[Tinyos-help] "Gray area" in simulated sensor network is different than "Gray area" in real sensor network!

[Philip Levis]

On Jan 28, 2009, at 9:34 AM, Nouha Baccour wrote:

> Hi all,
>
> I'm using TOSSIM (T2) simulator.
>
> Based on extensive simulations, I draw the curve PRR (Packet  
> reception rate) according to distance.
>
> This curve shows three reception regions (connected, gray and  
> disconnected), as it is observed in actual sensor networks.
>
> Inside the gray (called also transitional) area, the PRR is expected  
> to fluctuate between 0.15 and 0.85 and it is not correlated with  
> distance. However, in my curve (based on simulations), I found that  
> the PRR shows a relatively smooth decay with distance which is  
> uncommon . This is not acceptable because TOSSIM uses an accurate  
> link layer model.
>
> Note that I used the environment parameters as in the Tutorial: http://www.tinyos.net/dist-2.0.0/tinyos-2.x/doc/html/tutorial/usctopologies.html
>
> Can someone tell me where the problem can be (in the environment  
> setting, in the accuracy of TOSSIM, …)
>

To follow up: Nouha tried the experiment again and found a grey  
region. So it was either a mistake in the first experiment or  
misleadingly clean data.

Phil