No subject

Fri Jul 18 19:54:00 PDT 2008

rebooting intermittently.  If this happens, the sink node becomes out of
sync from all the other nodes, and it takes a while for the other nodes to
realize that they aren't in sync with the sink node and turn their RF chips
on permanently to listen for the next sync message.

Hope I'm in time, and wish the best of luck!
Tim

On Fri, Jun 13, 2008 at 10:20 PM, Anthony Dern <ebbie2007 at gmail.com> wrote:

> Esteemed list members,
>
>
>
> I have an application that was done in boomerang v1.4 (same issue if I
> upgrade to 1.5). my mote application is based on the standard Delta app from
> Moteiv, with some changes to the data packet to include some sensor data. It
> is running on TelosB motes (both Tmote Sky and Xbow). the app is built using
> "make tmote lowpower,5".  I am reading the sensor and sending the data
> packet from each mote every 10 minutes. Typical system of 80-100 motes.
> (though I see the same issues with a test system of 12 motes)
>
>
>
> I am using a Moteiv Tmote Connect as a gateway. I have a service that I
> wrote that runs on a windows machine that connects to the tmote connect,
> receives incoming packets, and routes the data to a sql database.
>
>
>
> The system seems to work fine (at least acceptable) if all of the motes are
> within single-hop distance of the gateway. When I get a network that
> actually relies on multihop, the system falls apart, with close to a 50%
> packet loss.
>
>
>
> My service is recognizing Uart Detect Request packets and sending the
> proper response. It also sends a Uart Keep Alive packet every 5 seconds. The
> symptom that I am seeing is that I will get packets from the motes normally
> for a while, then I stop getting packets from all but mote address 0 (on the
> gateway) this situation will continue for a while then I start getting
> packets normally for a while again. It seems to be acting like node 0 on the
> gateway is losing its "base station status" and all of the packets being
> transmitted by the other motes are going into the ether.
>
>
>
> One concern that I have is that even though I am sending the Uart Keep
> Alive message, I will continue to get Uart Detect Request packets from node
> 0. As I understand it, this should not be happening.
>
>
>
> I should also note that I was sent several patches by Rob Szewczyk. These
> patches replace files in the CC2420Radio, MultiHopLQI, netsync, and sp
> directorys at \opt\moteiv\tos\lib. These files were sent to me as fixes for
> an earlier lowpower mesh issue. They were not incorporated into the 1.5
> version of boomerang.
>
>
>
> Also to note, sometimes when the communication "comes back" I get a number
> of packets that were queued in the network somewhere.
>
>
>
> Thank you in advance for your time. I appreciate the help I hope I have
> included enough information.
>
> _______________________________________________
> Tinyos-help mailing list
> Tinyos-help at millennium.berkeley.edu
> https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
>

------=_Part_181563_6217202.1219037209703
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

<div dir="ltr">I have managed to successfully run a 25-node multi-hop network using Boomerang 2.0.4 from the TinyOS CVS.&nbsp; One thing to make sure of is that you patch &quot;NetSyncM.nc&quot; as per this page:<br><a href="http://docs.tinyos.net/index.php/Boomerang_Fixes">http://docs.tinyos.net/index.php/Boomerang_Fixes</a><br>
<br>Or just download from this link: <a href="http://sing.stanford.edu/klueska/tos_wiki_files/NetSyncM.nc">http://sing.stanford.edu/klueska/tos_wiki_files/NetSyncM.nc</a><br><br>From your symptoms, however, it seems that your sink node might somehow be rebooting intermittently.&nbsp; If this happens, the sink node becomes out of sync from all the other nodes, and it takes a while for the other nodes to realize that they aren&#39;t in sync with the sink node and turn their RF chips on permanently to listen for the next sync message.<br>
<br>Hope I&#39;m in time, and wish the best of luck!<br>Tim<br><br><div class="gmail_quote">On Fri, Jun 13, 2008 at 10:20 PM, Anthony Dern <span dir="ltr">&lt;<a href="mailto:ebbie2007 at gmail.com">ebbie2007 at gmail.com</a>&gt;</span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><p style="margin: 0in 0in 0pt;"><font face="OCR A Extended" size="3">Esteemed list members,</font></p>

&nbsp;
I have an application that was done in boomerang v1.4 (same issue if I upgrade to 1.5). my mote application is based on the standard Delta app from Moteiv, with some changes to the data packet to include some sensor data. It is running on TelosB motes (both Tmote Sky and Xbow). the app is built using &quot;make tmote lowpower,5&quot;. &nbsp;I am reading the sensor and sending the data packet from each mote every 10 minutes. Typical system of 80-100 motes. (though I see the same issues with a test system of 12 motes)

&nbsp;
I am using a Moteiv Tmote Connect as a gateway. I have a service that I wrote that runs on a windows machine that connects to the tmote connect, receives incoming packets, and routes the data to a sql database.

&nbsp;
The system seems to work fine (at least acceptable) if all of the motes are within single-hop distance of the gateway. When I get a network that actually relies on multihop, the system falls apart, with close to a 50% packet loss.

&nbsp;
My service is recognizing Uart Detect Request packets and sending the proper response. It also sends a Uart Keep Alive packet every 5 seconds. The symptom that I am seeing is that I will get packets from the motes normally for a while, then I stop getting packets from all but mote address 0 (on the gateway) this situation will continue for a while then I start getting packets normally for a while again. It seems to be acting like node 0 on the gateway is losing its &quot;base station status&quot; and all of the packets being transmitted by the other motes are going into the ether.

&nbsp;
One concern that I have is that even though I am sending the Uart Keep Alive message, I will continue to get Uart Detect Request packets from node 0. As I understand it, this should not be happening.

&nbsp;
I should also note that I was sent several patches by Rob Szewczyk. These patches replace files in the CC2420Radio, MultiHopLQI, netsync, and sp directorys at \opt\moteiv\tos\lib. These files were sent to me as fixes for an earlier lowpower mesh issue. They were not incorporated into the 1.5 version of boomerang.

&nbsp;
Also to note, sometimes when the communication &quot;comes back&quot; I get a number of packets that were queued in the network somewhere.

&nbsp;
Thank you in advance for your time. I appreciate the help I hope I have included enough information. 
 _______________________________________________ 
Tinyos-help mailing list 
<a href="mailto:Tinyos-help at millennium.berkeley.edu">Tinyos-help at millennium.berkeley.edu</a> 
<a href="https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help" target="_blank">https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help</a> </blockquote></div> </div>

------=_Part_181563_6217202.1219037209703--