search

akrv / deBr

Decentralized Brains in low data-rate, low power networks for collaborative maneuvers
Other
2 stars 1 forks source link

what is happening #1

Closed AnasIbrahim closed 4 years ago

AnasIbrahim commented 5 years ago

we should use this issue instead of email, it's easier. @akrv Ramesh and me will report here what we are doing

AnasIbrahim commented 5 years ago

Ramesh and I met last week to take a deeper look on the glossy code from LWB and Kasunch repo. I tried compiling both repos for our target with Contiki NG. So, for contiki vs contiki ng and latest contiki ng there is no difference at all, both repos work for current contiki ng.

We will use Contiki ng develop branch not the master, and yeah develop branch is not a specific release but the target names changed and we don't want to edit that later. just hope they name this edit as a release soon.

we also tried to check how much of the code should contain platform specific stuff (so we know how to properly organize the folders), and even for Kasunch repo the files that he imported from TI (rf****.h files) are really different from our chips rf files, so the code needs to be very platform specific.

I got the glossy files from LWB and build the folders hierarchy, now editing the make file, I will push these stuff soon here.

toothless90x commented 5 years ago

15-Nos Rucksack with Sensortag will b ready in one or two days. David and team doing the hardware test, I will also join them tomorrow. @AnasIbrahim finally own folder hierarchy . đź‘Ť I was getting lost with that.

Comparing Kasunch: this will get us an idea for Arm platform, but for the core development we should consider the LWB-git and LWB-generic glossy algorithm. I will setup our repo in halle and communicate with Anas further.

toothless90x commented 5 years ago

For Get-to-know meeting presentation, kindly get me your ur suggestions for the following sub-topics.

Topic : Sensor floor -Project idea (3 mins) (@akrv I need some pictures and a bit of explanation from you either here or in an email) -CC1350 MCU - dual band (1 min) -Rucksack (PCB) development and testing (~3 mins) -Glossy (~3 mins)

In total, I planned for 8~10 min presentation. @AnasIbrahim you can join at glossy section. Since, there will be almost all HiWi's particiapting, regardless of their domain and studies, we will keep it simple and elegant way. I will post the PPT file, once I finish.

akrv commented 5 years ago

For Get-to-know meeting presentation, kindly get me your ur suggestions for the following sub-topics.

Topic : Sensor floor -Project idea (3 mins) (@akrv I need some pictures and a bit of explanation from you either here or in an email) -CC1350 MCU - dual band (1 min) -Rucksack (PCB) development and testing (~3 mins) -Glossy (~3 mins)

In total, I planned for 8~10 min presentation. @AnasIbrahim you can join at glossy section. Since, there will be almost all HiWi's particiapting, regardless of their domain and studies, we will keep it simple and elegant way. I will post the PPT file, once I finish.

we will talk about that next week in person. it would be great to have the PPT ready.

akrv commented 5 years ago

reading kasunch code, I would say that his code should be reusable with just minimal changes to the rtimer acq and rf interfacing driver from glossy to the chipset specific rf-driver.

  1. is there any implementation of glossy for cc2650? which would accelerate most of the stuff we are trying. have you found anything there?

  2. have you already tried if hello-world example already the referenced version of Contiki-ng compiles for our board?

AnasIbrahim commented 5 years ago

for Kasunch code I guess the problems are: 1- the rf modules for cc1350 are totally different from cc2538, I couldn't see that the modules to are similar, so all of the extra functions he wrote will be worthless and we need to rewrite everything [this is the main point, verify if i'm wrong] 2- the code looks kinda not well formatted, everything is in one file, harder to read 3- he implemented an encryption do we need this?

So i think it's better to start from glossy itself and just refer to kasunch code as a reference on how to make some changes.

  1. for implementation of glossy for cc2650, I saw some papers refer to using some protocols on top of glossy, but no mention of the code being open.

  2. you mean just compiling the example for our board, then yes compiling works and the verification and the flashing using uni flash works. I didn't verify if the program is working on the board itself, but it should.

toothless90x commented 5 years ago

Current version: Presentation PDF: SensorFloor.pdf workfile PPTX: Sensorfloor1.pptx

Some texts will be edited and added in the final version. Project Idea has to be explained in first slide

akrv commented 5 years ago

image That’s already a good sign. Lets see how far it goes.

Best regards Aswin

On 8. Apr 2019, at 10:20, Ramesh notifications@github.com wrote:

Current version: Presentation PDF: SensorFloor.pdf https://github.com/akrv/deBr/files/3053282/SensorFloor.pdf workfile PPTX: Sensorfloor1.pptx https://github.com/akrv/deBr/files/3053293/Sensorfloor1.pptx Some texts will be edited and added in the final version. Project Idea has to be explained in first slide

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/akrv/deBr/issues/1#issuecomment-480733070, or mute the thread https://github.com/notifications/unsubscribe-auth/AEkNUBXlK2_Azg4AjreptaeTAvjXJaK5ks5vevvEgaJpZM4cXgia.

akrv commented 5 years ago
AnasIbrahim commented 5 years ago

for the echo, this is the hex file https://github.com/akrv/deBr/blob/master/apps/echo/echo.simplelink

akrv commented 5 years ago

Please check the file format in this repo: https://github.com/JelmerT/cc2538-bsl

And provide me that format please. Should be “hex” if I m not wrong

Aswin

Sent from my iPhone

On 6. May 2019, at 17:56, Anas I. Gouda notifications@github.com wrote:

for the echo, this is the hex file https://github.com/akrv/deBr/blob/master/apps/echo/echo.simplelink

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

AnasIbrahim commented 5 years ago

it's a .hex file. you can rename it. i used uniflash to flash it (http://www.ti.com/tool/UNIFLASH), i will check this other tool also.