in the <CODE><A href="MATH_FFTReal.html">MATH_FFTReal</A></CODE> function's output format into an array of real numbers in that same function's input format.
49</P>
50<P>
51In the explanation below, the value 2<sup>nShift</sup> (2 to the nShift power) is represented as N. <CODE>data</CODE> is an array of N <CODE>fx32</CODE> values in the <CODE><A href="MATH_FFTReal.html">MATH_FFTReal</A></CODE> function's output format. <I>sinTable</I> is the pointer to the array of N*3/4 length that has fx16 type sine values assigned that satisfy <CODE>sinTable[k] = <i>sin</i>( (2&#x3C0;/N) * k )  (k = 0, 1,..., N*3/4-1)</CODE>. This can be created using the <CODE><A href="MATH_MakeFFTSinTable.html">MATH_MakeFFTSinTable</A></CODE> function. <CODE>sinTable2</CODE> is the pointer to the array of length (<CODE>N/2)*3/4</CODE>, where fx16 type sine values are assigned that satisfy <CODE>sinTable2[k] = <i>sin</i>( (2&#x3C0;/(N/2)) * k )  (k = 0, 1,..., (N/2)*3/4-1)</CODE>. This can be created by passing a value smaller by 1 to the <CODE>nShift</CODE> argument with the <CODE><A href="MATH_MakeFFTSinTable.html">MATH_MakeFFTSinTable</A></CODE> function.
52</P>
53<P>
54The calculations are performed using fixed-decimal numbers, so if a large value is given as the input, there is a risk of overflow. When the input value is viewed as type s32, the absolute value should not be greater than or equal to <CODE>0x80000000/N</CODE>. Also note that the maximum error when performing both the forward transform and inverse transform is around several times <CODE>FX32_ONE</CODE>.
55</P>
56<h2>See Also</h2>
57<P><CODE><A href="MATH_MakeFFTSinTable.html">MATH_MakeFFTSinTable</A>, <A href="MATH_FFT.html">MATH_FFT</A>, <A href="MATH_IFFT.html">MATH_IFFT</A>, <A href="MATH_FFTReal.html">MATH_FFTReal</A></CODE></P>
58<H2>Revision History</H2>
59<P>
602005/07/21 Corrected the explanation for <CODE>sinTable</CODE>.<BR>2005/07/11 Corrected the explanation.<BR> 2005/05/13 Initial version.
61</P>
62<hr><p>CONFIDENTIAL</p></body>
63</html>
64