Thank you for the answers.
I have a question regarding the use of LPL in the CTP
I have made changes in order to make the lpl work in the following way
in "Boot.booted():"
call LowPowerListening.setLocalSleepInterval(250);
While sending message :
call LowPowerListening.setRxSleepInterval(&packet, 250);
and have used the proper wiring.
As I read through the mailing list you had mentioned that one needs to hack
into the ctp to send according to LPL for each packet.
However I found out ctp with lpl in the cvs
.
http://hinrg.cs.jhu.edu/git/?p=razvanm/tinyos-2.x.git;a=commit;h=9609b6b90101acae0da46deb835714f9a28d94c1
but seems like some files are missing e.g the Globals.h. Where can I find
the working version of LPL use in the CTP ?
regards,
Nicole
On Mon, Feb 14, 2011 at 2:59 AM, Omprakash Gnawali
<[email protected]>wrote:
> On Sun, Feb 13, 2011 at 9:15 AM, Nicole Chiesi <[email protected]>
> wrote:
> > I am facing a situation where , I use the TestNetwork Application in
> testbed
> > of about 45 telosb motes. I was trying to check the saturation point of
> ctp.
> > I increased the packet rate to some 1 packet per second for about 15
> nodes
> > and one packet every 8s for
> > the rest 20 motes. I still have a very good delivery ratio of about 98
> > percentage.
> > How much of data rate is ctp supposed to handle?
>
> CTP is not designed to work with high data rates. It does not limit
> the number of packets injected into the network. So, you should try to
> operate it under channel saturation. What rate will cause channel
> saturation? That will depend on the radio, MAC, network topology, ...
>
> > When I went down to 3packets /sec for the 15 nodes and one packet /8s
> for
> > the rest of the nodes , then was only significant drop in the delivery
> ratio
> > to about 74 percent.
> > I think the results I got are not accurate? i do not know if I did
> something
> > wrong here?
>
> You are sourcing 45+ packets/s. Many of these packets might be
> multi-hop which means the region around the sink is receiving and
> sensing many more packets. You might be way past channel saturation.
> If the channel is not saturated, it is reasonable to expect a decent
> performance.
>
> - om_p
>
_______________________________________________
Tinyos-help mailing list
[email protected]
https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
