Arpad,
If a model does not support training, then it should just leave out all of
the BCI parameters. Another way of doing this is making BCI_State a single
value "Off".
Walter
Walter Katz
<mailto:wkatz@xxxxxxxxxx> wkatz@xxxxxxxxxx
Office 978.461-0449 x 133
Mobile 720.417-3762
From: ibis-macro-bounce@xxxxxxxxxxxxx <ibis-macro-bounce@xxxxxxxxxxxxx> On
Behalf Of Muranyi, Arpad
Sent: Monday, October 21, 2019 2:22 PM
To: IBIS-ATM <ibis-macro@xxxxxxxxxxxxx>
Subject: [ibis-macro] Re: BCI Statistical BIRD AMI_Init or AMI_GetWave
Walter,
For the BCI_Training_Model parameter you define three values, "Init",
"GetWave" or "Dual"
and the default when the parameter is missing is "GetWave". Shouldn't there
be a way to
state that the model does NOT support any training?
Thanks,
Arpad
=======================================================================
From: ibis-macro-bounce@xxxxxxxxxxxxx
<mailto:ibis-macro-bounce@xxxxxxxxxxxxx>
[mailto:ibis-macro-bounce@xxxxxxxxxxxxx] On Behalf Of Walter Katz
Sent: Friday, October 18, 2019 1:57 PM
To: IBIS-ATM <ibis-macro@xxxxxxxxxxxxx <mailto:ibis-macro@xxxxxxxxxxxxx> >
Subject: [ibis-macro] BCI Statistical BIRD AMI_Init or AMI_GetWave
All,
I will proceed in the IBIS-ATM meeting using AMI_Init. If Ambrish would
prefer using AMI_GetWave for this purpose I suggest that he write an
alternative BIRD that we can evaluate.
Using AMI_Init for BCI statistical analysis allows the re-use of a the
arguments to AMI_Init during statistical training. I have made a change in
draft 3 (enclosed) that the **AMI_Parameters_out string contain a branch
(BCI .). The data in this branch shall be defined by the BCI Protocol, and
will be a much more efficient way of communicating data from one call to
AMI_Init to the next call to AMI_Init during statistical training. The
*AMI_Parameters_in contain the value of **AMI_Parameters_out from the
previous call to AMI_Init.
Walter
Walter Katz
<mailto:wkatz@xxxxxxxxxx> wkatz@xxxxxxxxxx
Office 978.461-0449 x 133
Mobile 720.417-3762