<html>
<head>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<meta name=Generator content="Microsoft Word 10 (filtered)">
<style>
<!--
/* Font Definitions */
@font-face
{font-family:Tahoma;
panose-1:2 11 6 4 3 5 4 4 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0cm;
margin-bottom:.0001pt;
font-size:12.0pt;
font-family:"Times New Roman";}
a:link, span.MsoHyperlink
{color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{color:purple;
text-decoration:underline;}
span.StyleCourrierlectronique17
{font-family:Arial;
color:navy;}
@page Section1
{size:612.0pt 792.0pt;
margin:72.0pt 90.0pt 72.0pt 90.0pt;}
div.Section1
{page:Section1;}
-->
</style>
</head>
<body lang=FR-CA link=blue vlink=purple>
<div class=Section1>
<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-CA
style='font-size:10.0pt;font-family:Arial;color:navy'>The duration of the pulse
code is directly responsible for this. Whenever the pulse spans two sampling
bins, two consecutive identical codes appear in the event channel. There are
also problems with codes that are emitted in close succession. If the signal
is decimated and two codes were presented in close succession, then the code
stored in the Biosemi file is the logical AND of the bits that were ON in any
of the bins that are merged together. The results depend on the exact timing
of the bins with respect to the duration of the emitted codes.</span></font></p>
<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-CA
style='font-size:10.0pt;font-family:Arial;color:navy'> </span></font></p>
<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-CA
style='font-size:10.0pt;font-family:Arial;color:navy'>My solution, when reading
the sequence of coded events is to remove codes that (a) are identical to the
previous code and (b) follow it by just one time step.</span></font></p>
<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-CA
style='font-size:10.0pt;font-family:Arial;color:navy'> </span></font></p>
<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-CA
style='font-size:10.0pt;font-family:Arial;color:navy'>If codes must be sent in
close succession, e.g. code 254 (Pause OFF) to start the acquisition and a
condition code, I now make sure to insert a reasonable delay between the two
(say 5 ms). In our first Biosemi application, with decimation, there was no
delay and we sometimes had three different codes at three consecutive times,
the first was 254, the last was our event code and the second was the logical bit-AND
of the two. The middle code had to be replaced by the third (event code) and
the third code deleted (since the event code appeared during the second time
bin). In a few less favourable cases, only code 254 and the logical AND were
present, such that the event code could not be identified unambiguously and was
therefore lost.</span></font></p>
<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-CA
style='font-size:10.0pt;font-family:Arial;color:navy'> </span></font></p>
<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-CA
style='font-size:10.0pt;font-family:Arial;color:navy'>An important
consideration, also, in sending codes to the Biosemi system, is to wait 1 s
after the last useful part of the recording, before sending out code 255 (Pause
ON), because any 1 s epoch currently incomplete when this code is received is
not saved (except perhaps in the very latest version of Actiview).</span></font></p>
<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-CA
style='font-size:10.0pt;font-family:Arial;color:navy'> </span></font></p>
<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-CA
style='font-size:10.0pt;font-family:Arial;color:navy'>Andre Achim</span></font></p>
<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-CA
style='font-size:10.0pt;font-family:Arial;color:navy'> </span></font></p>
<p class=MsoNormal style='margin-left:35.4pt'><font size=2 face=Tahoma><span
lang=FR style='font-size:10.0pt;font-family:Tahoma'>-----Message d'origine-----<br>
<b><span style='font-weight:bold'>De :</span></b>
eeglablist-bounces@sccn.ucsd.edu [mailto:eeglablist-bounces@sccn.ucsd.edu] <b><span
style='font-weight:bold'>De la part de</span></b> Jim Kroger<br>
<b><span style='font-weight:bold'>Envoyé :</span></b> 14 février 2005
23:14<br>
<b><span style='font-weight:bold'>À :</span></b> eeglablist<br>
<b><span style='font-weight:bold'>Objet :</span></b> [Eeglablist] double
event markers from Biosemi .bdf files</span></font></p>
<p class=MsoNormal style='margin-left:35.4pt'><font size=3
face="Times New Roman"><span style='font-size:12.0pt'> </span></font></p>
<p class=MsoNormal style='margin-right:0cm;margin-bottom:12.0pt;margin-left:
35.4pt'><font size=3 face="Times New Roman"><span style='font-size:12.0pt'>Hello,
I am wondering if anybody else has experienced this problem. We use a Biosemi
128-channel.<br>
Channel 129 has event markers. <br>
<br>
When using this channel to mark event times in EEGLAB, we have some kind of
ring or artifact that causes EEGLAB to mark two events instead of one. We also
try loading this into EMSE, and do not have the problem.<br>
<br>
We're having a bit of trouble understanding why this happens. Although we can
imagine ways to program around the problem, it would be best if we understood
it.<br>
<br>
Many thanks for any pointers.<br>
<br>
Jim<br>
<br>
<br>
</span></font></p>
</div>
</body>
</html>