<?xml version="1.0" encoding="utf-8" standalone="yes"?><html xmlns="http://www.w3.org/1999/xhtml">
<head>
<title>Simulation results</title>
<meta content="text/html; charset=utf-8" http-equiv="Content-Type" />
<meta content="text/css" http-equiv="Content-Style-Type" />
<script src="NED?action=retrieve&amp;identifier=dn00188761&amp;edition=1&amp;language=none&amp;coverage=global&amp;encoding=javascript&amp;component=data&amp;item=data" type="text/javascript" xml:space="preserve">
</script>
<?conversion name="Pub2XHTML" version="2.0" source="urn:mars:dn02259109:5:en:global:publishing_online_1_0:data:data:*:*:*" date="2005-02-28T16:16:52Z"?>
<link href="NED?action=retrieve&amp;identifier=dn01158124&amp;edition=1&amp;language=en&amp;coverage=global&amp;encoding=css&amp;component=data&amp;item=data" rel="stylesheet" title="Nokia Networks default online style" type="text/css" />
<meta content="draft" name="status" />
<?conversion name="SplitHTML" version="1.9" source="urn:mars:dn02259109:5:en:global:publishing_online_1_0:data:data:40:*:*" date="2005-02-28T16:16:52Z"?>
</head>
<body lang="en" xml:lang="en">
<p><a name="dec1221452213" shape="rect"></a>
<a name="ned_2" shape="rect"></a></p>

<div class="div">
<table width="100%">
<tr>
<td align="center" colspan="1" rowspan="1" width="20%"><?NED5 annotation?>
<br clear="none" /><?NED5 printview?>
</td>
<td align="center" colspan="1" rowspan="1" width="20%"></td>
<td align="center" colspan="1" rowspan="1" width="20%"></td>
<td align="center" colspan="1" rowspan="1" width="20%"><a href="NED?action=retrieve&amp;identifier=dn02259109&amp;edition=5&amp;language=en&amp;coverage=global&amp;encoding=xhtml_1_0&amp;component=data&amp;item=data&amp;pointer=ned_1#ned_1" onclick="sync(this);return true" shape="rect">EGPRS frequency planning </a></td>
<td align="center" colspan="1" rowspan="1" width="20%"><a href="NED?action=retrieve&amp;identifier=dn02259109_about&amp;edition=5&amp;language=en&amp;coverage=global&amp;encoding=xhtml_1_0&amp;component=data&amp;item=data&amp;pointer=ned_top#ned_top" onclick="openSmallPopup(event,'NED?action=retrieve&amp;identifier=dn02259109_about&amp;edition=5&amp;language=en&amp;coverage=global&amp;encoding=xhtml_1_0&amp;component=data&amp;item=data&amp;pointer=ned_top#ned_top');return false" shape="rect" target="_blank"><img alt="Document information" border="0" class="Metainfo" src="NED?action=retrieve&amp;identifier=dn01158003&amp;edition=1&amp;language=en&amp;coverage=global&amp;encoding=gif&amp;component=data&amp;item=data" /></a></td>
</tr>
</table>


<div class="div">
<h2>Simulation results</h2>

<p>The figures below show the results of link-level simulations.
They give a good estimation of the data rates achievable with EDGE and of
the effect of channel environment and frequency hopping.  </p>


<a name="dec1621634451" shape="rect"></a>
<div>
<img alt="urn:mars:dn02262436:1:en:global:tiff:data:data" border="0" src="NED?action=retrieve&amp;identifier=dn02262436&amp;edition=1&amp;language=en&amp;coverage=global&amp;encoding=gif&amp;component=data&amp;item=data" /><p class="figure-caption">Figure: 8PSK TU3 non-hopping, impairments included</p>
</div>


<a name="dec162163733" shape="rect"></a>
<div>
<img alt="urn:mars:dn02262448:1:en:global:tiff:data:data" border="0" src="NED?action=retrieve&amp;identifier=dn02262448&amp;edition=1&amp;language=en&amp;coverage=global&amp;encoding=gif&amp;component=data&amp;item=data" /><p class="figure-caption">Figure: 8PSK TU3 ideal frequency hopping, impairments not included</p>
</div>


<p>In observing the maximum available data rates, it can be
seen that MCS-9 with incremental redundancy offers the highest throughputs
but the lowest C/Is. Consequently, there is no hopping gain (in fact there
is a small hopping ‘loss’ of up to 1.5 dB). This is because MCS-9 has no forward
error correction (<a href="NED?action=retrieve&amp;identifier=general_glossary&amp;edition=13&amp;language=en&amp;coverage=global&amp;encoding=xhtml_1_0&amp;component=data&amp;item=data&amp;pointer=id09440#id09440" onclick="openSmallPopup(event,'NED?action=retrieve&amp;identifier=general_glossary&amp;edition=13&amp;language=en&amp;coverage=global&amp;encoding=xhtml_1_0&amp;component=data&amp;item=data&amp;pointer=id09440#id09440');return false" shape="rect" target="_blank">FEC</a>) on the user data and, on average,
frequency hopping can actually increase block error ratio (<a href="NED?action=retrieve&amp;identifier=general_glossary&amp;edition=13&amp;language=en&amp;coverage=global&amp;encoding=xhtml_1_0&amp;component=data&amp;item=data&amp;pointer=id011207#id011207" onclick="openSmallPopup(event,'NED?action=retrieve&amp;identifier=general_glossary&amp;edition=13&amp;language=en&amp;coverage=global&amp;encoding=xhtml_1_0&amp;component=data&amp;item=data&amp;pointer=id011207#id011207');return false" shape="rect" target="_blank">BLER</a>)
because of the action of randomising the error distribution. The same applies
to MCS-7 and MCS-8, although for these the loss is closer to zero.</p>


<p>Another issue with MCS-9 is that it is the <a href="NED?action=retrieve&amp;identifier=general_glossary&amp;edition=13&amp;language=en&amp;coverage=global&amp;encoding=xhtml_1_0&amp;component=data&amp;item=data&amp;pointer=id09745#id09745" onclick="openSmallPopup(event,'NED?action=retrieve&amp;identifier=general_glossary&amp;edition=13&amp;language=en&amp;coverage=global&amp;encoding=xhtml_1_0&amp;component=data&amp;item=data&amp;pointer=id09745#id09745');return false" shape="rect" target="_blank">MCS</a>
with the highest susceptibility to errors because of impairments, as there
is no FEC to correct any errors that may occur. Consequently, it may be more
appropriate in some C/I intervals to select MCS-7 or MCS-8, which, although
having a slightly worse C/I in the ‘ideal’ case, might perform better in a
situation with impairments.</p>


<p>Nokia has performed simulations that indicate typical data
throughputs that might be achievable in practical networks.</p>


<p>The following static simulation results show the throughput
in a 3-sector site with 50% load and re-use pattern 3/9. </p>


<a name="mar22412647" shape="rect"></a>
<div>
<img alt="urn:mars:dn0448126:1:none:global:tiff:data:data" border="0" src="NED?action=retrieve&amp;identifier=dn0448126&amp;edition=1&amp;language=none&amp;coverage=global&amp;encoding=gif&amp;component=data&amp;item=data" /><p class="figure-caption">Figure: Data rates – re-use 3/9 load =50%</p>
</div>


<p>The new C/I target values resulting in a coverage are highly
dependent on frequency re-use factors. On the other hand, higher re-use factors
decrease the spectrum efficiency. A smaller re-use factor limits those higher
data rates on a smaller coverage near the base station and reduces to best
effort during the peak hours.</p>


<p>Expected real scenarios should demonstrate average data
rates of nearly 3 times the GPRS data rates (roughly the increase is because
of higher modulation scheme).</p>


<a name="dec1621640185" shape="rect"></a>
<div>
<img alt="urn:mars:dn02262475:1:en:global:cgm_fixed:data:data" border="0" src="NED?action=retrieve&amp;identifier=dn02262475&amp;edition=1&amp;language=en&amp;coverage=global&amp;encoding=gif&amp;component=data&amp;item=data" /><p class="figure-caption">Figure: Data performance vs. re-use and range</p>
</div>


<p>The following figures show the throughput from static simulations
for MCS-1 and MCS-5. The closer to the <a href="NED?action=retrieve&amp;identifier=general_glossary&amp;edition=13&amp;language=en&amp;coverage=global&amp;encoding=xhtml_1_0&amp;component=data&amp;item=data&amp;pointer=id03816#id03816" onclick="openSmallPopup(event,'NED?action=retrieve&amp;identifier=general_glossary&amp;edition=13&amp;language=en&amp;coverage=global&amp;encoding=xhtml_1_0&amp;component=data&amp;item=data&amp;pointer=id03816#id03816');return false" shape="rect" target="_blank">BTS</a>, the higher
the probability to achieve the maximum data rates. The system simulated is
close to a 7/21 re-use.</p>


<a name="dec1621641226" shape="rect"></a>
<div>
<img alt="urn:mars:dn02262487:1:en:global:tiff:data:data" border="0" src="NED?action=retrieve&amp;identifier=dn02262487&amp;edition=1&amp;language=en&amp;coverage=global&amp;encoding=gif&amp;component=data&amp;item=data" /><p class="figure-caption">Figure: MSC-1 performance </p>
</div>


<a name="dec1621642207" shape="rect"></a>
<div>
<img alt="urn:mars:dn02262506:1:en:global:tiff:data:data" border="0" src="NED?action=retrieve&amp;identifier=dn02262506&amp;edition=1&amp;language=en&amp;coverage=global&amp;encoding=gif&amp;component=data&amp;item=data" /><p class="figure-caption">Figure: MCS-5 performance </p>
</div>


<p>Control channel performance is also of primary concern
as it conditions the traffic channel's data rates. The figure below shows
the probability of error free reception of control blocks, for example, access
grants. The results show that, for the radio link control (<a href="NED?action=retrieve&amp;identifier=general_glossary&amp;edition=13&amp;language=en&amp;coverage=global&amp;encoding=xhtml_1_0&amp;component=data&amp;item=data&amp;pointer=id08942#id08942" onclick="openSmallPopup(event,'NED?action=retrieve&amp;identifier=general_glossary&amp;edition=13&amp;language=en&amp;coverage=global&amp;encoding=xhtml_1_0&amp;component=data&amp;item=data&amp;pointer=id08942#id08942');return false" shape="rect" target="_blank">RLC</a>)
block to work properly at a BLER of 5%, re-use factors of at least 3/9 are
needed. Depending on the real layout of base stations, a higher frequency
re-use might be required.</p>


<a name="dec1621644179" shape="rect"></a>
<div>
<img alt="urn:mars:dn02262518:1:en:global:cgm_fixed:data:data" border="0" src="NED?action=retrieve&amp;identifier=dn02262518&amp;edition=1&amp;language=en&amp;coverage=global&amp;encoding=gif&amp;component=data&amp;item=data" /><p class="figure-caption">Figure: Control channel performance vs. range </p>
</div>
</div>
</div>
</body>
</html>