TCP/IP data vs BDF data

Post Reply
kkalogia
Posts: 7
Joined: Thu Jun 28, 2012 9:55 am

TCP/IP data vs BDF data

Post by kkalogia » Mon Aug 13, 2012 3:01 pm

Hello,
I would like to pose a question regrading the data recieved from actiview through TCP/IP connection and their corresponding .bdf file.

For a given recording, the .bdf data extracted from actiview and the 2D array received from the TCP/IP connection (stocked via own program) are identical?
Or there is a further transformation step before the creation of the .bdf file?

Thank you in advance for your time.

Best Regards,

Konstantina

Coen
Site Admin
Posts: 925
Joined: Fri Mar 26, 2004 7:00 pm
Location: Amsterdam, Netherlands
Contact:

Re: TCP/IP data vs BDF data

Post by Coen » Mon Aug 13, 2012 3:15 pm

The data format received via TCP/IP differs from the BDF format, see viewtopic.php?f=2&t=196

Best regards, Coen (BioSemi)

kkalogia
Posts: 7
Joined: Thu Jun 28, 2012 9:55 am

Re: TCP/IP data vs BDF data

Post by kkalogia » Tue Aug 14, 2012 11:05 am

Dear Coen,
Thank you for your answer.
So after the transformation of the data coming via TCP into a 2D array (samples X signals)
with the way described in the post you sent me ,
we have the same values as in the bdf file(that can be tested via the digital and physical max and min values I suppose...)?
Or there is any other transformation needed to pass to the bdf data?
Thank you
Best Regards,
Konstantina

Coen
Site Admin
Posts: 925
Joined: Fri Mar 26, 2004 7:00 pm
Location: Amsterdam, Netherlands
Contact:

Re: TCP/IP data vs BDF data

Post by Coen » Tue Aug 14, 2012 11:37 am

You are correct, after the transformation to a 2D array you have the same data as in the BDF file (use also the same 3-bytes/24-bit to 32-bit Integer conversion). Actiview streams exactly the same 24-bit samples to both BDF file and via TCP/IP.

Best regards, Coen (BioSemi)

kkalogia
Posts: 7
Joined: Thu Jun 28, 2012 9:55 am

Re: TCP/IP data vs BDF data

Post by kkalogia » Wed Aug 29, 2012 1:11 pm

Dear Coen,
I finallly managed to transform the data coming from the TCP/IP to the ones that correspond to the bdf file.

Just another question now.

Some time ago I did some recordings with the same program as I do now.

In the header of the .bdf file I found the max and min real and digital values(where I am based for the computation of the real values).
I realised though that for example for the respiration channel the min_real_value was -2621445(previous recordings)
while now is -2097152.
The only difference from the previous recordings as regards to the new ones, is the biosemi. Before it was the MK1 and now is the MK2.

Is it possible that this factor makes the difference between the values or is another parameter?

I am asking this question because I want to compare the values obtained from the two recordings periods.

Thank you

Konstantina

Coen
Site Admin
Posts: 925
Joined: Fri Mar 26, 2004 7:00 pm
Location: Amsterdam, Netherlands
Contact:

Re: TCP/IP data vs BDF data

Post by Coen » Wed Aug 29, 2012 2:29 pm

The different real min/max values depend on the type of respiration belt that is used (-262144/262143 for the older Nihon Kohden belt, -2097152/2097151 for the current SleepSense belt). You cannot compare the absolute values between these two different types of belts.

Best regards, Coen (BioSemi)

kkalogia
Posts: 7
Joined: Thu Jun 28, 2012 9:55 am

Re: TCP/IP data vs BDF data

Post by kkalogia » Mon Sep 03, 2012 1:43 pm

Dear Coen,
I am sorry for my continuous questions but I really dont understand what is going on.

I have the Nihon Kohden belt and I had the same one before also.
But here is a table with the comparison of the values at the old and the new recordings.

Labels Phys_Min_old recordings Phys_Min_new recordings
GSR1 262143 262143
GSR2 262143 262143
Erg1 524286 2097151
Erg2 524286 2097151
Resp 262143 2097151
Plet 4096000 4095875
Temp 8388 8388

I really don't get it why I have such a difference.

Thank you for your time.
Konstantina

Coen
Site Admin
Posts: 925
Joined: Fri Mar 26, 2004 7:00 pm
Location: Amsterdam, Netherlands
Contact:

Re: TCP/IP data vs BDF data

Post by Coen » Mon Sep 03, 2012 1:58 pm

The values of the new recording are correct. The older system may have had channels with different gain factors.

Best regards, Coen (BioSemi)

Post Reply