(keitai-l) Re: International Usage

From: Benjamin <bkml_at_mac.com>
Date: 07/04/02
Message-Id: <451EADD8-8F24-11D6-A07B-003065FB21DC@mac.com>
On Thursday, July 4, 2002, at 05:25 , Christopher Kobayashi wrote:

> Now .. Maybe it's a good time to look into getting Japanese phones to
> work in the US. Like a small area at first .. a wireless hotspot for the
> phones to work with most of their features. Can you buy equipment from
> Japan like that? Prepackaged features in a box with an antenna? From
> NEC, Fujitsu. Small wireless company with phones from Japan. Is it like
> 802.11 where you don't have to worry about purchasing your own
> frequency?

Absolutely not !!! (although there is an alternative)

I had explained to you that Japanese CDMA had to be customised so it 
would fit in to the horrible mess of Japanese cellular spectrum. The 
consequence of this is also that there is absolutely no way that anybody 
could deploy hot spots of Japanese wireless in the US or elsewhere. The 
Japanese have isolated themselves from the rest of the planet and the 
only way to make phones interoperable are ...

- dual mode phones (or more accurately two phones in one), ie. KDDI 
passport
- cleanup of the Japanese spectrum and migration to US CDMA or Euro-GSM


The reasons behind this are plentiful. You can search the archive of 
Keitai-L and you will find a number of threads were we have been 
discussing this and the ins and outs of the Japanese "going it alone".

This has to do with the way the Japanese spectrum is organised. 
Typically, wireless phone systems work on two frequencies in parallel so 
you can listen and speak at the same time (duplex telephony). Because 
the radio tower transmits at much higher power than mobile terminals, 
there is a potential problem with interference between the transmission 
and the receiving channels if they are too close together. As a result, 
wireless systems are designed such that there is a safe distance between 
transmitting and receiving channels, or uplink and downlink. This is 
called frequency separation.

In the rest of the world, across wireless standards, the frequency 
separation has traditionally been 45MHz in 800 and 900 MHz bands. 
However, the Japanese decided to do everything they could in order to 
protect their domestic market from foreigners and so they made a number 
of design decisions which where not driven by good design principles. 
First they chose to reverse uplink and downlink and In the case of 
frequency separation they chose 130Mhz, in the process occupying 
***both*** the US 800 MHz ***and*** the Euro 900 MHz band.  When they 
ran out of bandwidth they then had difficulties to find additional 
spectrum at the neighbouring bands. The problem was that they needed to 
find free spectrum not only neighbouring one band (ie 800 MHz) but also 
a corresponding patch neighbouring the other band (ie 900 MHz). Well, 
they couldn't find any such spectrum. They then gave up the idea of 
keeping the 130 MHz separation for the extended band, but still they 
couldn't find spectrum. Eventually they ended up with a myriad of 
patches which are bolted on all over the place and all of them have 
different frequency separations. A radio planner's nightmare.

And the more you dig for details, the more examples of "we want to be 
different just to be different" buerocrat's design you will find.

No way, you get any other country to allow such a mess to be deployed in 
their spectrum !!!

There was one little chance for KDDI to make seamless international 
roaming possible without the need for dual mode handsets:

In all of the Japanese cellular spectrum, there is one tiny little patch 
of 2 MHz which has a corresponding 2 MHz patch that could be used to 
create a 2x 2 MHz patch of paired spectrum with a 45 MHz separation 
within the US 800 MHz band.

If these two patches had been reassigned and used as a 2x 2 MHz US 
conform CDMA patch, and also reversing the uplink and downlink back to 
standard, then any Japanese CDMA phone today could roam abroad and any 
non-Japanese CDMA phone could roam in Japan, without the need of a 
dual-mode.

Roaming would be confined to just this smallest common denominator 2 MHz 
patch (actually less because of the way CDMA is designed) and that could 
at times mean congestion and a degradation of quality in this "roaming 
band", but it would work out of the box.

The reason why this was not done this way, is most likely because it 
would have required some shuffling around of most of the other patches. 
In any event, it will be even more difficult to do now, because there 
are many more cellphones in Japan today and bandwidth is scarce again. 
Shuffling patches of spectrum around for reassignment would require some 
capacity to be temporarily taken off the air. Also, new handsets would 
be required, at least for KDDI's CDMA service. So, the chances for such 
a realignment are rather dim.

However, a cleanup of Japanese cellular spectrum is overdue and may be 
undertaken some day (perhaps when there is an administration that is not 
a bunch of weenies who are afraid to make decisions - although at 
present the lookout on that seems like waiting for the Messiah).


In other words, making Japanese phones work in the US, the way you 
suggested isn't going to happen and making them work the way it should 
be doesn't seem to be happening any time soon.

There is however one piece of Japanese technology which has been totally 
overlooked and underestimated by the West and as a result now even the 
Japanese don't seem to be all to bullish about it anymore ... It is PHS.

PHS is one of those technologies, which have suffered a handicap as a 
result of being far ahead of their time. PHS has already implemented 
concepts that are still science fiction for the 3G standard developers, 
there is plenty of stuff on the "to do" or "future outlook" wish lists 
of 3G standards bodies and designers, which are a reality in PHS today.

In respect of roaming, three things are interesting about PHS:

- other than the Japanese cellular system (PDC) PHS has been a great 
international success (albeit as a cordless technology as opposed to 
cellular)

PHS is virtually everywhere on the planet other than in Europe and 
Korea. Last time I checked it had been in field trials in the US and 
IIRC it is now approved by the FCC.

- PHS is a supercharged cordless phone system

PHS does not assign spectrum to operators. The entire PHS band is shared 
by all operators and channels are assigned on demand when they are 
needed. PHS does not use paired spectrum. Transmission power is very low 
for both base and terminals. All this allows to operate PHS in 
unlicensed spectrum with little or no regulation of setting up hot spots.

- the Japanese have already got dual-mode PDC/PHS handsets

DoCoMo provides a service called Doccimo, whereby a dual mode PDC/PHS 
terminal can switch between PDC cellular, PHS public and PHS cordless. 
Japanese manufacturers have experience in the manufacturing of such dual 
mode cellular/PHS terminals.


This means that PHS could be used as a key to make Japanese handsets 
work in other countries the way you suggested. You could probably set up 
PHS hot spots in the US and serve Japanese visitors with a roaming 
service of some kind.

There are just under 6 million PHS users in Japan, PHS handsets are 
available from three different service operators (KDDI, DoCoMo, Astel) 
and PDC/PHS dual mode handsets are available from DoCoMo. Thus, there is 
enough opportunity for the average Japanese traveller to get a phone 
that could be used in overseas hot spots.

However, there are some obstacles to this ...

- Spectrum assignment

Unfortunately, the Japanese PHS spectrum is not exactly matching the US 
unlicensed spectrum set aside for cordless and any approval of PHS in 
the US by the FCC would have been for the US unlicensed spectrum, not 
for the Japanese PHS spectrum.

However, both the Japanese PHS spectrum and the US unlicensed spectrum 
are in the lower 1900 MHz band, so it would not be very difficult to 
adjust handsets such that they can operate in both bands, but 
adjustments would be inevitable.

- PHS has no roaming facility

In order to provide a seamless roaming service, you will need to be able 
to charge for roaming usage and also you may want to be able to provide 
roamers with the ability to receive incoming calls made to their PHS 
phone number back home.

To do both of this, a roaming facility is required. However, PHS, which 
had been designed as a lightweight system, does not have any 
international roaming facility, no protocol, no processes, nothing ! 
There were some white papers circulated within the PHS MoU some time ago 
on how roaming could be defined within the PHS standard, but nothing 
seems to have come of it.

Therefore, any company who wanted to operate PHS hot spots overseas to 
serve Japanese visitors, would be on their own. The most likely 
workaround would be to have the Japanese visitors sign up to get on the 
air, perhaps at a PHS desk at the airport when they arrive, but that 
means it would not be much different from a phone rental based service 
and given that the coverage will be limited it is unlikely to draw 
enough interest if there is a sign-up activity involved. Also, there is 
still the problem with how do you get incoming calls to be diverted from 
Japan to the hot spot networks for a roaming PHS user.


Mind you, your idea is not new. I have been working on an alternative 
roaming system which can solve the above mentioned problems and I have 
also worked on designs how this system could be used as a roaming 
facility for PHS. So far, there has been some interest in this by 
Chinese companies because PAS (as PHS is called in China) is the mobile 
phone of choice for most Chinese (less than 5% of Chinese can afford 
CDMA or GSM, but over 80% can afford PAS) and China's PAS was in fact 
the driver behind the efforts on PHS adaptation. However, there is no 
reason why this could not be used in a hot spot based service context in 
the US.

So, if you are serious about your idea to build roaming hot spots for 
Japanese visitors in the US, you may want to look into PHS and you may 
contact me off-list about the roaming technology you will most certainly 
need to make such a service seamless.

kind regards
benjamin
Received on Thu Jul 4 11:02:27 2002