(keitai-l) Re: Blip, blip, zap! 3G in Europe & roaming...

From: Maria Pienaar <maria.pienaar_at_solidtech.com>
Date: 10/15/01
Message-ID: <004501c15571$759ec790$ce01a8c0@us.solidtech.com>
On Friday, October 12, 2001 Benjamin Kowarsch wrote:

"The difficulty with PHS is that PHS has not equivalent roaming
international roaming functionality and therefore such a gateway cannot
simply translate but it would also have to provide all the missing
functions.

However, the ZEBRA roaming protocol has this functionality already
built-in because it caters to situations where home networks do not
support ZEBRA. This is called non-cooperative mode. There is also no
need to deliver incoming calls via GSM/MAP because when the user signs
in to a foreign network they are assigned a local account and a local
phone number which is then sent back to the home network. If the home
network wants to provide this service they can then use an ordinary call
forwarding service to deliver incoming calls to their roaming customer.

The home network need not even be a mobile network. A fixed network can
also participate. Thus, you could have a PHS cordless phone (without
public PHS service) and if it was a GSM/PHS dual mode handset then you
could roam via ZEBRA into a GSM network and receive incoming calls to
your house or office on your mobile while roaming, or have voicemail
forwarded to yourself."


Benjamin

I have been following the discussion with quite a lot of interest. Where can
I find out more about the Zebra roaming protocol? Are you referring to the
Zebra routing protocol for IPv6 or something completely different?

You also seem to be mixing up a number of different roaming areas between
voice and data roaming, which makes the discussion sometimes a bit hard to
follow as these are such different areas at this point in time and unlikely
to change.

Regarding the topic of inter-standard voice roaming, which seem to be the
current topic under discussion the following:

The main problem with inter-standard roaming is not the technical solutions
to it, but the commercial agreements that tie these together as well as the
potential fraud risk. The potential subscriber fraud risk is potentially
quite high when you look at roaming, and it is the home network operator's
risk, not the visited network operator. Credit cards are in a lot of cases
not accepted as a back-up means to the potential credit risk, due to the
high incidence of credit card fraud in a lot of countries.

There has been numerous technical solutions in the past, from very easy ones
to more complicated ones, but as you suggested in one of your emails
referring to CAMEL - which is more seen as a key signalling protocol to
enable real-time rated prepaid roaming, there has not been an active drive
by all network operators to implement this.

On the technical side, one would need a dual mode user device that will
support both types of subscriptions. For argument sake, let's use the
example of a CDMA subscriber roaming into a GSM network using a dual-mode
CDMA/GSM handset. The handset would have to support a SIM card, with a valid
subsctiption to a GSM network, as well as the subscriber information
programmed into the handset as supported by the IS-41 standard for CDMA and
TDMA. On the GSM side, it is unlikely that network operators would
authenticate a phone with no valid network operator or roaming partner SIM
on their networks. It is not just a simple case of "authenticating" the
IS-41 subscription back to the home CDMA network via a gateway. The only way
a "local" GSM roaming account would be created in the visited GSM network is
possibly if a "local" SIM card or valid roaming partner SIM card is used,
which implies that the home CDMA network operator would need to have access
to such a SIM card for it's roaming phones. At this point in time, non-GSM
networks are not able to get access to their own GSM SIM cards as only
network operators owning GSM networks can have this. The other option is to
own subscriptions from valid GSM operators such as Swisscom, but the fraud
risk for these subscriptions are still the home network's.

Creation of "local" account at the visited network:
I can only see this potentially happening for prepaid where the potential
subscriber fraud risk is much lower since the payments are done upfront to
the visited network. I cannot see a visited network just randomly "opening
up" access to subscribers it has potentially no control over and where the
home network does not take responsibility for the fraud of it's subscribers
visiting another network. So, if we are looking at prepay for this scenario,
then it would make sense to calls forwarded to your "local visited preapy
account, in a more seamless way, which means you need to be able to identify
this as a roaming prepay subscription on your network - as you know, there
are some technical difficulties with this and many cluegy solutions. And, on
top of this, the question is where does the initial amount come from for the
account so the subscriber could use it.

Also, one thing you seem to be continuing to ignore is the impact of local
languages for local services - if you are using a "local" server for e.g.
prepay or unified messaging, then it is normally supported in the local
language, not the language of the visitor. So, this would imply support for
more langauges on these "local visited" platforms. One could argue that
english would be a good medium for business travelers, but when you start
looking at the touris section this may not apply. Not all operators would be
open to do this on their "local visited" platforms. Also, forwarding of
calls from the home network to the visited network is not always a cheap
alternative, as the charges are still based on the home network operator
charges. Even in the case of your suggested "selective forwarding" - I would
hate to be a subscriber to get a voice call just telling me that someone
with a valid PIN is phoning me at 2 am in the morning because I happen to be
in a different time zone unknowing to this person. Even if I "reject" this
call, I will most likely still be charged by the home network for this
forwarded call.

I doubt if Zebra will be able to help address any of the commercial issues
behind what makes roaming work and the choice of operators to support it.
From your descriptions, it sounds like another attempt to revive the old
"one number service" concept. These types of "personal" and "personal
assistant" services have been tried numerous times in the market and has
failed misserably. As a subscriber, give me seamless roaming services that
are affordable and don't make it more difficult for me or more frustrating
for people calling me to reach me with PIN codes and other delaying means.

From the operator side, roaming is all about controlling the fraud risk and
where it sits. I believe only real-time rated prepay roaming could bring
roaming into the hands of the normal vacationing traveler in a least fraud
risk manner - providing this type of service is easy to manage by the
various operators and supports the various issues of top-ups, different
currency controls and language issues.

Now, let's start adding the various data services roaming to the picture
which is what 2.5G and 3G adds to the picture..... ;-)

Maria



[ Need archives? How to unsubscribe? http://www.appelsiini.net/keitai-l/ ]
Received on Mon Oct 15 14:48:30 2001