Re: Request for DECT LINKTYPE

[Matthias Wenzel <tcpdump () mazzoo de>]

Guy Harris wrote:
> On Dec 22, 2008, at 1:18 PM, Matthias Wenzel wrote:
>
> > Guy Harris wrote:
> >
> > > So DECT and DECT 6.0 are close enough that they can use
> > > LINKTYPE_DECT/DLT_DECT?
> >
> > Yes. DECT has 10 radio channels, DECT 6.0 has 5, they all fit into one
> > 0-63 channel table scheme in ETSI EN 300 175-2, Annex F (mostly
> > unlicensed channels).
>
> So what would the data in the packet start with?  Would these be
> MAC-layer messages (possibly with some metadata)?

We currently start with metadata such as rx/tx mode, channel, timeslot,
frame/multiframe number, RSSI value. Then follows the packet almost
verbatim as seen on air, starting with a preabmle, packet type, payload
and trailing CRCs.
I wrote "almost verbatim" as our device already does some descrambling
of parts of the payload.

> So perhaps I'll just call it WDCT (that's what Siemens appears to call
> it, and they appear to be the ones who developed it), hence
> LINKTYPE_WDCT/DLT_WDCT.

good, but see below.

> Are there any published specifications for it, or is that only specified
> in internal Siemens documentation?

Forget all I said about WDCT, it's all within the same standard ETSI EN
300 175. Now I had a look at that part of the standard, and all the
standard defines is another set of channels in various ISM bands as
915MHz, 2.4GHz and 5.8GHz.
The channelhopping story is not part of the standard but exists as
proprietary manufacturer extensions. Otherwise one couldn't get any
quality on those noisy ISM bands.
As a result there's no interoperability between different manufacturer's
phones and basestations in WDCT.
We can't care too much about proprietary extensions, so I suggest just
one LINKTYPE_DECT, and we'll add a metadata field for the band.

best regards,

matthias
-
This is the tcpdump-workers list.
Visit https://cod.sandelman.ca/ to unsubscribe.