SIM Toolkit and NMR in idle mode

Hi,

>In general the answer will depend on the phone manufacturer.
>The specs tell manufacturers what they must do to claim
>compliance, and many do more than the bare minimum. It's
>really up to them.

>I use STK to read and log NMR from Siemens phones. I'm able to
>get much the same data (and more) using specific commands on a
>Wavecom modem, and I find no real discrepancy between my
>Siemens and Wavecom data (ignoring RXQUAL, which is hopeless in
>idle mode). Signal strengths (serving cell and 6 strongest
>neighbours) will vary a bit, but that's to be expected.
>Reassuringly, I get good agreement on the BSIC and BCCH of the
>neighbouring cells, even in dynamic environments. And most
>importantly, BSIC and BCCH always pair with each other exactly.

>GSM 04.08, as mentioned in GSM 11.14. And that raises an
>interesting issue. One of the parameters buried within NMR is
>the MEAS-VALID bit which "indicates if the measurement results
>for the dedicated channel are valid or not" (GSM 04.08, Table
>10.5.47). A strict interpretation of GSM 11.14 indicates that
>we shouldn't rely on this bit's value, but my feeling is that
>it's probably OK. It certainly seems to be dependable on
>Siemens phones.

>As to how often they're updated, my Siemens experience suggests
>at least every 10 seconds.

>John


John, could it be possible for you to post (*) a code example (whatever
language it could be) of such a use of STK for retrieving NMR informtion
from Siemens phones ?
By NRM, I mean RXlevel from neighbour cells as well as their cellid (or
BCCH) and the current cell ID.

I've just acquired a S65 for doing such things (among others) and would
like to develop a location application as a proof of concept.

BTW, do you know if it exists, for Siemens phones, a gateway between KVM
and STK in order to directly interpret the information retrieven by the SIM.

Best of all would be to retrieve directly these informations (NMR) from
within the KVM but ... I guess it's not time up yet for mobile
manufacturers to publish such things.

Regards,

db


(*) Or to send me a mail to the following address (insert a '@' and a
'.' in the respective blanks).
--

Courriel : usenet blas net