You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
636 lines
19 KiB
636 lines
19 KiB
<!DOCTYPE html>
|
|
|
|
<!--Converted with LaTeX2HTML 2002-2-1 (1.71)
|
|
original version by: Nikos Drakos, CBLU, University of Leeds
|
|
* revised and updated by: Marcus Hennecke, Ross Moore, Herb Swan
|
|
* with significant contributions from:
|
|
Jens Lippmann, Marek Rouchal, Martin Wilck and others -->
|
|
<HTML>
|
|
<HEAD>
|
|
|
|
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
|
|
|
|
|
<TITLE>Delay networks</TITLE>
|
|
<META NAME="description" CONTENT="Delay networks">
|
|
<META NAME="keywords" CONTENT="book">
|
|
<META NAME="resource-type" CONTENT="document">
|
|
<META NAME="distribution" CONTENT="global">
|
|
|
|
<META NAME="Generator" CONTENT="LaTeX2HTML v2002-2-1">
|
|
<META HTTP-EQUIV="Content-Style-Type" CONTENT="text/css">
|
|
|
|
<LINK REL="STYLESHEET" HREF="book.css">
|
|
|
|
<LINK REL="next" HREF="node109.html">
|
|
<LINK REL="previous" HREF="node107.html">
|
|
<LINK REL="up" HREF="node104.html">
|
|
<LINK REL="next" HREF="node109.html">
|
|
</HEAD>
|
|
|
|
<BODY >
|
|
<!--Navigation Panel-->
|
|
<A ID="tex2html2171"
|
|
HREF="node109.html">
|
|
<IMG WIDTH="37" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="next"
|
|
SRC="next.png"></A>
|
|
<A ID="tex2html2165"
|
|
HREF="node104.html">
|
|
<IMG WIDTH="26" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="up"
|
|
SRC="up.png"></A>
|
|
<A ID="tex2html2159"
|
|
HREF="node107.html">
|
|
<IMG WIDTH="63" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="previous"
|
|
SRC="prev.png"></A>
|
|
<A ID="tex2html2167"
|
|
HREF="node4.html">
|
|
<IMG WIDTH="65" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="contents"
|
|
SRC="contents.png"></A>
|
|
<A ID="tex2html2169"
|
|
HREF="node201.html">
|
|
<IMG WIDTH="43" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="index"
|
|
SRC="index.png"></A>
|
|
<BR>
|
|
<B> Next:</B> <A ID="tex2html2172"
|
|
HREF="node109.html">Recirculating delay networks</A>
|
|
<B> Up:</B> <A ID="tex2html2166"
|
|
HREF="node104.html">Time shifts and delays</A>
|
|
<B> Previous:</B> <A ID="tex2html2160"
|
|
HREF="node107.html">Time shifts and phase</A>
|
|
<B> <A ID="tex2html2168"
|
|
HREF="node4.html">Contents</A></B>
|
|
<B> <A ID="tex2html2170"
|
|
HREF="node201.html">Index</A></B>
|
|
<BR>
|
|
<BR>
|
|
<!--End of Navigation Panel-->
|
|
|
|
<H1><A ID="SECTION001130000000000000000"></A>
|
|
<A ID="sect7.network"></A>
|
|
<BR>
|
|
Delay networks
|
|
</H1>
|
|
|
|
<P>
|
|
|
|
<DIV ALIGN="CENTER"><A ID="fig07.03"></A><A ID="7867"></A>
|
|
<TABLE>
|
|
<CAPTION ALIGN="BOTTOM"><STRONG>Figure 7.3:</STRONG>
|
|
A delay network. Here we add the incoming signal to a delayed
|
|
copy of itself.</CAPTION>
|
|
<TR><TD><IMG
|
|
WIDTH="55" HEIGHT="219" BORDER="0"
|
|
SRC="img679.png"
|
|
ALT="\begin{figure}\psfig{file=figs/fig07.03.ps}\end{figure}"></TD></TR>
|
|
</TABLE>
|
|
</DIV>
|
|
|
|
<P>
|
|
If we consider our digital audio samples <IMG
|
|
WIDTH="36" HEIGHT="32" ALIGN="MIDDLE" BORDER="0"
|
|
SRC="img669.png"
|
|
ALT="$X[n]$"> to correspond to
|
|
successive moments in time, then time shifting the signal by <IMG
|
|
WIDTH="11" HEIGHT="14" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img28.png"
|
|
ALT="$d$"> samples
|
|
corresponds to a
|
|
<A ID="7870"></A>
|
|
<I>delay</I>
|
|
of <IMG
|
|
WIDTH="31" HEIGHT="32" ALIGN="MIDDLE" BORDER="0"
|
|
SRC="img680.png"
|
|
ALT="$d/R$"> time units, where <IMG
|
|
WIDTH="15" HEIGHT="14" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img36.png"
|
|
ALT="$R$"> is the sample rate.
|
|
Figure <A HREF="#fig07.03">7.3</A> shows one example of a
|
|
<A ID="7873"></A><I>linear delay network</I>:
|
|
an assembly of delay units, possibly with amplitude
|
|
scaling operations, combined using addition and subtraction. The output
|
|
is a linear function of the input, in the sense that adding two signals at the
|
|
input is the same as processing each one separately and adding the results.
|
|
Moreover, linear delay networks create no new frequencies in the output that
|
|
weren't present in the input, as long as the network remains time invariant,
|
|
so that the gains and delay times do not change with time.
|
|
|
|
<P>
|
|
In general there are two ways of thinking about delay networks. We can think
|
|
in the
|
|
<A ID="7875"></A><I>time domain</I>,
|
|
in which we draw waveforms as functions of time (or of the index <IMG
|
|
WIDTH="13" HEIGHT="13" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img75.png"
|
|
ALT="$n$">), and
|
|
consider delays as time shifts. Alternatively we may think in the
|
|
<A ID="7877"></A><I>frequency domain</I>,
|
|
in which we dose the input with a complex sinusoid (so that its output is a
|
|
sinusoid at the same frequency) and report the amplitude and/or phase change
|
|
wrought by the network, as a function of the frequency. We'll now look at the
|
|
delay network of Figure <A HREF="#fig07.03">7.3</A> in each of the two ways in turn.
|
|
|
|
<P>
|
|
|
|
<DIV ALIGN="CENTER"><A ID="fig07.04"></A><A ID="8340"></A>
|
|
<TABLE>
|
|
<CAPTION ALIGN="BOTTOM"><STRONG>Figure 7.4:</STRONG>
|
|
The time domain view of the delay network of Figure <A HREF="#fig07.03">7.3</A>.
|
|
The output is the sum of the input and its time shifted copy.</CAPTION>
|
|
<TR><TD><IMG
|
|
WIDTH="440" HEIGHT="257" BORDER="0"
|
|
SRC="img681.png"
|
|
ALT="\begin{figure}\psfig{file=figs/fig07.04.ps}\end{figure}"></TD></TR>
|
|
</TABLE>
|
|
</DIV>
|
|
|
|
<P>
|
|
Figure <A HREF="#fig07.04">7.4</A> shows the network's behavior in the time domain. We
|
|
invent some sort of suitable test function as input (it's a rectangular pulse
|
|
eight samples wide in this example) and graph the input and output as functions
|
|
of the sample number <IMG
|
|
WIDTH="13" HEIGHT="13" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img75.png"
|
|
ALT="$n$">. This particular delay network adds the input to a
|
|
delayed copy of itself.
|
|
|
|
<P>
|
|
A frequently used test function is an
|
|
<A ID="7886"></A><I>impulse</I>,
|
|
which is a pulse lasting only one sample. The utility of this is that, if we
|
|
know the output of the network for an impulse, we can find the output for
|
|
any other digital audio signal--because any signal <IMG
|
|
WIDTH="31" HEIGHT="32" ALIGN="MIDDLE" BORDER="0"
|
|
SRC="img80.png"
|
|
ALT="$x[n]$"> is a sum of
|
|
impulses, one of height <IMG
|
|
WIDTH="29" HEIGHT="32" ALIGN="MIDDLE" BORDER="0"
|
|
SRC="img682.png"
|
|
ALT="$x[0]$">, the next one occurring one sample later
|
|
and having height <IMG
|
|
WIDTH="29" HEIGHT="32" ALIGN="MIDDLE" BORDER="0"
|
|
SRC="img683.png"
|
|
ALT="$x[1]$">, and so on. Later, when the networks get
|
|
more complicated, we will move to using impulses as input signals to show
|
|
their time-domain behavior.
|
|
|
|
<P>
|
|
On the other hand, we can analyze the same network in the frequency domain
|
|
by considering a (complex-valued) test signal,
|
|
<BR><P></P>
|
|
<DIV ALIGN="CENTER">
|
|
<!-- MATH
|
|
\begin{displaymath}
|
|
X[n] = {Z^n}
|
|
\end{displaymath}
|
|
-->
|
|
|
|
<IMG
|
|
WIDTH="74" HEIGHT="28" BORDER="0"
|
|
SRC="img684.png"
|
|
ALT="\begin{displaymath}
|
|
X[n] = {Z^n}
|
|
\end{displaymath}">
|
|
</DIV>
|
|
<BR CLEAR="ALL">
|
|
<P></P>
|
|
where Z has unit magnitude and argument <IMG
|
|
WIDTH="14" HEIGHT="13" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img27.png"
|
|
ALT="$\omega $">. We already know that the
|
|
output is another complex sinusoid with the same frequency, that is,
|
|
<BR><P></P>
|
|
<DIV ALIGN="CENTER">
|
|
<!-- MATH
|
|
\begin{displaymath}
|
|
H {Z^N}
|
|
\end{displaymath}
|
|
-->
|
|
|
|
<IMG
|
|
WIDTH="37" HEIGHT="24" BORDER="0"
|
|
SRC="img685.png"
|
|
ALT="\begin{displaymath}
|
|
H {Z^N}
|
|
\end{displaymath}">
|
|
</DIV>
|
|
<BR CLEAR="ALL">
|
|
<P></P>
|
|
for some complex number <IMG
|
|
WIDTH="18" HEIGHT="14" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img25.png"
|
|
ALT="$H$"> (which we want to find). So we write the output
|
|
directly as the sum of the input and its delayed copy:
|
|
<BR><P></P>
|
|
<DIV ALIGN="CENTER">
|
|
<!-- MATH
|
|
\begin{displaymath}
|
|
{Z^n} + {Z^{-d}} {Z^n} = (1 + {Z^{-d}}) {Z^n}
|
|
\end{displaymath}
|
|
-->
|
|
|
|
<IMG
|
|
WIDTH="200" HEIGHT="28" BORDER="0"
|
|
SRC="img686.png"
|
|
ALT="\begin{displaymath}
|
|
{Z^n} + {Z^{-d}} {Z^n} = (1 + {Z^{-d}}) {Z^n}
|
|
\end{displaymath}">
|
|
</DIV>
|
|
<BR CLEAR="ALL">
|
|
<P></P>
|
|
and find by inspection that:
|
|
<BR><P></P>
|
|
<DIV ALIGN="CENTER">
|
|
<!-- MATH
|
|
\begin{displaymath}
|
|
H = 1 + {Z^{-d}}
|
|
\end{displaymath}
|
|
-->
|
|
|
|
<IMG
|
|
WIDTH="92" HEIGHT="25" BORDER="0"
|
|
SRC="img687.png"
|
|
ALT="\begin{displaymath}
|
|
H = 1 + {Z^{-d}}
|
|
\end{displaymath}">
|
|
</DIV>
|
|
<BR CLEAR="ALL">
|
|
<P></P>
|
|
We can understand the frequency-domain behavior of this delay network
|
|
by studying how the complex number <IMG
|
|
WIDTH="18" HEIGHT="14" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img25.png"
|
|
ALT="$H$"> varies as a function of the
|
|
angluar frequency <IMG
|
|
WIDTH="14" HEIGHT="13" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img27.png"
|
|
ALT="$\omega $">. We are especially interested in its argument and
|
|
magnitude--which tell us the relative phase and amplitude of the sinusoid
|
|
that comes out. We will work this example out in detail to show how the
|
|
arithmetic of complex numbers can predict what happens when sinusoids are
|
|
combined additively.
|
|
|
|
<P>
|
|
|
|
<DIV ALIGN="CENTER"><A ID="fig07.05"></A><A ID="8344"></A>
|
|
<TABLE>
|
|
<CAPTION ALIGN="BOTTOM"><STRONG>Figure 7.5:</STRONG>
|
|
Analysis, in the complex plane, of the frequency-domain behavior of
|
|
the delay network of Figure <A HREF="#fig07.03">7.3</A>. The complex number <IMG
|
|
WIDTH="15" HEIGHT="14" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img20.png"
|
|
ALT="$Z$"> encodes
|
|
the frequency of the input. The delay line output is the input times
|
|
<IMG
|
|
WIDTH="34" HEIGHT="16" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img24.png"
|
|
ALT="$Z^-d$">. The total (complex) gain is <IMG
|
|
WIDTH="18" HEIGHT="14" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img25.png"
|
|
ALT="$H$">. We find the magnitude and argument
|
|
of <IMG
|
|
WIDTH="18" HEIGHT="14" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img25.png"
|
|
ALT="$H$"> by symmetrizing the sum, rotating it by <IMG
|
|
WIDTH="27" HEIGHT="32" ALIGN="MIDDLE" BORDER="0"
|
|
SRC="img26.png"
|
|
ALT="$d/2$"> times the
|
|
angular frequency of the input.</CAPTION>
|
|
<TR><TD><IMG
|
|
WIDTH="355" HEIGHT="207" BORDER="0"
|
|
SRC="img688.png"
|
|
ALT="\begin{figure}\psfig{file=figs/fig07.05.ps}\end{figure}"></TD></TR>
|
|
</TABLE>
|
|
</DIV>
|
|
|
|
<P>
|
|
Figure <A HREF="#fig07.05">7.5</A> shows the result, in the complex plane, when the
|
|
quantities <IMG
|
|
WIDTH="11" HEIGHT="13" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img262.png"
|
|
ALT="$1$"> and <IMG
|
|
WIDTH="32" HEIGHT="17" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img676.png"
|
|
ALT="${Z^{-d}}$"> are combined additively. To add complex numbers
|
|
we add their real and complex parts separately. So the complex number <IMG
|
|
WIDTH="11" HEIGHT="13" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img262.png"
|
|
ALT="$1$"> (real
|
|
part <IMG
|
|
WIDTH="11" HEIGHT="13" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img262.png"
|
|
ALT="$1$">, imaginary part <IMG
|
|
WIDTH="11" HEIGHT="13" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img179.png"
|
|
ALT="$0$">) is added coordinate-wise to the complex
|
|
number <IMG
|
|
WIDTH="32" HEIGHT="17" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img676.png"
|
|
ALT="${Z^{-d}}$"> (real part <!-- MATH
|
|
$\cos(-d \omega)$
|
|
-->
|
|
<IMG
|
|
WIDTH="68" HEIGHT="32" ALIGN="MIDDLE" BORDER="0"
|
|
SRC="img689.png"
|
|
ALT="$\cos(-d \omega)$">, imaginary part
|
|
<!-- MATH
|
|
$\sin(-d \omega)$
|
|
-->
|
|
<IMG
|
|
WIDTH="66" HEIGHT="32" ALIGN="MIDDLE" BORDER="0"
|
|
SRC="img690.png"
|
|
ALT="$\sin(-d \omega)$">). This is shown graphically by making a parallelogram,
|
|
with corners at the origin and at the two points to be added, and whose
|
|
fourth corner is the sum <IMG
|
|
WIDTH="18" HEIGHT="14" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img25.png"
|
|
ALT="$H$">.
|
|
|
|
<P>
|
|
As the figure shows, the result can be understood by symmetrizing it about
|
|
the real axis: instead of <IMG
|
|
WIDTH="11" HEIGHT="13" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img262.png"
|
|
ALT="$1$"> and <IMG
|
|
WIDTH="32" HEIGHT="17" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img676.png"
|
|
ALT="${Z^{-d}}$">, it's easier to sum the
|
|
quantities <IMG
|
|
WIDTH="35" HEIGHT="17" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img691.png"
|
|
ALT="${Z^{d/2}}$"> and <IMG
|
|
WIDTH="45" HEIGHT="17" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img692.png"
|
|
ALT="${Z^{-d/2}}$">, because they are symmetric about
|
|
the real (horizontal) axis. (Strictly speaking, we haven't properly defined
|
|
the quantities <IMG
|
|
WIDTH="35" HEIGHT="17" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img691.png"
|
|
ALT="${Z^{d/2}}$"> and <IMG
|
|
WIDTH="45" HEIGHT="17" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img692.png"
|
|
ALT="${Z^{-d/2}}$">; we are using those expressions
|
|
to denote unit
|
|
complex numbers whose arguments are half those of <IMG
|
|
WIDTH="22" HEIGHT="17" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img693.png"
|
|
ALT="$Z^d$"> and <IMG
|
|
WIDTH="32" HEIGHT="17" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img694.png"
|
|
ALT="$Z^{-d}$">, so
|
|
that squaring them would give <IMG
|
|
WIDTH="22" HEIGHT="17" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img695.png"
|
|
ALT="${Z^{d}}$"> and <IMG
|
|
WIDTH="32" HEIGHT="17" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img676.png"
|
|
ALT="${Z^{-d}}$">.)
|
|
We rewrite the gain as:
|
|
<BR><P></P>
|
|
<DIV ALIGN="CENTER">
|
|
<!-- MATH
|
|
\begin{displaymath}
|
|
H = {Z^{-d/2}} ({Z^{d/2}} + {Z^{-d/2}})
|
|
\end{displaymath}
|
|
-->
|
|
|
|
<IMG
|
|
WIDTH="183" HEIGHT="28" BORDER="0"
|
|
SRC="img696.png"
|
|
ALT="\begin{displaymath}
|
|
H = {Z^{-d/2}} ({Z^{d/2}} + {Z^{-d/2}})
|
|
\end{displaymath}">
|
|
</DIV>
|
|
<BR CLEAR="ALL">
|
|
<P></P>
|
|
The first term is a phase shift of <IMG
|
|
WIDTH="50" HEIGHT="32" ALIGN="MIDDLE" BORDER="0"
|
|
SRC="img697.png"
|
|
ALT="$-d \omega / 2$">. The second term is
|
|
best understood in rectangular form:
|
|
<BR><P></P>
|
|
<DIV ALIGN="CENTER">
|
|
<!-- MATH
|
|
\begin{displaymath}
|
|
{Z^{d/2}} + {Z^{-d/2}}
|
|
\end{displaymath}
|
|
-->
|
|
|
|
<IMG
|
|
WIDTH="92" HEIGHT="25" BORDER="0"
|
|
SRC="img698.png"
|
|
ALT="\begin{displaymath}
|
|
{Z^{d/2}} + {Z^{-d/2}}
|
|
\end{displaymath}">
|
|
</DIV>
|
|
<BR CLEAR="ALL">
|
|
<P></P>
|
|
<BR><P></P>
|
|
<DIV ALIGN="CENTER">
|
|
<!-- MATH
|
|
\begin{displaymath}
|
|
= (\cos(\omega d / 2) + i \sin(\omega d / 2)) +
|
|
(\cos(\omega d / 2) - i \sin(\omega d / 2))
|
|
\end{displaymath}
|
|
-->
|
|
|
|
<IMG
|
|
WIDTH="384" HEIGHT="28" BORDER="0"
|
|
SRC="img699.png"
|
|
ALT="\begin{displaymath}
|
|
= (\cos(\omega d / 2) + i \sin(\omega d / 2)) +
|
|
(\cos(\omega d / 2) - i \sin(\omega d / 2))
|
|
\end{displaymath}">
|
|
</DIV>
|
|
<BR CLEAR="ALL">
|
|
<P></P>
|
|
<BR><P></P>
|
|
<DIV ALIGN="CENTER">
|
|
<!-- MATH
|
|
\begin{displaymath}
|
|
= 2 \cos(\omega d / 2)
|
|
\end{displaymath}
|
|
-->
|
|
|
|
<IMG
|
|
WIDTH="93" HEIGHT="28" BORDER="0"
|
|
SRC="img700.png"
|
|
ALT="\begin{displaymath}
|
|
= 2 \cos(\omega d / 2)
|
|
\end{displaymath}">
|
|
</DIV>
|
|
<BR CLEAR="ALL">
|
|
<P></P>
|
|
This real-valued quantity may be either positive or negative; its absolute value
|
|
gives the magnitude of the output:
|
|
<BR><P></P>
|
|
<DIV ALIGN="CENTER">
|
|
<!-- MATH
|
|
\begin{displaymath}
|
|
|H| = 2 |\cos(\omega d / 2) |
|
|
\end{displaymath}
|
|
-->
|
|
|
|
<IMG
|
|
WIDTH="129" HEIGHT="28" BORDER="0"
|
|
SRC="img701.png"
|
|
ALT="\begin{displaymath}
|
|
\vert H\vert = 2 \vert\cos(\omega d / 2) \vert
|
|
\end{displaymath}">
|
|
</DIV>
|
|
<BR CLEAR="ALL">
|
|
<P></P>
|
|
The quantity <IMG
|
|
WIDTH="26" HEIGHT="32" ALIGN="MIDDLE" BORDER="0"
|
|
SRC="img30.png"
|
|
ALT="$\vert H\vert$"> is called the
|
|
<A ID="7917"></A><I>gain</I>
|
|
of the delay network at the angular frequency <IMG
|
|
WIDTH="14" HEIGHT="13" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img27.png"
|
|
ALT="$\omega $">, and is graphed in
|
|
Figure <A HREF="#fig07.06">7.6</A>. The frequency-dependent gain of a delay network (that
|
|
is, the gain as a function of frequency) is called the network's
|
|
<A ID="7920"></A><I>frequency response</I>.
|
|
|
|
<P>
|
|
Since the network has greater gain at some frequencies
|
|
than at others, it may be considered as a
|
|
<A ID="7922"></A><I>filter</I>
|
|
that can be used to separate certain components of a sound from others.
|
|
Because of the shape of this particular gain expression as a function of
|
|
<IMG
|
|
WIDTH="14" HEIGHT="13" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img27.png"
|
|
ALT="$\omega $">, this kind of delay network
|
|
is called a (non-recirculating)
|
|
<A ID="7924"></A><I>comb filter</I>.
|
|
|
|
<P>
|
|
|
|
<DIV ALIGN="CENTER"><A ID="fig07.06"></A><A ID="8359"></A>
|
|
<TABLE>
|
|
<CAPTION ALIGN="BOTTOM"><STRONG>Figure 7.6:</STRONG>
|
|
Gain of the delay network of Figure <A HREF="#fig07.03">7.3</A>, shown as a function
|
|
of angular frequency <IMG
|
|
WIDTH="14" HEIGHT="13" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img27.png"
|
|
ALT="$\omega $">.</CAPTION>
|
|
<TR><TD><IMG
|
|
WIDTH="326" HEIGHT="137" BORDER="0"
|
|
SRC="img702.png"
|
|
ALT="\begin{figure}\psfig{file=figs/fig07.06.ps}\end{figure}"></TD></TR>
|
|
</TABLE>
|
|
</DIV>
|
|
|
|
<P>
|
|
The output of the network is a sum of two sinusoids of equal amplitude, and
|
|
whose phases differ by <IMG
|
|
WIDTH="22" HEIGHT="14" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img703.png"
|
|
ALT="$\omega d $">. The resulting frequency response agrees
|
|
with common sense: if the angular frequency <IMG
|
|
WIDTH="14" HEIGHT="13" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img27.png"
|
|
ALT="$\omega $"> is set so that an integer
|
|
number of periods fit into <IMG
|
|
WIDTH="11" HEIGHT="14" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img28.png"
|
|
ALT="$d$"> samples, i.e., if <IMG
|
|
WIDTH="14" HEIGHT="13" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img27.png"
|
|
ALT="$\omega $"> is a multiple of
|
|
<IMG
|
|
WIDTH="37" HEIGHT="32" ALIGN="MIDDLE" BORDER="0"
|
|
SRC="img704.png"
|
|
ALT="$2\pi/d$">, the output of the delay is exactly the same as the original signal,
|
|
and so the two combine to make an output with twice the original amplitude. On
|
|
the other hand, if for example we take <!-- MATH
|
|
$\omega = \pi/d$
|
|
-->
|
|
<IMG
|
|
WIDTH="61" HEIGHT="32" ALIGN="MIDDLE" BORDER="0"
|
|
SRC="img705.png"
|
|
ALT="$\omega = \pi/d$"> so that the delay is
|
|
half the period, then the delay output is out of phase and cancels the input
|
|
exactly.
|
|
|
|
<P>
|
|
This particular delay network has an interesting application: if we have a
|
|
periodic (or nearly periodic) incoming signal, whose fundamental frequency is
|
|
<IMG
|
|
WIDTH="14" HEIGHT="13" ALIGN="BOTTOM" BORDER="0"
|
|
SRC="img27.png"
|
|
ALT="$\omega $"> radians per sample, we can tune the comb filter so that the peaks in
|
|
the gain are aligned at even harmonics and the odd ones fall where the gain is
|
|
zero. To do this we choose <IMG
|
|
WIDTH="61" HEIGHT="32" ALIGN="MIDDLE" BORDER="0"
|
|
SRC="img706.png"
|
|
ALT="$d=\pi/\omega$">, i.e., set the delay time to exactly
|
|
one half period of the incoming signal. In this way we get a new signal whose
|
|
harmonics are <!-- MATH
|
|
$2\omega, 4\omega, 6\omega, \ldots$
|
|
-->
|
|
<IMG
|
|
WIDTH="98" HEIGHT="29" ALIGN="MIDDLE" BORDER="0"
|
|
SRC="img707.png"
|
|
ALT="$2\omega, 4\omega, 6\omega, \ldots$">, and so it now has a new
|
|
fundamental frequency at twice the original one. Except for a factor of two,
|
|
the amplitudes of the remaining harmonics still follow the spectral envelope of
|
|
the original sound. So we have a tool now for raising the pitch of an incoming
|
|
sound by an octave without changing its spectral envelope. This octave doubler
|
|
is the reverse of the octave divider introduced back in Chapter
|
|
<A HREF="node75.html#chapter-modulation">5</A>.
|
|
|
|
<P>
|
|
The time and frequency domains offer complementary ways of looking
|
|
at the same delay network. When the delays inside the network are smaller than
|
|
the ear's ability to resolve events in time--less than about 20
|
|
milliseconds--the time domain picture becomes less relevant to our
|
|
understanding of the delay network, and we turn mostly to the frequency-domain
|
|
picture. On the other hand, when delays are greater than about 50
|
|
msec, the peaks and valleys of plots showing gain versus frequency (such as
|
|
that of Figure <A HREF="#fig07.06">7.6</A>) crowd so closely together that
|
|
the frequency-domain view becomes less important. Both are nonetheless valid
|
|
over the entire range of possible delay times.
|
|
|
|
<P>
|
|
<HR>
|
|
<!--Navigation Panel-->
|
|
<A ID="tex2html2171"
|
|
HREF="node109.html">
|
|
<IMG WIDTH="37" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="next"
|
|
SRC="next.png"></A>
|
|
<A ID="tex2html2165"
|
|
HREF="node104.html">
|
|
<IMG WIDTH="26" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="up"
|
|
SRC="up.png"></A>
|
|
<A ID="tex2html2159"
|
|
HREF="node107.html">
|
|
<IMG WIDTH="63" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="previous"
|
|
SRC="prev.png"></A>
|
|
<A ID="tex2html2167"
|
|
HREF="node4.html">
|
|
<IMG WIDTH="65" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="contents"
|
|
SRC="contents.png"></A>
|
|
<A ID="tex2html2169"
|
|
HREF="node201.html">
|
|
<IMG WIDTH="43" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="index"
|
|
SRC="index.png"></A>
|
|
<BR>
|
|
<B> Next:</B> <A ID="tex2html2172"
|
|
HREF="node109.html">Recirculating delay networks</A>
|
|
<B> Up:</B> <A ID="tex2html2166"
|
|
HREF="node104.html">Time shifts and delays</A>
|
|
<B> Previous:</B> <A ID="tex2html2160"
|
|
HREF="node107.html">Time shifts and phase</A>
|
|
<B> <A ID="tex2html2168"
|
|
HREF="node4.html">Contents</A></B>
|
|
<B> <A ID="tex2html2170"
|
|
HREF="node201.html">Index</A></B>
|
|
<!--End of Navigation Panel-->
|
|
<ADDRESS>
|
|
Miller Puckette
|
|
2006-12-30
|
|
</ADDRESS>
|
|
</BODY>
|
|
</HTML>
|
|
|