Yes I am also concerned about slotted ACKs. Section 5.2 considers the fact that it is not neccesary to fire a timer every 20 symbols and proposes Adaptive Timer Resolution mechanism. but later it quotes proposed mechanism could not solve the starvation problem at the data rate of 2 frames/sec.
As paper title gives a view that TinyOS concepts is the limitation for supporting beacon mode. But I think the real bottleneck is hardware (MCU) capabilities itself. I am also trying understand how any other modren operating system with preemptive scheduling (for ex. t-kernel) can handle the situation. Thanking you, D. Manjunath On Tue, 8 Aug 2006, [iso-8859-1] Diego Bartolomé Arquillo wrote: > Hi all, > I am agree with Phil about the section 5.1 of the paper. It is not neccesary > to fire a timer every 20 symbols. I think the statement: "In order to > communicate in a beacon-enabled network, a timer expiring every 20 symbols, > i.e. 320 us, is needed" is not correct. The only real problem that I see, > could be the related in section 5.3. If it is true that the transmission of > an acknowledgment from software require 132 symbols, it will not be possible > to carry out the timing requirements for slotted acknowledgments. > BR, > Diego > > -----Mensaje original----- > De: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] En nombre de Philip Levis > Enviado el: martes, 08 de agosto de 2006 0:00 > Para: [EMAIL PROTECTED] > CC: tinyos-help@millennium.berkeley.edu > Asunto: Re: [Tinyos-help] Re: IEEE-802.15.4 - Beacon Mode > > On Aug 5, 2006, at 4:30 AM, [EMAIL PROTECTED] wrote: > > > > > Dear friends, > > > > The paper entitled "EVALUATING THE USE OF MOTES AND TINYOS FOR > > A MOBILE SENSOR PLATFORM" > > evaluates 802.15.4 (beaconmode) on a real sensor network of micaz and > > telos motes. > > Suprisingly, it says beacon mode can not be supported on > > capabilities that present motes posses. Any comments on the paper > > will be very useful....... > > > > please google for the paper as attachment mails are bouncing back!!!! > > I'm not sure if I understand all of the conclusions made in the paper; among > other things, Jonathan Hui and Joe Polastre implemented > 802.15.4 beaconing mode as part of the implementation of SP. > > For example, in section 5.1 you state that 15.4 requires a periodic 20 symbol > timer, and that this makes it impossible to implement properly. I don't think > your statement is correct. It requires a timer with a fidelity of 1 symbol, > but there's nothing there that means that you need to fire a timer every 20 > symbols... why would you need to do so? You're picking a random number in the > range of random > (2^(BE-1)) unit periods... > > Phil > _______________________________________________ > Tinyos-help mailing list > Tinyos-help@Millennium.Berkeley.EDU > https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help > > > > -- > This message has been scanned for viruses and > dangerous content by MailScanner, and is > believed to be clean. > > -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. _______________________________________________ Tinyos-help mailing list Tinyos-help@Millennium.Berkeley.EDU https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
