[sim-wg] RE: Incorporation of Mica2 simulation

[Venkatesh S.]

Hello everyone:

This is in response to Phil's mail.  

These are the things that we have done for Mica2 simulation.

1. Noise model for CC1000 radio

The basic idea is to provide users a noise model that works close to actual running code in sensor nodes.  
We use Phil's paper - "Improving wireless simulation through noise modelling" and the associated CPM code.
In this implementation we sampled the outdoor environment noise and statistically charecterized using both Naïve and CPM models.  The obtained noise samples are fed to the higher layer.  

2. CSMA - CCA:  The exisiting T2 code is based on SNR-PRR curve for CC2420.    
We tried to approach this for CC1000 by including the ADC0_RSSI port sampling.  We introduced a "channel entity" which provides information to each node about "busy" channel or otherwise.  Using this, we successfully solved the packet drop problem caused by MAC "ACK_WAIT" state at transmitters.  
My assessment is this CSMA-CCA approach probably computes transmit/receive energy consumption closer to actual hardware since the radio is byte level SPI.  


Phil, I couldnt get you in saying "wasn't able to test the connectivity".  Could you be a bit clearer? 

--
Cheers, 
Venkatesh S



-----Original Message-----
From: Philip Levis [mailto:pal at cs.stanford.edu]
Sent: Wed 6/20/2007 5:52 AM
To: Venkatesh S.
Cc: simwg; Chad Metcalf
Subject: RE: [sim-wg] RE: Incorporation of Mica2 simulation
 
I was able to get it to compile fine but wasn't able to test
connectivity. My one comment, looking at the code, is that it might be
useful to leverage the existing SNR based TOSSIM code for clear channel
assessment and interference.

Phil





-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.millennium.berkeley.edu/pipermail/sim-wg/attachments/20070704/425257b9/attachment.htm