To get to the bottom of this, without expensive signal analyzers,
perhaps you should try to do the speed measurement in the basestation
itself and send a message with the cumulative values over the UART
once in a while. That would at least eliminate the suspected bottleneck
and see if there is any reason for heroic efforts to get higher BW.
This link was graciously supplied a couple days ago:
http://www.moteiv.com/community/Change_the_default_UART_baud_rate
but is for Tmotes...there might be something similar for mica's,
so again, double check that you have really changed the baud rate.
If you can get a scope on the Rx line you should see some change.
Changing the speed of the UISP download is probably unnecessary
since that's not used during 'normal' operation. I'm a little
surprised that it worked because, I'd think that you need to change
something on the MIB end as well.
Also try eliminating the SerialForwarder and just use Listen
or something directly connected to the serial port. One less
piece of code in the way. And you might want to consider using
the e'net base station board (MIB520 or something, I forget).
As to how to extract part of the radio data for UART forwarding,
you want to remove the current UART stuff from TOSBase and replace
it with the lower level UARTM interface which doesn't have any
software protocol layer. You will also need to modify the PC end
to just get raw bytes w/o protocol checks and crcs and all. But
you may end up needing some kind of message sync behavior anyway,
so I don't know how worthwhile it would be.
MS
Jeong Kwanhee wrote:
Hello, all!!
I did change some files in order to execute 115200 serial speed.
First of all, I added –dspeed=115200 below :
uisp -dprog=mib510 -dspeed=115200 -dserial=COM1 -dpart=ATmega128
--wr_fuse_e=ff --erase --upload if=build/micaz/main.srec
And I added makerule.
Second, I did change to outp(7, UBRR0L); in platform/mica2/HPLUART0M.nc.
Because I know both micaz and mica2 use same file. Is right??
Finally, I did test using java application.
java net.tinyos.sf.SerialForwarder -comm [EMAIL PROTECTED]:
<mailto:[EMAIL PROTECTED]:57600>115200
As result this test, I gained only 57kbps performance.
Is this method wrong???? Help me everyone!!!
------------------------------------------------------------------------
_______________________________________________
Tinyos-help mailing list
[email protected]
https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
_______________________________________________
Tinyos-help mailing list
[email protected]
https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
