Jan Floras blog

Wee, as of last night I finally had a symbol period granularity timer module working for the Micro.4 platform. I’ve rewritten the virtual timer interface to be able to convert an n-bit wide hardware timer with symbol period granularity into a 32-bit wide virtual timer with symbol period granularity. This is very handy when it is impossible to drive the hardware timers at a 62,5 KHz clock as needed for symbol period time. The Sensinode Micro.4 MSP430 timers are now run at a 125 KHz clock making them 15-bit wide and the Freescale DIG528 MC9S08GB60 timers are run at a 31,25 KHz clock making them 17-bit wide. Both are converted to 32-bit symbol period timers usable by Tiny15Four though the generic virtual timer module.

Next up is fighting the CC2420 radio. Hopefully things will go a little more smooth now that I have some documentation for the Micro.4 platform (BTW, thanks to Pekka Saavalainen from Sensinode for supplying me with the relevant documentation for both Micro and Nano).

Also Rostislav Spinar has suggested that I set up a web page at google code to make the Tiny15Four source more easily available. I’m planning on doing just that as soon as time allows me to. Thanks to Rosta for the good suggestion and offer to help me out on that one

Today I went by Martin Leopolds new place of employment and picked up a Sensinode Micro and Nano (with a CC2420 and CC2430 respectively). Martin has done alot of work porting TinyOS to different mcs51 (8051) based platforms. If you are interested in such platforms you can check his page out here:

TinyOS 8051 working group 

The plan is now to supply full 802.15.4 support for the popular CC2430 SoC through the use of Tiny15Four. I will report on my progress here..

I’ve released my TinyOS for JN5121 code in its current state in the download section. It is far from complete, but I guess some people might find use for it, so why not make it public

I’ve started planning the porting of Tiny15Four to a CC2420 platform, probably the Sensinode Micro series (although outdated, this is what I have access to). This will be my last work on my master degree, but hopefully I’ll be able to continue my work on Tiny15Four as a Ph.d. student.

With regards to the Jennic JN5121 platform, I’m planning on releasing whatever code I currently have for TinyOS on this site. I don’t know how much more work I’ll put into disassembling the Jennic code. I seems realistic that I would be able to provide drivers for every module in the chip except for the radio. And will not be of much use with regards to Tiny15Four.

It finally happened. I succesfully defended my thesis friday and received the heighest grade (12). I’m very pleased with the outcome and the fact that it is overwith

With regards to compiling the Tiny15Four source code and getting to a release 1.0, it seems that Jonas Fonseca, who is also working hard with the Re-Mote testbed (http://remote-testbed.googlecode.com/ and http://remote-testbed.blogspot.com/), is going to setup the remote compiling environment again. This means that there will hopefully be a tested 1.0 release soon

I’ve decided to release my Tiny15Four source in a “not tested” version for those of you asking for it. My problem right now is that I don’t have access to a HCS08 compiler so I can’t test my code on a real platform. I know that it compiles in NesC though. No documentation or anything. Write me if you are interested.

Download the source code here: Tiny15Four 0.9

Finally I’ve had some time to look at the jn5121 port. Since I still have no documentation on the platform what so ever, I’ve decided to figure things out myself. I’ve written a disassembler that runs on the chip, so that I can figure out, what all the functions supplied in ROM by Jennic actually does.

I’ve managed to write native drivers for the timers and the SPI during the last couple of days. So things are starting to progress. I’m a little worried about, if I can get enough information about the transceiver out of the Jennic code, since it is not directly accessible though any of the APIs provided by Jennic.

As soon as I have some more modules ready, I’ll publish some code in here.

Wheeeee.. My thesis is done and I have submitted it to printing and binding today at 9:30.. So tomorrow I’ll pick it up at 10 from the print shop and will hopefully be able to turn it in before noon. Now, I just have to wait for my thesis defense. As my advisor, Philippe Bonnet, is currently on paternity leave, the thesis defense will be held until Q1 of 2008. In the meanwhile my plan is to make some money (3 months of thesis writing can really destroy your economy) and look a little more into the Jennic platform.

Be the first to read my thesis available here: Tiny15Four: An IEEE 802.15.4 implementation in TinyOS

Wow, it’s really been awhile since I’ve had the chance to work with all WSN related project. Mostly because I had to complete the obligatory military service, although I didn’t actually serve my time in the military because of my pacifism

But I’m back again and I begun writing my master thesis just this friday. And it is really progressing well. I guess have done around 1/10 of it already. I’m (big surprise) writing about my TinyOS 802.15.4 radio stack. And in that connection, I’m going to make a version 1.2 release that will both be available here and of course through the TinyOS CVS repository at Sourceforge.net. I have no time estimate on the release just yet. It very much depends on how things are progressing. The thesis will also be available from the blog as soon as it is done as will slides from my defence.