cell re-selection

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
Report Content as Inappropriate

cell re-selection

i did a lot of resarch and testing on cell selection and re-selection
process the last two week.
the cell selection process, network selection process (manual and
automatic) and mobility management process were already implemented in
OsmocomBB a long time, but turned out to be buggy and incomplete. i made
test drives to check the process and debugged it.
the re-selection process is new. it is used to track surrounding cells
while listening to the BCCH of the current cell (camping on a cell).
special extension to the layer1 firmare is used to measure neighbour
cells. if an neighbour cell becomes 'better', the mobile switches to
that cell, depening on different criteria. now it is possible to move
with OsmocomBB.
the re-selection process is not handover! handover is a process where a
phone switches between cells while doing a call. handover is one next
step to implement. the process is a little more complex, because it
requires not only neighbour cell measurements, but also syncing to them
without interrupting the traffic channel. most layer 3 stuff of handover
is already implemented.
if you like to play and test your moving OsmocomBB, you can check out
the "jolly/roaming" branch. it contains the extension to layer1, as well
as sim reader and fixes from "sylvain/testing" branch. use both "mobile"
and "layer1" firmware from this branch.
in order to see some process at VTY, you can do:
monitor network 1 (continously display the strongest cell and neighbour
show ms 1 (to see current states)
show neighbour-cells 1 (to see a more detailed current list of

Reply | Threaded
Open this post in threaded view
Report Content as Inappropriate

Re: cell re-selection

This post has NOT been accepted by the mailing list yet.
Hello Sir:
I am working on the implementation of call handover (in dedicated mode) to the osmocom-bb code. Any progress from your side or any insights that you feel are necessary for this implementation would be highly appreciated.

Asghar Mustafa.