From adsl5qfc at tpg.com.au Tue Oct 12 07:43:02 2004 From: adsl5qfc at tpg.com.au (adsl5qfc@tpg.com.au) Date: Tue Oct 12 07:43:54 2004 Subject: [Smac-users] Making a node sleep in active cycle Message-ID: <1097592182.416bed76ed722@postoffice.tpg.com.au> Hi, is it possible to make each node decide whether it ought to turn itself on during the period of the listen cycle for DATA packets based on a state variable? In other words, after the SYNC period of the listen cycle, each node will have to check the value of a state variable before deciding whether to go to sleep for the rest of the duty cycle or to listen for RTS/DATA packets. If it is possible, which parts of the smac code in ns-2 do I have to alter? Thank you! Kind Regards, Eric From weiye at ISI.EDU Tue Oct 12 08:48:33 2004 From: weiye at ISI.EDU (Wei Ye) Date: Tue Oct 12 08:49:57 2004 Subject: [Smac-users] Making a node sleep in active cycle In-Reply-To: <1097592182.416bed76ed722@postoffice.tpg.com.au> References: <1097592182.416bed76ed722@postoffice.tpg.com.au> Message-ID: <1097596113.15449.3.camel@wee.isi.edu> You can check how it works with nav (network allocation vector), neighborNav (a timer to track neighbors' nav) and adaptive listen as an example. -Wei Ye On Tue, 2004-10-12 at 07:43, adsl5qfc@tpg.com.au wrote: > Hi, > > is it possible to make each node decide whether it ought to turn itself on > during the period of the listen cycle for DATA packets based on a state > variable? In other words, after the SYNC period of the listen cycle, each node > will have to check the value of a state variable before deciding whether to go > to sleep for the rest of the duty cycle or to listen for RTS/DATA packets. If it > is possible, which parts of the smac code in ns-2 do I have to alter? Thank you! > > > Kind Regards, > Eric > _______________________________________________ > Smac-users mailing list > Smac-users@mailman.isi.edu > http://mailman.isi.edu/mailman/listinfo/smac-users From adsl5qfc at tpg.com.au Tue Oct 19 01:04:43 2004 From: adsl5qfc at tpg.com.au (adsl5qfc@tpg.com.au) Date: Tue Oct 19 01:05:18 2004 Subject: [Smac-users] wireless transmission range Message-ID: <1098173083.4174ca9b46caa@postoffice.tpg.com.au> Hi, I'm trying to reduce the wireless transmission range of nodes by changing the appropriate values in Phy/WirelessPhy (e.g. RXThresh, Pt_, CSThresh_). However, no matter how small I set the values to be, the transmission ranges don't seem to be affected. For example, when I run a 360x360 topology with 20 nodes inside, a SYNC broadcast transmitted by any node in the topology will be heard by every other node. I am running DSDV as my routing protocol, the height of my antenna is 1.5 and the antenna gains are set at 1.0. May I know how I can resolve my problem? Thank you, Eric From fll at iis.fraunhofer.de Tue Oct 19 02:02:42 2004 From: fll at iis.fraunhofer.de (Christian =?ISO-8859-1?Q?Fl=FCgel?=) Date: Tue Oct 19 02:03:17 2004 Subject: [Smac-users] S-MAC and mobile Nodes Message-ID: Hi all! According to the publications concerning s-mac, a node which overhears more than one schedule may adopt both of them to improve inter-cluster communication. What about mobile nodes? In our project nodes will be mobile and will move at random through our network at speeds of up to 10 mph since people are wearing them. So what happens if a node leaves one cluster and joins another? Is this possible? Will this node keep adopting schedules until it is awake at all times or will he be able to drop old schedules which he no longer needs to adhere to? Are there any publications which have measured performance of s-mac when nodes are mobile? Christian -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.isi.edu/pipermail/smac-users/attachments/20041019/a6c1b0da/attachment.html From johnh at ISI.EDU Tue Oct 19 09:05:59 2004 From: johnh at ISI.EDU (John Heidemann) Date: Tue Oct 19 09:06:25 2004 Subject: [Smac-users] S-MAC and mobile Nodes In-Reply-To: Message-ID: <200410191606.i9JG5xtP003412@dash.isi.edu> On Tue, 19 Oct 2004 11:02:42 +0200, Christian =?ISO-8859-1?Q?Fl=FCgel?= wrote: >Hi all! > >According to the publications concerning s-mac, a node which overhears more >than one schedule may adopt both of them to improve inter-cluster >communication. What about mobile nodes? In our project nodes will be mobile >and will move at random through our network at speeds of up to 10 mph since >people are wearing them. > >So what happens if a node leaves one cluster and joins another? Is this >possible? Will this node keep adopting schedules until it is awake at all >times or will he be able to drop old schedules which he no longer needs to >adhere to? > >Are there any publications which have measured performance of s-mac when nodes >are mobile? With S-MAC as described in the ToN paper, the mobile node would need to adopt any new schedules it encountered. In addition, there is a latency to discovering new schedules. Although this is tunable, this may prove to be a problem for mobile nodes. (Although perhaps not at walking speeds.) We have done some new work about S-MAC with multiple schedules, though, see . This paper includes both experimental data about the number of schedules we observed in a large network, and a new algorithm that allows all nodes in the network to migrate to the same schedule. With lots of moving nodes I would strongly encourage use of this "global schedule algorithm". There's also a prototype implementation checked in to our current CVS version of S-MAC, although it's not on by default. -John From m_aswathkumar at yahoo.co.in Thu Oct 28 03:36:52 2004 From: m_aswathkumar at yahoo.co.in (Aswath Kumar) Date: Thu Oct 28 03:38:02 2004 Subject: [Smac-users] Power Management Message-ID: <20041028103652.79369.qmail@web8510.mail.in.yahoo.com> Hi, I am using tinyOS 1.1.0 and trying to understant the implementation of SMAC. I want to know, whether SMAC puts the microcontroller (ATmega128L) into power down mode, when SMAC is in sleep. I understant that the microcontroller runs continuously even during sleep (Processing the Clock fired every 1 ms). I would like to confirm this and would know its details. Regards, Aswath Kumar --------------------------------- Do you Yahoo!? Yahoo! Mail Address AutoComplete - You start. We finish. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.isi.edu/pipermail/smac-users/attachments/20041028/9c926c2e/attachment.html From weiye at ISI.EDU Thu Oct 28 09:07:03 2004 From: weiye at ISI.EDU (Wei Ye) Date: Thu Oct 28 09:08:04 2004 Subject: [Smac-users] Power Management In-Reply-To: <20041028103652.79369.qmail@web8510.mail.in.yahoo.com> References: <20041028103652.79369.qmail@web8510.mail.in.yahoo.com> Message-ID: <1098979623.24143.15.camel@wee.isi.edu> Yes, you're right. The original implementation does not take care of the CPU sleeping, and only puts radio into sleep mode. -Wei On Thu, 2004-10-28 at 03:36, Aswath Kumar wrote: > Hi, > > I am using tinyOS 1.1.0 and trying to understant the implementation of > SMAC. I want to know, whether SMAC puts the microcontroller > (ATmega128L) into power down mode, when SMAC is in sleep. I understant > that the microcontroller runs continuously even during sleep > (Processing the Clock fired every 1 ms). I would like to confirm this > and would know its details. > > Regards, > > Aswath Kumar > > > > ______________________________________________________________________ > Do you Yahoo!? > Yahoo! Mail Address AutoComplete - You start. We finish. > > ______________________________________________________________________ > _______________________________________________ > Smac-users mailing list > Smac-users@mailman.isi.edu > http://mailman.isi.edu/mailman/listinfo/smac-users