Page 1 of 1

Biosemi & Actiview values--peripheral signals

Posted: Thu Jul 19, 2012 11:02 am
by kkalogia
hello,
I am new to BioSemi and I have some basic questions that I would appreciate if you can help me.
I am trying to use my own program to acquire the data form the biosemi and then send them to actiview.
I would like to know though, which exactly transformations are performed in Actiview in order to pass from the digital data with dig.max=8388607 and dig.min=-8388608, to the values in actiview with phys.max=8388 and phys.min=-8389, and then from those how we pass to the normal values, for example for temperature ( normal for epidermal temperature==around 32 celcius), that are depicted as an output to actiview.

I am working on an affective game and I am recording peripheral signals.

My goal is to aquire the data with biosemi pass it to actiview through tcp/ip connection and then according to the values acquired that indicate stress or boredom adjust the level of the game.

I know that my question is basic, but i need some help to understand!

Thank you.
Konstantina

Re: Biosemi & Actiview values--peripheral signals

Posted: Thu Jul 19, 2012 2:55 pm
by Coen
Via TCP/IP you receive 24-bit samples (range is from -8388608 to -8388607).

Use the following scaling factors:

- Electrode channels: LSB = 31.25 nV (divide the 24-bit number by 32 to get a value in microVolt)
- Temperature channel: LSB = 1 millidegree Celsius (divide the 24-bit number by 1000 to get a value in degree Celsius)

Best regards Coen (BioSemi)

Re: Biosemi & Actiview values--peripheral signals

Posted: Thu Jul 19, 2012 3:07 pm
by kkalogia
Thank you so much for your instant reply.
One more question.
What for the GSR electrode?
I suppose I also have to divide with something and then proceed as described here?
viewtopic.php?f=7&t=400&p=1011&hilit=he ... alue#p1011
Thank you.
Konstantina

Re: Biosemi & Actiview values--peripheral signals

Posted: Thu Jul 19, 2012 4:48 pm
by Coen
Use the following scaling factor for the GSR channel:

LSB = 31.25 picoSiemens (divide the 24-bit number by 32 to get a value in nanoSiemens)

Best regards, Coen (BioSemi)

Re: Biosemi & Actiview values--peripheral signals

Posted: Fri Aug 03, 2012 12:22 pm
by kkalogia
Dear Coen,
Thank you for your reply.
I just need some further explanations and your help.

As explained before I use the .bdf data to control my system which is a Tetris game.

The control of the system is going to happen online, that means that I have to compute whether the player is stressed or bored according to his physiological values and then adjust the system according to his state.
Therefore I need to be very precise about the values so that I can compare them to the physiological values that exist in the literature and then proceed to the adjustment of the game’s level.

If I understand well, the relation between the digital values and the physical values is a linear system.
So for a given digital value we can compute the physiological value.
The solution of this system, as I computed, gives an equation and for example for the temperature the solution is:

Real value=-0.5+ 1000*digital value.

In your post before, you suggested that for temperature I should divide all the values with 1000 though, in the solution I computed there is a difference of -0.5 which in case of body temperature is an important difference.

Please tell me If I am wrong on that and what should I do for the real value computation.

Thank you in advance,
Konstantina

Re: Biosemi & Actiview values--peripheral signals

Posted: Fri Aug 03, 2012 12:33 pm
by Coen
If you need a very accurate temperature readout, then you should calibrate the system with a reference thermometer.

Best regards, Coen