(keitai-l) Re: International Usage

From: Benjamin <bkml_at_mac.com>
Date: 07/07/02
Message-Id: <39C38510-91D5-11D6-B314-003065FB21DC@mac.com>
On Saturday, July 6, 2002, at 11:23 , Ken Chang wrote:

> why should support a technology that has no market ... WAP?

You better check up on the background before you make like that. PHS has 
a huge market and is the main cordless technology and one of the major 
WLL technologies on the planet.

Your comparison with WAP is more likely to match DECT, the European flop 
that has clearly turned out the big looser in the cordless market. It 
does not, however, fit PHS, the winning technology, where PACS is 
already extinct and DECT is only still there because the Europeans have 
not allowed for it to be exposed to competition from PHS. DECT may still 
die nevertheless. It's been a big flop. PHS on the other hand has been a 
huge success.

No market ? far from it. UT Starcom, whose business is entirely PHS 
based are one of the world's most profitable wireless telecom companies.

> I happen to be one who don't like PHS so much,

What does it matter ? Just because you don't like it doesn't mean that 
you can dismiss it.

>  though there're
> a lot of clever designs

Indeed. For example the preservation of bandwidth. In fact PHS is the 
only technology out there today that has a sensible solution to the 
problem of bandwidth being a limited resource for which demand is on the 
rise.

And I repeat, the roadmap for the evolution of 3G to 4G which labs and 
radio agencies are working on today, pretty much describes a future 
technology that resembles PHS. The Japanese government funds a research 
program for 4G mobile and PHS technology is considered the basis for 
major parts of the program. PHS was and is ahead of its time.

Examples of items on the wish list for future (4G) mobile systems are

- do away with paired spectrum and fixed frequency separation
- dynamic channel allocation on demand
- operators to share the spectrum
- spectrum efficiency through SDMA

All of which is accomplished by PHS today.

>  and it's playing a good role as "poor
> man's WLAN".

This is just a value added service, not central to the concepts of PHS 
and it has absolutely nothing to do with what I am talking about. You 
may as well bring up the polycarbonate material from which the cases of 
PHS phones are made - it has nothing to do with the PHS technology. It 
is just a detail of it's roll-out in the market place. If there was no 
data services at all, still PHS would be groundbreaking because of the 
concepts listed above.

> PHS was designed as "poor man's mobile"

Definitely not. What you are talking about is perception not design. PHS 
was designed as a highly efficient and cost effective multi purpose 
wireless telephone system, where the main pillars of multi-purpose are

- WLL
- cordless consumer phone
- wireless PABX
- public mobile

Because of the latter three, I dubbed it "supercharged cordless".

In the early days of PHS, there were a number of limitations in favour 
of simplicity which have caused problems in the market, ie. the cell 
handover was too slow and calls dropped if you moved faster than bicycle 
speed. It is understandable that this has contributed to a poor man's 
mobile perception, but this is a misperception, particularly as such 
issues have been addressed and the PHS we have today is much more mature 
than it was back then. Likewise, early PDC/GSM/CDMA implementations had 
limitations and problems that were later improved upon.

>  and had a fundamental
> problem in what marketing position it should take

Again, you are talking about perception and not design.

>  - it never
> even had chance to compete with the costly PDC network.

Oh no ? Think again. DDI is returning a profit on their PHS service even 
though they operate in a market environment where a government mostly 
concerned with protecting DoCoMo from competition has fixed the charges 
that PHS operators can charge. On top of that, PHS operators have been 
severely limited by the government's policy of letting NTT maintain a 
stranglehold on the PHS operators (transit network).

And why did they do that ? Because they feared that PHS would have wiped 
out PDC, and they had the best reasons to believe that. You bet PHS 
would have long wiped out PDC if

- PHS operators wouldn't be strangled by NTT's monopoly on transit;
- PHS operators could set tariffs as they pleased.

In 1994 the Japanese government "forecasted" 6 million PHS users for the 
year 2000. And that is exactly how many there were. Do you think they 
had a magic ball to see those figures ? Why would their forecast have 
been so accurate for a technology that hadn't even been launched at the 
time ? Don't fool yourself, they have regulated the market and 
effectively throttled (or shall I say strangled) the PHS operators to 
match the forecasted figures.

Remind you that PDC operators had to bring prices down when they faced 
competition from PHS initially as they faced massive churn. However, PHS 
operators were not allowed to respond to that - they had to stay on the 
tariffs set by the government.

How could you possibly confuse that for "never had a chance to compete 
with PDC". Well, in a way you are right ... PHS never had a chance to 
compete with PDC because the government didn't allow the PHS operators 
to competitively respond.

But that doesn't mean PHS could not have if it only had been allowed to.

In Taiwan, FITEL is very successfully competing in a highly competitive 
GSM market even though it was launched at a time when the market had 
already had the second highest mobile penetration in the world after 
Finland (82% or so). At that point, nobody would have dared to enter a 
market near saturation if the business had been based on GSM. How is 
that for a technology that can't compete ?

In China the central government in Beijing is ***desperately*** trying 
to get rid of PHS, but various attempts have failed. The mostly 
government owned GSM and CDMA operators have no other way to deal with 
the threat from PHS other than pushing for government restrictions. And 
this despite the fact that the cellular industry in China has access to 
capital and government assistance, while the PHS services operated by 
the metropolitan telecom operators have to deal with all kinds of 
artificial obstacles and get little or no funding.

Without PHS's incredible price/performance ratio, 80% of Chinese would 
not be able to afford mobile telephone services. These people can afford 
mobile services because of PHS. They could never afford CDMA or GSM 
services.

The Beijing government has called PHS a technology with an unfair 
advantage over CDMA and GSM and they have even tried to shut the PHS 
services down. But PHS is still there because every serious attempt by 
the government to shut down or limit services has resulted in public 
uprising and the government had to back down. How is that for a 
technology that can't compete ?

>   it's
> market overlaps more with the public phone boxes.

Now you are simply repeating the Japanese politician's PR statements and 
-with all due respect- it is nothing but blah blah. I could also state 
that any CDMA, GSM or PDC phone service overlaps with public phone boxes 
and back this even up with statistics from all over the world where 
public phones have been on the decline since the introduction of mobile 
phone networks (independent of technology). If anything at all the 
statement only confirms that PHS is an equal member of the family of 
mobile technologies.

> in the early 1990s, 3 operators provided CT-2 service in the UK
> and all of them bankrupted 3 years later when the Japanese
> started to roll out the PHS service in 1995.

You make it look like the launch of PHS was the reason for the downfall 
of CT-2. In fact the two have nothing to do with each other 
***whatsoever***; they are totally unrelated.

Not only are the technologies totally different, but also the way the 
Rabbit service in the UK was designed is totally different from PHS. In 
fact it could not possibly more different.

Rabbit was a service that offered only outgoing calls. You could not 
receive calls. Also, it was a hot spot network. You could only make 
calls in a few meters distance from selected public phone boxes. And 
even then, it was often not possible to make a call because someone in 
the public phone box serving your CT-2 phone already occupied the line 
and you may as well have queued to get into the phone box and not use 
the Rabbit service, which is what people eventually did in their 
frustration of never being able to make any calls.

CT-2, as a technology was very much inferior to PHS, but the way the 
Rabbit service had been implemented was absolutely nonsensical and the 
demise could have been predicted in advance.

It is totally misleading to compare CT-2 and PHS (the technologies). The 
two have nothing in common. As a matter of fact, PHS had borrowed a lot 
from GSM. However, It is even more misleading to compare Rabbit (the 
CT-2 based service in the UK) with PHS services in Japan or anywhere 
else. There is not the faintest similarity.

>   now all the three
> Japanese operators had already bankrupted in 1998-99, ... in a
> Japanese way.

Only two of the PHS services incurred losses. DDI Pockets started to 
return a profit after three years of service. Also, even though Astel 
and NTT Personal did not return profits on a country wide basis, 
individual regions within the respective groups did become profitable. 
This all shows that it is not the technology itself which has caused 
losses to be incurred.

Besides, TuKa is a good example for a non-performing technically 
bankrupt PDC operator.

In China, the cost for a call on a PHS service is a fraction of 1 yen 
although most of the infrastructure comes from Japan, Taiwan and the US 
just like GSM and CDMA infrastructure comes from foreign vendors. GSM 
and CDMA tariffs are similar to cellular tariffs in other countries. It 
couldn't possibly be local labour cost that is responsible as whatever 
cost advantage there is in China applies to GSM, CDMA and PHS in China.

In 1995, PHS operators trusted assurances by the Japanese government to 
introduce competition and gradually remove NTT's stranglehold (transit 
network). However, the government did not deliver on their promises and 
they basically screwed their own people here. It took a threat from the 
US government to impose sanctions for the Japanese government to finally 
start to get going on wholesale rates (last year). Only now do we begin 
to see wholesale rates to be lowered gradually. Had the PHS operators 
had the benefit of hindsight, they would most likely have invested in 
building an alternative backbone instead of waiting for the Japanese 
government to fulfill its promises. UT Starcom, the US company which 
builds the PHS infrastructure in China is best proof for this. They 
learned from the Japanese situation and built networks such that they 
would not be exposed to bullying from government owned competitors which 
made PHS services in China even more competitive and pretty much 
resilient against government policies designed to strangle PHS services.

> before we do anything, we should have a good study of the market,
> set conditions as the base of business/technology development.

The market for PHS is there. Particularly in Japan where most of the 
population is concentrated in regional clusters. Also, the cellular 
companies in many countries have started to mimic PHS's ability to roam 
between house/office and public service (VHE). However, the cellular way 
is to use location information to accomplish this. You simply get a 
cheap tariff when you are close to your house or in your house, you do 
not actually use your landline, you are still on cellular service and 
there is no cost advantage as far as the network is concerned.

Besides, if we really "study the market" in Japan, we see that TuKa (on 
PDC) is probably the world's worst performing cellular phone service. 
This tells us, that it is not the technology that makes the difference 
here. TuKa is PDC based and it is technically bankrupt, while DDI 
Pockets is PHS based and it is profitable. At the same time there is 
DoCoMo and Astel, showing the opposite pattern of a profitable PDC 
network and a mostly loss leading PHS service.

However, in the PHS segment of the market, it doesn't matter that Astel 
is loosing money because they do not waste any spectrum, thanks to the 
way PHS was designed. TuKA on the other hand is an incredible waste of 
spectrum while DoCoMo desperately needs more spectrum. With PDC like 
other cellular technologies, spectrum is assigned to operators. 
Therefore there is no easy way to let DoCoMo use part of the TuKa 
spectrum, at least that which TuKa doesn't really need due to their 
underperforming in the market.

> in wireless, the use patterns of the voice and data services are
> a little bit different in that the voice needs more coverage and
> mobility.  for voice and light data we have GSM already and for
> heavier data we have Wi-Fi, where is the room left for PHS?

PHS concepts are the basis of G4 mobile. So the answer to your question 
is =while unlikely to be called PHS then= that eventually, PDC, CDMA, 
GSM and UMTS will be replaced by G4 mobile technology that will more 
closely resemble the PHS of today than any of the others.

Besides, there is no good reason to assume that any one mobile 
technology will eventually serve all applications. In urban centres with 
dense population, clearly PHS is the better concept while in rural areas 
something more closely resembling CDMA-450 or GSM-450 is better suited 
and cost effective than for example UMTS. After all, UMTS is not a new 
standard for one single wireless system but instead a meta-standard of 
multiple systems and how they will hopefully inter-operate. While this 
is in principle not such a bad idea, the implementation will have to go 
through various evolutionary steps before the meta-standard portfolio is 
composed of the most suitable technologies for the various areas of 
application.

It is reasonable to assume that 4G will be a result of learning from the 
experience of 3G and more closely resemble a portfolio of technologies 
each one close to best suitable for a particular application. At present 
the vision for this portfolio is

- WLAN systems for data services in urban areas
- W-CDMA system for data services in areas with no WLAN coverage
- PHS like telephone system for densely populated urban areas
- long range maxi-cell based system and VSAT for rural areas

As most of the wealthy population is in urban areas, it would seem that 
PHS and its successors is closer to the real needs of the market than 
most of what we see today.

> some words about WAP & i-mode again:

As I said, data service are just another value added service and the 
relationship is the same to any wireless phone service, whether that is 
PHS or PDC or GSM or CDMA doesn't matter. PHS is a ground breaking 
technology because of its design objectives and accomplishments as 
described above; not because PHS operators in Japan have found a market 
by providing high speed data over PHS. Even if PHS was strictly voice 
only, the concepts behind it would still be valid and make it superior 
to any other wireless phone system there is today.

> PHS is more expensive than PDC if one tries to provide similar
> coverage,

Why don't you use a satellite phone all the time ? It's the most cost 
efficient way to cover the entire planet you know ?!

Ahhh, you don't think it is worth while paying the higher cost *all the 
time* where another technology could offer more cost efficient 
service ?! So, you don't want to pay for the odd chance you may be in 
the middle of the desert or ocean etc etc etc ?!

Well, the same principle applies to PHS versus cellular. In most urban 
areas, even in most of the Japanese country side, PHS is more cost 
efficient. Wherever there are buildings, PHS can be deployed cost 
efficiently. As for that phone call while on a walk in the woods 
kilometers away from the next village ... there are so few people using 
a phone service in such areas, that even today's cellular systems which 
have a typical range of about 35 km can often not provide cost efficient 
cover.

Under the 4G vision, those areas would be served with a celluar system 
that trades in long range (to reduce the number of radio towers) for 
capacity (as there will be fewer people using the service). Thus, you 
would have pico-cell low power PHS like (no paired spectrum and dynamic 
channel assignment) in cities, and perhaps GSM-450 with 150 Km range* in 
rural areas. An SDR handset would then switch between the different 
protocols. Services will be priced according to the cost/benefit ratio 
of the area you are in. Just like today, where a satellite phone call 
from the middle of the Pacific will cost you $$$ and a call from a 
mobile phone in HKG costs a fraction of that because there are many 
folks and networks to create economies of scale. On the Ocean however, 
you won't have economies of scale and therefore it will remain more 
expensive than most land based services, no matter how far prices will 
come down (given latest ideas to launch self navigating balloons etc etc 
etc).

* Ericsson Australia has tested a customised GSM-450 with reduced number 
of time slots in order to allow for a longer range and their tests have 
shown reliable service with distances of up to 135 Km from the base. 
Contrary to common belief, it is the number of time slots which limits 
the range of (TDMA based) cellular services, not so much the radio 
signal itself.

>  and its last stronghold, the 64 kb/s data, will be
> eliminated by WLAN in half to one year's time

Diminshed, yes, very likely - Eliminated, no, most certainly not.
WLAN will need much longer to get the same coverage and with an 
inflation in the number of hot spots and operators, it will take years 
to sort out billing and settlement or until consolidation kicks in to 
avoid the issue of bazillions of roaming agreements and clearing houses.

>  (you can have
> unlimited voice/video phone calls with WLAN).

Not with the reliability of either PHS or 3G.

Why are people still using international phone services where there are 
close to zero cost internet based phone services available, ie 1 cent / 
minute to anywhere in the US. Sure, the internet telephony services have 
made an impact, but traditional long distance is still around.

In fact, if WLAN becomes ubiquitous, PHS is likely to benefit, while 3G 
is likely to suffer. UT Starcom's PHS system has been modified to use IP 
as a backbone for transit instead of the Japanese PHS system which uses 
ISDN and which is why NTT has a stranglehold on PHS operators here.

Let's do an imagination experiment, shall we ?!

Let's assume that in a parallel universe, there is a country called 
Japan' which is pretty much like the Japan we know, except that, a US' 
company, UT Starcom' has bought one of the Japanese' PHS networks, say 
Astel'. Now, UTS' would take the PHS network of Astel' and rebuild the 
backbone based on their technology which they deployed in China'. And 
they would be able to buy bandwidth cheaply from a new WiFi provider 
operating a 100Mbit/s wireless backbone in the 5GHz band offering 
amongst other services, corporate VPN. UTS' uses this to connect all of 
their PHS base stations to their switching centres, in the process 
cutting out NTT' completely and reducing the cost of phone calls so 
drastically that they can undercut domestic phone tariffs of NTT'. Now, 
you pay -say- 3 yen for a phone call on your PHS from anywhere to 
anywhere in populated Japan' instead of NTT' tariffs, starting at 10 
yen. In other words, if the end user on the street can use WLAN, so can 
a PHS operator.

This is not as hypothetical as it would seem because in China, UT 
Starcom are doing something pretty similar, just with the difference 
that they don't have a wireless but a fibre based backbone. Obviously, 
if PHS bases were to be connected via WLAN to their backbone, you could 
save even more money because there is no wiring required. In many cases, 
you may even be able to solar power the entire base, so you don't even 
require a power line.

In fact that is where the Bangladeshi's are heading for with their PHS: 
Cheap phone coverage in rural areas where the last mile is PHS, the 
backbone is wireless, too and the village's one or two PHS base stations 
are solar powered.

So much for your assumption that PHS is too expensive to provide 
coverage where cellular can. More often quite the opposite is the case, 
particular when combined with wireless IP based backbone infrastructure.

> thanks to PHS, Japanese mobile operators used to be reluctant
> to extend repeaters and leak-cables to the underground, that
> Japan is still suffering.

I assume what you mean to say is, that because PHS was able to provide a 
cost efficient internet service equal or above modem speeds, nobody 
bothered to try to compete with PHS and roll out other means of internet 
service provision.

Do you realise that this is negating what you started out with ? that 
PHS was not cost efficient and it wasn't capable to compete ? Now you 
are basically saying that PHS is so cost efficient that it gained an 
unfair advantage nobody else could match.

Q.E.D.



In any event, what I was talking about before in this thread was PHS as 
a technology which could be deployed as self sufficient hot spot 
networks in places of interest to Japanese visitors overseas, in 
particular in the US. This is not because PHS has a better roaming 
facility than GSM -in fact it doesn't- but because there is no other 
technology deployed in Japan which would make roaming hot spots possible 
without spectrum conflicts and major cost.

Let's assume, you deploy PHS in the major casinos and hotels in Las 
Vegas in order to accommodate Japanese tourists. You could do so with 
PHS because it has been approved by the FCC - you couldn't do so with 
PDC and Japanese CDMA because you'd need spectrum which is already 
assigned elsewhere. Very likely you would deploy UT Starcom's PHS system 
using IP instead of ISDN as a backbone. What's more, the hotels could 
run such a service as a wireless PABX within their buildings and have 
PHS handsets in the rooms for the perusal of not only Japanese visitors, 
but also any other guests. The investment to deploy such as system is 
not much different from upgrading the in-house wireline phone system 
with new terminals. A hotel could be tempted to do so, simply because 
they can expect more usage if guests can carry their hotel phone with 
them. Also, people would probably favour the hotel's PHS phone over 
their mobile because they wouldn't get charged for receiving a call.

It is likely to be more expensive to upgrade all the TV sets in a hotel 
than to deploy a PHS based wireless PABX and set up roaming agreements 
with other hotels. The hotels in Las Vegas are mostly owned by consortia 
anyway, so doing something like this on the corporate level will make 
roaming agreements and settlement far more easy than a roaming agreement 
between two different VodaFone owned networks. Now, what is the ROI for 
TV sets in hotel rooms ? Whatever it is, I bet that a PHS based hotel 
telephone system has a far better ROI.

As this scenario shows, there could be cases where PHS hot spots in the 
US may be viable even without taking Japanese roamers into account for 
revenue. So, don't dismiss PHS in the context of hot spots so easily, 
just because you don't like it.

Sure, I hear you when you say that you could do this with WiFi. In fact, 
I remember that we had a heated discussion last year here on Keitai-L 
where I was arguing that WLAN would become a threat to 3G not only in 
respect of data services but also in the context of voice. I was pretty 
alone back then. The consensus on the list was that it would not make 
sense for any operator to make the investment needed for soooo many base 
stations. One year later we have exactly that happening. However, back 
then, I said that it wasn't even necessary to have operators deploy base 
stations. I said that eventually every coffee shop, bar, restaurant, 
hotel lobby etc etc would deploy a 200$ base simply because they want to 
attract customers and they wouldn't need to charge for the WLAN service 
they provide. I described a scenario where WLAN would eventually become 
ubiquitous and I said that this would be the point when internet phone 
operators would probably show up to provide VoIP over WLAN whereby 
initially IP telephony client software on notebooks and PDAs would be 
used but eventually a new device could appear, an WLAN IP mobile phone. 
People then quoted the UK's CT-2 disaster (as you did here) to proof 
that this could never happen and if it did it would flop like CT-2. I 
then provided some rough ideas of how it would be possible to deliver 
inbound calls to a mobile IP phone roaming between various different hot 
spots, such as a forwarding service bound to a universal phone number 
from a unified messaging service provider where the incoming calls to 
the universal phone number would either be forwarded via IP if the WLAN 
mobile phone had notified the UMS service of its whereabouts and IP 
address or a voicemail service would kick in to then take a voice 
message which would be delivered to the IP phone as soon as it shows up 
in another hot spot. I also said that such a mobile IP phone would 
probably come as a dual mode phone where one mode was traditional mobile 
mode (ie GSM or CDMA or even PHS) and the other mode was VoIP over WiFi.

So, as you can see, I am pretty much on your wavelength here.

However, you have to realise that developments like this take their time 
to reach the necessary critical mass to convince vendors to develop and 
manufacture products. And from experience we should realise that there 
will always be multiple alternative technologies with overlapping 
applications.

The situation with PHS is that all the technology is already there and 
it has reached the critical mass needed to get economies of scale. Even 
the dual mode terminals for PHS/GSM and PHS/PDC are there. In fact, one 
of the main reasons why many countries have approved PHS is the ability 
to provide a simple solution of a solid Virtual Home Environment, 
whereby you use PHS as a cordless phone at home, on the wireless PABX in 
your companies offices (even worldwide) and GSM when you are outside of 
these areas. VHE is more and more considered a future killer app and 
anything based on 2G or 3G is a fake as it relies on location based 
services to pinpoint you to your home or office and service still comes 
from the next radio tower. The Europeans are trying to place DECT but 
they have basically lost the battle for market share to PHS a long long 
time ago. Outside of Europe, PHS is the technology for this, not DECT. 
So, even if the Japanese PHS operators were to close shop as far as 
public service is concerned, there would still be a case for PHS hot 
spots, or perhaps, the case for PHS hot spots might even be stronger. 
Either way, you can't dismiss PHS as easily as you did.

regards
benjamin
Received on Sun Jul 7 21:14:21 2002