40 Million Euros in Funding for Galileo Apps - and Some Hard Questions

40 Million Euros in Funding for Galileo Apps - and Some Hard Questions
http://sidt.gpsworld.com/gpssidt/content/printContentPopup.jsp?id=3D57=
8617

Jan 30, 2009
By: Alan Cameron
GPS World

This column brings you an excerpt from the panel I moderated January 27, =
in the=20
Charlemagne Conference Center in Brussels. The Growing Galileo 2009 Confe=
rence was=20
sponsored by the European GNSS Supervisory Authority, on the occasion of =
the FP7 Second=20
Call for Galileo exploratory applications. Representatives of NavTeq, STM=
icroelectronics,=20
eRide, and SANEF gave their views on the GNSS business climate, related t=
heir successes=20
using GPS, and looked ahead to the prospects of the sat nav market and Ga=
lileo=20
market-readiness.

This month only, the new Ask Galileo feature becomes Galileo Asks, focusi=
ng primarily on=20
the question, =E2=80=9CWho will buy?"

Alan Cameron: What do you think will be the strategy of chipset suppliers=
concerning the=20
integration of GNSS signals in chip?

Option A: Every chipset will be enabled for all four potential constellat=
ions (GPS,=20
Galileo, GLONASS, Compass) even if they are not totally compatible.
Option B: Chipsets will only be enabled for those GNSS signals that are f=
ully compatible=20
and interoperable.
Option C: Chipsets will be enabled with GNSS signals according to the req=
uirements of the=20
application integrators.

Philip Mattos, Chief Engineer, GPS and Navigation, STMicroelectronics R&D=
Ltd: It=E2=80=99s=20
largely a marketing issue. We see that the people who want us to put a fu=
nction into the=20
chipset, not actually because the function is usable today, but because t=
hey feel they=E2=80=99ll=20
sell their product more easily if they tick the box on the side that says=
=E2=80=9CEGNOS-ready=E2=80=9D or=20
actually =E2=80=9CWAAS-ready=E2=80=9D in the States, it=E2=80=99s the sam=
e signal, it=E2=80=99s compatible. So they can=20
tick the box and say =E2=80=9Csupports WAAS/EGNOS.=E2=80=9D This year the=
y can tick the box and say=20
=E2=80=9CGalileo-ready.=E2=80=9D That is a very important function that w=
ill bias it toward the =E2=80=9Cput=20
everything in anyway=E2=80=9D discussion.

We have to look at the costs of putting it in, and I mean the costs on a =
per-chip basis.=20
Putting GLONASS FDMA in is very expensive. So whereas we=E2=80=99re study=
ing that option, I can=20
quite confidently see that our manufacturing partners that we sell to wil=
l finally come to=20
an agreement that, no, the extra cost of putting FDMA in is not worthwhil=
e. But we are=20
doing the study.

On the other hand, the cost of putting Galileo into a combined chipset wi=
th GPS is almost=20
zero. It=E2=80=99s only development costs. Yes, there=E2=80=99s a tiny co=
st because we need to sample a=20
little bit faster, very little. We need a little bit of memory for the me=
mory codes=20
instead of chip register codes. But whereas that might have made a 10 per=
cent difference=20
in a GPS chipset, we don=E2=80=99t make GPS chipsets, we make an entire a=
pplication chipset. So if=20
GPS was only a quarter of the chip, and we=E2=80=99re making a 10 percent=
difference to that,=20
that=E2=80=99s a 2.5 percent difference to the whole chip. It=E2=80=99s t=
rivial, and well worth getting=20
the ticked box.

We think that nearly everything that would have been a GPS chipset will h=
ave Galileo in=20
it, from sometime in 2010. We=E2=80=99ll have ours out in 2009. By 2010, =
people will start putting=20
it into every new chipset.

I=E2=80=99m very much of two views on GLONASS. I would say when GLONASS g=
oes to CDMA, everyone=20
will put it in. But while it=E2=80=99s FDMA, I=E2=80=99m slightly biased =
against. We=E2=80=99re doing a study of=20
it, but only a few people in the highest-priced markets will be willing t=
o put it in.

As for the Chinese system, at the moment I think a definite no, based on =
a lack of=20
knowledge. Not on my lack of knowledge, but a lack of knowledge of the ma=
rket. When we=20
fully understand the spec of the signal, the availability and so on, and =
how much drive=20
there will actually be from China =E2=80=94 will we be able to sell chips=
ets into China? Because=20
there will be a huge patriotic drive, everything going to China will have=
to have that in=20
it. So I won=E2=80=99t say yes or no, I=E2=80=99ll just say, open mind on=
that one.

Slightly negative on GLONASS, but totally positive on Galileo. Every chip=
set will do both=20
[GPS and Galileo]. And EGNOS is already in, it=E2=80=99s just as good. To=
load the EGNOS software=20
or not load the EGNOS software, it=E2=80=99s actually in the hardware of =
every chip.

Moni Malek, General Manager and Board Member, eRide Europe GmbH: I agree =
with this as=20
well, In fact, we also have EGNOS in every chip, it=E2=80=99s easy to swi=
tch on. He=E2=80=99s right, it=E2=80=99s=20
going to cost 10 percent more on a GPS chip, on a full chip it=E2=80=99s =
2.5 percent. For the=20
benefit of supporting what we believe is a future standard. It=E2=80=99s =
definitely going to happen.

Compass, we=E2=80=99ll have to see how the market develops. GLONASS, FDMA=
is probably too=20
expensive. I=E2=80=99ll let you deal with that.

Alan Cameron: Before I ask the next question, I remind the audience, when=
you hear the=20
word LBS, think also Galileo, LBS uptake, Galileo uptake. The questions a=
nd the answers go=20
cross-platform, so to speak. To substitute Galileo for LBS, you might mov=
e the timeframe=20
out one or two, or three or four years. I believe many of the issues will=
be the same.

[Note: In a presentation regarding the FP7 Second Call for application pr=
oposals, Boris=20
Kennes, applications officer, European GNSS Supervisory Authority, stated=
that LBS is=20
expected to be the most important (that is, generating the most revenue) =
application of=20
Galileo.]

What is currently the challenge with LBS uptake from the consumer perspec=
tive?

Pia Vuohelainen, Partner Manager, Partner & Developer Programs, NAVTEQ: W=
hen you look at=20
current devices, there are still minor things with the battery life, very=
often the=20
battery drains, and also getting the fix very fast, the GPS fix when you =
are in the urban=20
areas, that takes time. These are the type of things that Galileo will ad=
d value.

I use a lot of these navigation tools [on my mobile]. Nowadays they are w=
orking better and=20
better because there are improvements like a compass, because then my dev=
ice knows where=20
north is and south is, so that helps quite a bit in the overall user expe=
rience, because=20
the overall user experience is the key.

I think one of the reasons that people aren=E2=80=99t using these service=
s is that they are not=20
aware of these services, because they haven=E2=80=99t been using them, th=
ey have no experience. I=20
think we should be the advocates, going around showing that these are the=
things that are=20
possible today, and these will be enhanced by Galileo in the future.

Moni Malek: I cannot agree more. We are doing some trials, and the result=
s are quite=20
interesting. Pia=E2=80=99s right, the power consumption is still a proble=
m. If you switch on your=20
iPhone with GPS, it probably runs out the iPhone in five hours, so that=E2=
=80=99s not too=20
exciting. We=E2=80=99re solving that with AGPS. What we do is you have tw=
o phones running this=20
Friend-Finder application, and the first thing is, you don=E2=80=99t run =
GPS, you first look at=20
the cell IDs. When the friends have the same cell ID, then you switch to =
an AGPS mode, and=20
that way, we can have a friends-finder application running for the whole =
length of the=20
phone, without wasting GPS power. It=E2=80=99s clear, when you=E2=80=99re=
not in the same cell, you=E2=80=99re not=20
going to get near to each other. We can get a fix when you=E2=80=99re in =
two adjoining cells as=20
well. That needs a bit of clever software, there are companies who are do=
ing that, I=E2=80=99m a=20
business angel as well and I invest in that kind of company. You=E2=80=99=
ve got to get that feel,=20
because that=E2=80=99s the haptic system solution to solve that kind of p=
ower problem [haptic=20
technology or communication interfaces with the user through the sense of=
touch =E2=80=93 Ed.].=20
You don=E2=80=99t solve it by running GPS all the time and waiting for yo=
ur friend to walk by.

The second thing with friends-finder [as an example of LBS uptake], if yo=
u look at our=20
generation, I want to know where my friends are but I don=E2=80=99t reall=
y want other people to=20
know where I am. So you use that attitude, the friends finder is dead.

If you go to the young people, it=E2=80=99s the other way round. Their so=
cial network says, if=20
people don=E2=80=99t know where I am, no one=E2=80=99s going to call me. =
So they want their information=20
exposed, it=E2=80=99s completely the reverse. The new generation of kids =
are afraid that people,=20
that their friends don=E2=80=99t know where they are, they want to be ope=
n. That makes the friends=20
finder for them realistic.

It=E2=80=99s getting all those things right, together, that enable an app=
lication. We=E2=80=99re doing=20
trials now, we=E2=80=99ve combined with Facebook, the major social networ=
king. You need companies=20
that combine the application, that create the application and combine it =
[with the right=20
partners]. You go to the iPhone and you have trouble putting background a=
pplications in=20
because Steve Jobs controls everything. But we can do a good job on a G p=
hone, that=E2=80=99s the=20
Google phone, which is coming out pretty soon.

So, keep your fingers crossed and we=E2=80=99ll get it all right. With ce=
ll IDs, with GPS, the=20
trick is that kids want their friends to know where they are, they don=E2=
=80=99t mind exporting=20
their data. So you get all these things together, they=E2=80=99re making =
switches in the=20
availability of LBS.

Pia Vuohelainen: And I think the last point is the availability of the ma=
p data, remaining=20
in the device when you get it. Or some kind of a map application, the act=
ual data doesn=E2=80=99t=20
need to reside in the device, but actually in the application. The consum=
er doesn=E2=80=99t know=20
where to download this type of application, downloading software to the d=
evice might be a=20
big inhibitor.

AC: Your answers to my question seem to revolve around capabilities. From=
my viewpoint,=20
the lack of uptake of LBS services, and by extension of Galileo services =
to the consumer=20
in the future, stem from the fact that no one wants to buy, no one wants =
to pay $9.99 or=20
10 euros per month for a location service. They want it to come free with=
the contract=20
they=E2=80=99re already paying, or with what they=E2=80=99ve paid already=
for the hardware. They don=E2=80=99t=20
want to pay per use, they don=E2=80=99t want to pay per month. That I=E2=80=
=99ve heard so far, there are=20
two models for paying for LBS, either the user pays, or second, is advert=
ising, some form=20
that the user consents to receive in order to get this free location serv=
ice. Is this the=20
only other model? Who=E2=80=99s paying, and how are they to be made willi=
ng to pay? Because right=20
now, nobody=E2=80=99s willing.

Earlier, I quoted from a market research study of mobile subscribers down=
loading=20
navigation routes, which while predicting tremendous growth in user numbe=
rs and revenues=20
over the next three years, also admitted that =E2=80=9Cmost mobile naviga=
tion users only use the=20
service during a free trial period. The key challenge will be to convert =
these test-users=20
into paying subscribers."

Is this what Galileo faces once it becomes available to consumers =E2=80=94=
that everyone wants it=20
for free? They will get their location for free, obviously, on the Open S=
ervice, but no=20
one will be willing to pay additional for any service that actually makes=
that location=20
useful in some way. Thus no Galileo uptake.

Moni Mallek: Actually, it=E2=80=99s strange. Take the friends finder serv=
ice. There=E2=80=99s an SMS=20
service, you can ask for the cell ID of your friend, and the friend has t=
o accept that=20
their location is given, and that works with kids now. They sell applicat=
ions and you=20
download it, and the first five or ten are free, and after that you pay 2=
0 cents, and=20
they=E2=80=99re finding that people still are prepared to pay 20 cents to=
find their friends. They=20
get the first finds free, and they say, this is a good application, I kno=
w where my wife is.

Pia Vuohelainen: I actually don=E2=80=99t agree with you, Alan, that no o=
ne will be willing to pay=20
for these services. Are you sure of that? Because I work with a lot of ap=
plication=20
developers, and I=E2=80=99m seeing that consumers are not wiling to pay h=
uge amounts of money. But=20
for example a child-tracking application, the kids are very dear to us, s=
o are our elderly=20
people. There are things that we are very willing to pay, if they are wor=
king well.

Also when I look at some of the fleet management activities, I have custo=
mers in Russia=20
and one of their views was that with a very nice fleet management, they s=
tarted to save 30=20
percent in fuel costs. I said, that=E2=80=99s a bit of a high number, and=
the person actually told=20
me, it=E2=80=99s not about being more efficient in routing, it=E2=80=99s =
about preventing reselling of the=20
fuel from the trucks. So there are clearly certain applications that have=
clear paybacks,=20
where you can easily calculate the ROI.

I believe the willingness to pay is there.

Lastly, the advertising. Advertising is always about the numbers game. Fr=
om the=20
advertisers=E2=80=99 perspective, they still require a large number of ey=
eballs in order to get=20
into a certain model. At the moment we are still lacking an advertising a=
ggregator model=20
for all of our services within Europe, but there will be companies going =
into that space.

Question from the Audience

Jac Spaans, Netherlands Institute of Navigation: I would like to come bac=
k to the earlier=20
question on interoperability, compatibility. I think the compatibility of=
the four systems=20
is sort of guaranteed in the ICG, the International Committee on GNSS, wh=
ere these issues=20
are discussed. There will be no interference or jamming of signals. The i=
nteroperability,=20
I fully agree with what Philip said, I think what he said is valid for th=
e Open Service.=20
My question is, what=E2=80=99s going to happen with the Commercial Servic=
e, and Safety-of-Life=20
Service. If I want to certify my safety-of-life system as a combined rece=
iver of GPS and=20
Galileo, I get integrity from Galileo, but I don=E2=80=99t get it from GP=
S.

In these applications, will we have combined receivers, or do we have Gal=
ileo-onlys? And=20
this is valid for the Commercial Service, for PRS, and for safety-of-life=
=2E

Philip Mattos: Initially, I should say my expertise is entirely the mass =
market, of course=20
it=E2=80=99s the Open Service. Talking from the safety point of view, the=
situation would be=20
similar, except for the problems of different bands and so on. Because ma=
ss-market chips=20
can=E2=80=99t handle different frequency bands. But the other services ar=
e largely a matter of=20
certification and so on, and I see it as a premium item that there will n=
ever be a=20
combined ICD (interface control document) for all four services. So if yo=
u did go into=20
certified mode, as it were, your box would have to do four separate outpu=
ts, with a little=20
NME to get each output to say valid, valid, valid, valid, and it could ne=
ver give you a=20
combined output in certified fashion. But I do say that=E2=80=99s not my =
area of expertise.

Moni Malek: I=E2=80=99d just say that EGNOS gives you the integrity signa=
l for GPS as well, so=20
there is some value in that. As Philip said, each system in the certified=
mode has to do=20
what it can, it can=E2=80=99t do anything more. You can=E2=80=99t find a =
system that solves everyone in=20
this mode. The open signal is the most important one.