ParkerVision 2005 AEA
November 8, 2005
Jeff Parker, CEO ParkerVision
Jeff Parker: Thank you for joining us at
AeA today. This is going to be webcast so just FYI we will put this up on our
website for a while so anybody that wants to see the presentation or whoever
wants to look at this, it will be available.
The theme of our discussion today and this is something that you, probably many
of you, saw introduced on our new website a couple weeks ago, is breaking the
RF rules. This company has invested a lot of time and energy and dollars in
creating a new architecture and a new way of processing radio signals and
really breaking out of the old RF architectures and the rules that those
architectures inherently limit a whole variety of important aspects up and down
the food chain, whether it's consumers who want more battery life or network
service providers who are trying to get more capacity on their network, or more
features in their handsets. Of course, there's our obligatory partner
statement. You know, you have seen that. We can move on.
So let's talk about the, a little bit about our discussion today is going to
walk through. We're going to talk a little bit about our focus as a company. As
you know we have exited the retail market that we were working on last year
with some of our WiFi products that used our energy signal processing
technology and we have now moved to a pure focus as an IP and technology
provider to the OEM marketplace. We will discuss that further.
We are going to walk through what makes the company's opportunity as robust and
exciting as it is, based on our technologies. What is energy signaling
processing? How does that break down into the two macro components of a direct
to power technology for power amplification and direct to data for RF receiver
applications and what we have done to protect the intellectual properties that
we've developed.
Our target market we're going to discuss a little bit. What markets are we
going after? How big are those markets? What are the trends in those markets?
Why do we have a competitive advantage? Why ParkerVision, why now? Why do we
believe we can win in these markets? And market strategies: how do we fit into
what our customers are trying to do? How do we help them get to their goals?
The convergence of technology and market needs I think is the most exciting
aspect of where we are as a company today. This is not a technology we have
developed that is multiple years ahead of the needs of the market. We really
are converging with where OEM and service providers are trying to go and really
fitting the market needs for people who are trying to get to the next step,
especially in 3G handsets and beyond.
What do you watch for in our company from here? What's going to happen over the
next quarter and two quarters and three quarters? Perhaps up through the end of
next year.
Our focus as a company is, very focused on commercialization of our proprietary
RF technology products and how we can advance those OEM products and services
that people are working on for the next generation. If we kind of step back a
year, at last AeA, you would have heard a couple of interesting questions again
and again at every presentation.
Question No. 1: Why are you pursuing a retail business model? If I didn't hear
that questions fifty times I didn't hear it once. And we explained then that it
was a very important goal for the company to showcase our technology and a
product that people could get their arms around and understand that energy
signal processing. hat we have done in converting old tired analog radio
architecture to digital, can bring spectacular benefits that you can really see
and feel and touch, no matter where you are in the food chain. We'll talk a
little more about that, but this was extremely helpful in getting the traction
that OEM marketing has today in our discussion and I'll talk a bit more about
that later.
And then the second question has occurred again and again is "Hey, what's
your strategy for pursuing a broader OEM opportunity? If your chips and
technology can be as good as they claim they are, why aren't you guys focused
on the OEM?" What we explained back then was we were working diligently to
advance our technology to a place we thought OEMs would really get engaged with
us and especially the top tier OEMs that we wanted to target. It was maybe two,
three, four months after AeA that we announced our Direct to Power technology.
We did get engaged in dialog with OEMs. A couple months after that, we decided
that this is absolutely the time for this company to focus on a pure OEM play
and we exited retail and today we are pure IP and technology centric company
pursuing OEM design.
Since that d2p announcement back in the January or February time frame of this
year, we've narrowed our focus out of the retail activities. Everybody in this
company is 100% focused on the sales, the marketing, the development,
especially of the piece of our technology that we call the d2p or Direct to
Power which is a technology that takes data signals, could be analog data,
could be digital data, and directly converts them in a single unified to a
radio carrier and tower, we're going to talk a lot about that here in a few
minutes. In the process, we've been able to reduce our use of capital about 25%
because we no longer have to support the marketing expenses of a retail
campaign. All of the senior staff in this company is focused exclusively on the
business development opportunities and activities of an OEM business.
After we made those introductions the d2p in press releases in the beginning of
this year shortly thereafter we were lined up to start doing demonstrations and
really then some of those have transitioned into business discussions ‑
how we can do business with these OEMs, but our year has been marked in the
majority of the year with the out with the OEMs and showing them the technology
actively engaged in how the technology can help their next generation products
and what kind of business relationships they're looking for from ParkerVision.
What's our near term focus? Our near term focus, as I've already publicly
stated multiple times, is secure design for the technology. It's great that we
have a wonderful technology, but now its time to start getting the OEMs to
commit to embracing this inside their products.
The way we've been engaged with OEMs is we've been out with the prototypes,
actually several different prototypes and different generations of prototypes
that demonstrate how the d2p technology works and the value of that for their
products, we're going to walk through some of that here in a minute. Our target
has been on top tier customers. Our initial goal was focus on the top tier,
especially in the cellular and handset space. And depending on the reaction
there, we might expand to include other kinds of OEMs and possibly other market
spaces. As good fortune would have it, we've been so actively engaged with so
many OEMs in the top tier markets that that's where our focus has stopped and
that's where the company is exclusively in dialog today.
I believe you'll see the company over the next two, three, or four quarters
emerge with a business model that's not purely a chip model. I think you'll see
a combination of both chips that companies are interested in purchasing from
us, some of them more shorter term and I think you'll also see licensing
business model come from us as certain OEMs are interested in how do they
include this technology either within other system chips they're developing, or
running on fab that they've run on other than the fab that we're running our
chips on, which happens to be an IBM fab. In the near term you'll see us
continue to increase market awareness about our technology and about our products.
The website you saw us revise and turn on a week or two ago, you'll see a lot
more information going on that website that'll help you to continue to follow
the story of the technology and how it fits into the markets and why OEMs, we
believe, will adopt the technology. And hopefully that site over the course of
the next year will also start popping up with important partners and official
customers in technology that are the top‑tier OEMs that we're speaking to
today.
So today we are going to spend a little bit of time talking to you about what
we call Energy Signal Processing, which is the umbrella over the components
that we're developing. Energy Signal Processing being a technology approach
that takes the old analog circuit architectures that have been around now for
almost 100 years‑‑I mean, they're not vacuum tubes anymore but
they're the same circuits. Marconi would recognize those circuits‑‑and
replace them with what we call Energy Signal Processing, which is a digital
processing implementation of radio signals both on the receiver as well as the
transmitter.
The benefit of ESP technology is you are really liberating people from the
constraints of the old analog architectures. And we'll talk a lot about what it
means to break these old RF rules and the benefits of doing that, because there
are many, many benefits of doing that.
The ESP technology is a result of our engineers and mathematicians really
stepping outside of the design limitations that the analog circuits impose upon
the designers, and really looking at how to process radio signals in a
completely different light. And one of the things that I think is the most
exciting about ESP technology in the longer term is not just what it can do for
products that people are trying to design today for the next generation of 3G
or 3.5G products, but the potential to influence future generations of wireless
standards.
Today those standards are based on certain assumptions of what the radio
transceiver can and cannot do‑‑how many people can I pack into a
spectrum, what kind of adjacent channel rejection can I get, what kind of data
rates are practical. Those are strongly influenced by the performance of radio
transceivers. We will liberate people from a lot of the constraints that today
they have to work around because of these older analog architectures.
So when I try to give a sort of high‑level analogy of what is ESP
technology, what could you liken it to or try to compare it to. If you look at
most of the wireless, well in fact pretty much all of the wireless digital
communication products today, they are based on DSP‑‑Digital Signal
Processors. And of course DSPs are in a lot more than just the wireless
products, they're in audio music products and video products and really
anyplace that you're processing analog audio, video, multimedia signals. DSPs
have really become the standard, the norm, the de facto.
And if you think about why they've become so broadly adopted, think about the
old music players or cassette tapes or eight‑tracks or records. It didn't
make any difference how much money you spent on those analog machines, there
was a certain amount of quality you were going to get out of the audio. It was
going to have a certain amount of distortion, there was going to be a certain
amount of background noise, and it really didn't make any difference how much
money you spent.
DSPs come along and the CD digital music era ushers in. I can still remember
hearing my first CD and going "Wow. That is crisper, clearer, more depth,
less distortion at much less money than any of the old analog processed
audio." So, sure, the audio still comes out of a speaker in an analog
format, an analog signal, but the processing of that signal through DSPs became
digital.
This is the same thing that ESP ‑ energy signal processors ‑ that
we've developed are going to do for wireless communications.
When people have purchased our WiFi product, and they've gone from that 1000
foot outdoor distance to a mile, they say, "Why does it do that?"
"How do you do that?" It wasn't through stronger transmitters, it was
through having ESP. It was through having signal processing that was so much
cleaner and noise floors that were so much lower and data packets that were so
much better behaved that you could get much much closer to the theoretical
perfection of what you'd like to as an engineer in an actual produced product.
So, ESPs have a big role to play in the future. I can tell you that it's
ludicrous for me to believe or think that anybody could embrace the concept
that analog radio transceiver technology that's been used for a hundred years
could possibly be the right foundation going forward to some of the world's
most advanced consumer products ever ‑ wireless digital communications.
It's crazy!
So, if you can turn that analog processing digital, why wouldn't you do it? Not
why would you do it, why wouldn't you do it?
Today we're going to talk about what ESPs do and how some of the benefits can
be measured and seen. We can spend a whole day seminar on this. Hopefully, the
day will come when we have day seminars for our customers on ESP technology.
But today we've only got 45 minutes, so we're going to have to move through
this pretty quickly.
What distinguishes energy signal processors from analog is they have the
ability to process a radio signal optimally to the energy of that radio signal.
Analog cannot process a radio signal in an optimal fashion. There's all kinds
of inherent distortion and limitations to the analog processors.
When you can optimally process a radio waveform, you have the ability now to
eliminate many circuit process that you don't need anymore. Some of them are
redundant, some of them are large, they take up power, they take up size, and
ESPs can eliminate that.
Last but not least, ESPs then become the fundamental umbrella over which our
direct‑to‑RF power building block (d2p) and direct‑to‑data
building block (d2d), receivers and transmitters, can spring forth from.
We're going to spend most of our time today talking about d2p because, frankly,
that's where we're getting the traction with the OEMs.
It's a transmit chain replacement technology. If you look at the way a
transmitter today is built ‑ let's use your cell phone handset as an
example ‑ there are a chain of events that happen from the data domain up
to the RF that goes out of your antenna. A d2p replaces all of that hardware
from the data domain all the way to the radio domain.
So you no longer need transmitters, you no longer need power amplifiers ‑
at least not the analog ones, and you don't need any of the corresponding
filters which are required to clean up some of that analog distortion that
happens.
D2p technologies are not base stand dependent, meaning they can take any data
waveform ‑ any analog or digital signal and convert that to the RF
carrier without being specifically, a priori knowledgeable to what that data
signal looks like. And they can produce a carrier at the RF power output that
you're looking for in a single, unified step. There are no steps between the
data domain and the RF domain at power in a d2p architecture ‑ it's one
unified step from point A to point B.
Man 1: Explain that again.
Jeff Parker: OK, I have to show you a
block diagram ‑ the way it's done traditionally, and then I'll show you
how we convert it. Maybe that will help put some more detail to that.
Why are OEMs excited about d2p? Really it's three reasons, categorically.
Well, there's actually a fourth I should put up there now that I'm thinking
about it. But number one is size. Today they're trying to pack more features
into handsets: mp3 players and video conferencing and cameras and email data
services.... And so, how do I pack all of this and not make my handset balloon?
Number two: performance. Building a radio transmitter in a mobile product is a
difficult, challenging process especially for 3G and beyond that applications.
How do I make those things behave properly over temperature, over voltage, over
power ‑ meaning different powers depending upon where I am to the base
station and over frequency, because I'm on different channels. It's a very
complicated puzzle for analog, their hardware to deal with. And it's a lot of
the limitations analog hardware imposes. It causes performance issues in mobile
products.
Efficiency: How do I get a stable transmitter over all those operating
conditions that doesn't use a lot of power? Because frankly, I've got a little
battery. I don't have a lot of power.
Multimode capability: How do I get all the different things that I want to
offer to my customer as an OEM in terms of different networks, different
geographical solutions.... It's CDMA here, it's Y‑band CDMA there, it's
EDGE someplace else, it's GSM, ultimately with WiFi.... How do I get all that
packed into this little product called a handset?
And last but not least, my fourth bullet: How do I get it out at a cost that I
can make money?
And these are the four reasons why OEMs are excited about d2p. We address all
four of those in no small way.
If you look at a traditional handset ‑ and this is kind of hard for you
guys to see from where you're sitting. But what you see here: this blue box on
the top is a kind of floor plan of a block diagram of a handset. And here I've
got my base band processor ‑ my BSP that's putting data. And ultimately
I'm trying to get that data from here to go out my antenna.
And what I have to do today is, I have to take that data into a traditional
radio frequency transmitter where I go from the data domain to the RF domain.
But the transmitted signal is small. It's not enough to go any place. I'm going
to have to boost that signal. But even with the little, tiny signal that I just
generated for this transmitter there's distortions because of the analog
process.
So you know what I have to do? I have to put a filter here. And if I've got a
multi‑band transmitter for different frequencies I have multiple filters.
And then I take that filter output into power amplifiers that are analog power
amplifiers. And I amplify the signal to be large enough in signal power to get
to the receiver I'm trying to talk to.
But I've got to go through another filter because, you know what? The signal's
been distorted again. And so, it's kind of a one step forward, half a step
back, one step forward, half a step back process and that's what [xx] power and
size and cost.
The chain here that we call the transmit chain on a single band 3G handset....
You know, this looks fairly simple. It looks like it's only four building blocks.
Well, those four building blocks translate probably to 50, 60, 70 components.
If it's a multi band you could be up to a couple hundred components. So it's
not a particularly elegant affair and it's what people deal with today in the
analog world.
We translate that entire chain into a single IC where you take the data out of
the base stand into our chip. It translates that into single steps from the
data domain to the radio domain, eliminates all those filters that would have
normally been here and then comes out in a radio carrier that's at the power
level. And yes, we still can go through a filter, though, we don't have a
perfect transmit signal. But it's the same filter they're using today, so we
don't impose any more requirements on them than they currently have, and then
up to the antenna.
This can be a single chip. It can be a single mode, single band. It can be a
single mode, multi band. It can be a multi mode, multi band. Typically you'll
see for a chip like a d2p you'll have about 10 to 20 little, supporting
components depending upon how many bands and how many modes of operation you're
trying to deliver.
So that's now a high level block diagram overview of the difference between a
traditional approach and a d2p.
This is my favorite part of the discussion. We could get lost here for the next
few hours but my CFO won't let me ‑ she'll throw something at me because
she says we have a lot more to cover than just these [xx] steps. But frankly
this is why OEMs want to do business with Parker Ridge and where we are right
now.
When you walk in to an OEM with a new technology and you're one of hundreds
that's trying to get their business what you have to quickly do to distinguish
yourself is help them understand that not only do you understand their problems,
but you have a solution for their problems that doesn't take something with one
hand and give it back with another. What we're going to show you here in five
or ten minutes is just a taste of how the d2p balances competing objectives
that analog transmit chains are not able to balance like a d2p digital
architecture can.
So what I'm showing you here is this: We take a d2p demo board, which is a
little circuit board that has our d2p technology on it, and we go into an OEM
and we hook it up to some Agilent test gear which is pretty much the Cadillac
standard, the gold standard in the RF industry. All the OEMs we do business
with either have Agilent test gear or the competing gear from a company called
Rohde and Schwarz, which is maybe 20% of the market out there.
What that gear allows you to do is to take a radio signal on the transmitter
and plug into that equipment and it will do an automatic assessment for you of
all of the values of what you're generating. And it will quickly help an OEM
determine: Have you really created a transmitted waveform that meets the
standard, that meets the government regulations, that really does what is
necessary to make it into a product? And so you've got to pass this litmus
test. I mean, there's even tests beyond this, but this is a good leap to
getting an OEMs attention.
What I'm showing you here is this is a GSM transmitted signal that we put into
this piece of Agilent test gear. And this is the screen of the test gear on
this screen capture. What you see here is a circle ‑ well, it looks like
an ellipse. If this TV were perfect it would look like a circle. And the way
this modulated analyzed signal works is the Agilent test gear takes these four
areas, which are the vectors of the GSM data and draws a line between them. And
if it's a perfect, ideal radio signal you would have a perfect circle.
So a zero degree error in the phase of that circle would be... or in the phase
would result in an ideal GSM modulation. The GSM standard says you can have up
to five degrees of error which is the distortion. But the peak error can be up
to 20 degrees. And this error will relate to data packets that are being lost.
It will relate to other non‑ideal conditions that as an OEM you want to
avoid. The typical transmitter is going to be probably in the three, four
degree range here, and, you know, 10, 12, 15 degree range here. We are at point
one eight degree of error.
And this is an error that, you don't see this kind of quality even from base
station equipment which is a card, not chips that are plugged into the wall
with unlimited amounts of power.
The peak error, which can be up to 20 degrees, is less than half a degree. So,
when they see this‑‑and I've actually been in meetings with very
experienced RF engineers who will put their nose right up on the screen, and
they'll say, "I've never seen anything like that." And that's the
first thing that we'll typically show someone.
Now, you'll see this is a 1.9 gigahertz frequency output from this chip. And
it's running at full power: two watts for this particular implementation. And
the reason those are important is that's the PCS band, which is a harder thing
to generate than the cellular band. And this 33 dBm is full power. If you
wanted to fudge this, you'd go down in power. You'd say, "Look at this
perfect circle!" And then, as you run up in power, you'd see the circle
distort. So, we're starting at full power.
The next thing we do is we take the d2p and we change the frequency from 1.9
gigahertz to 1.8 gigahertz. We've moved 100 megahertz over. That's many
channels. And guess what happens to the phase error. The phase error goes from
0.18 to 0.14. And the peak goes from 0.48 to [xx]... It got a little better.
And they would look at that and they would say, "Wow. That's pretty impressive,
over 100 megahertz of frequency spectrum." So we say, "Well, let's go
the other direction." And we go just under two gigahertz.
So we've got a 200 megahertz wide spectrum, and your phase error hasn't
changed. And it's the best they've ever seen. That's covering, now, the Korean
PCS band, the Japanese PCS band, the American PCS band, the European‑‑that
doesn't miss any of the PCS bands. And there's another 3G band at 2.1 gig. But
to cover 200 megahertz of spectrum, that's an impressive result.
Next thing we do is we put up this test. This is a wideband CDMA test. And
we're showing here these are the data vectors that have been analyzed. And what
you're seeing here in these yellow lines is the transmitted signal is drawing
these data vectors, and this machine is putting that up. And what you want your
transmitter to do is to have the smallest possible dot you can, because that
gives you the highest data throughput you can get. If you overshoot and
undershoot those constellation points, you begin to degrade the signal.
The way that's measured is two ways. One measurement's called rho.
Our rho is 0.99945. And I don't even know if the test here can test beyond
that. But you don't see this in the market. You don't see this any place. You
don't even see this in base stations, OK?
The EVM, the error vector magnitude, that says: how accurate are these vector
points being represented? How much overshoot and undershoot does your
transmitter have? You can have up to 17.5 percent. We have 2.3 percent. These
are spectacular numbers.
Now, this is run at full power. But let me show you what's very interesting
about this. This is 1.8 gigahertz in the PCS spectrum. The voltage power in
this chip is only 2.4 volts. And let me tell you the significance of that. The
significance of that is your battery in your handset, at full charge, is going
to be at 4.2 volts. When it drops, it's going to go down to 2.7. Below 2.7,
your handset's not going to work, so you would normally just test this down to
2.7.
But the reason we show this at 2.4 is this shows you that you have margin in
your components here, so it'll handle over temperature and manufacturing
tolerances, and that you're better than what they require from the batteries
they're currently using.
So we then go from 2.4 volts up to 4.2. And you can say, "Well, maybe when
you go into the other side of the voltage that you're going to distort the
signal." And what you'd see is you go from a 0.999 rho to 0.998. And then
we take this signal and we move it from 1.8 gig to almost two gig, back down to
2.4 volts, and you're still at 0.9999 rho and 2.59 percent on the error vector.
What this is showing you is this is not a technology that you're trying to
balance BBs on a razor blade, as our CTO likes to say. You have a lot of robust
forgiveness in this. The same thing you'd expect to get out of the digital
signal processor that you're getting at base band signals, you're getting out
of an energy signal processor on the D2P component here.
This is the EDGE waveform. Remember, we saw the four points of the GSM. So you
take the GSM signal and you'd make more data points out of it, and you get the
EDGE, which is the 2.75G standard. This is a very challenging waveform to
create. This is kind of: how do you pack 10 pounds into an 8‑pound bag?
This is how the GSM people said, "We can upgrade our networks less
expensively before we get to 3G, by layering some things on top of the GSM
network."
Well, that's true, but they've put a tremendous burden on the mobile products,
because to make this data constellation is not trivial. And you can see this in
the specs. You can have up to a nine percent EVM and 30 percent peak. Well, we do
this at less than 1.5 percent on the average EVM and less than six percent on
the peak. So here, you're sitting at full power, at 3.6 volts, at 1.9
gigahertz, and you've got these wonderful performance metrics.
And now I want to show you something a little different. One of the things an
OEM wants to see is, "OK, you've shown me that your signal is pristine;
it's really nice. Well, what about the spectrum around the signal that I'm not
wanting?" In other words, you have a radio carrier, and if you have a
perfect radio carrier, you'd have this nice rectangular‑looking piece of
spectrum that you're using, and around that, you wouldn't be using any spectrum
or you wouldn't be creating any spectrum.
But any non‑linearities that you have will grow radio, will grow
frequency transmitted that you don't want. And we grow it. We don't have a
perfectly linear process, so we have some regrowth, just like the analog does.
But what these numbers show you here is the equipment here measures different
distances off the center of the frequency you're generating. And if you had
zeros up and down here, that would say that you passed the standard. You're not
regrowing unwanted spectrum beyond what we allow you to as a standard, but you
have no margin for manufacturing. You have no margin over temperature. You have
no margin over other operating conditions.
So, what you see here is we have anywhere from 6.8 DDC of margin, minimum, all
the way up to ‑21. The bigger the negative number here, the more margin
you have. Many transmitters you would see would have margins of one, two, or
maybe they'd have numbers that look like this at one operating condition, but
they'd be just skirting the edge on the edge of the frequency band, or low
power conditions, or some other operating condition.
So, this would tend to be its worst at high voltage because it would tend to
want to re‑grow in nonlinearity more quickly and you'll see excellent
performance there. So, if I was a skeptic I'd say, "Well, maybe your
technology regrows at the low voltage."
So, we go down to the low voltage. We go down to the low voltage, this one, and
you'll see we still have excellent results. And then we do one more thing, this
is a full power output out of the transmit, out of the V/P. We then drop the
power down, from of a watt down to less than 0 dBm.
And what you'll see here is our numbers aren't as good. And this one here you
know just skirts by. But that's under an extraordinary condition where you're
3/10 of a volt below what the carrier [sp] can even deliver. So, what this is
showing you is that this technology is extraordinary in its quality: its
ability to generate the kind of quality waveform that you would expect out of
our [sp] transmitter.
This is my last chart of our technology. This is an interesting one. This is
showing you‑‑so here's the frequency spectrum and here's power. And
this is actually a CDMA 2000 transmitter waveform. Remember I told you before,
if you could take a perfect signal, it would be like a rectangle. It'd be like
flat here, straight down here, and straight down here.
This blue line is a traditional CDMA transmitter. And you can see it can't get
that nice, crisp, rectangular look that you want. So, it's just skirting by
here on those specs that are measured at the various points along the way.
Well, the green line is V/P. While it's not perfect, it is so much closer to
theoretically what you want to generate. And it's not even a close contest. So,
the difference between here and here is what V/P does, which is an enormous
benefit. How? More users can get on the spectrum. I can now put people closer.
I mean we'll get to that in the presentation. There are a lot of reasons why
you want to do that.
CDMA is probably the most challenging transmit waveform to generate because it
requires a lot of linearity which is why you see people having problems in
here. So, I show you this one because out of all of them, this would be the
most difficult one for us to make look like that.
Now, I'm just going to spend a minute on the traditional technology: so what
people use today. Well, this is the chart right off of the datasheet of a very
high quality wideband CDMA power amp. I won't tell you which company, but it's
a very well known company and again a very high quality component. And this is
a very great component for an analog component.
What you're going to see here is the challenge people have in trying to use
these in a handset. So, this chart here shows you here's the power out of the
PA. I can make the power go up. And what these numbers are showing you on this
axis is that regrowth area that we don't want. Right? We want this to be as big
a negative number as possible. This is the area that's outside of the channel
I'm interested in.
Well for wideband CDMA, for this particular measurement you want to be ‑33
or lower. So you can see, "Oh this is very high speed behavior here Jeff.
What's the big deal man? The form looks pretty good."
Well, let me show you where the problem comes. You'll see it runs really nice
and flat until you get up here to about 25 or 26 dBm and then all of a sudden
it goes catastrophically bad.
What that is is these analog power amplifiers all have a compression, what's
called a P1 compression point. It uses that. You push the power out of the
amplifier as far as you can push it before it goes nonlinear. Once it goes
nonlinear, that's what you're going to get. You're going to get a nonlinear
result.
So, here's the rub. Why do people push them close to this point? Why do you
care about this area? Well, let me tell you why we care about this area. That
same component, these are the efficiency curves.
So, here's what it shows you. When you're at about 27 or 28 dBm of power out
which is the full power of a wideband CDMA transmitter, you're running at about
40, 45% efficiency. That's pretty good.
But guess what? If I'm pushing power out at that point, look at how close I am
to going nonlinear: very close. So, what do I do? Nobody can run them here. You
run them back here on the flat part of the curve because you've got to assume
over temperature, voltage changes, other operating conditions, I can't go here.
So, where they really run these is right about here: 25dbm. Well, this isn't 25
dBm. What's the efficiency at 25? Well, it's no more than 45 or 50%. At 25 it's
just under 30% efficiency.
So when you look at a PA, and one of the important reasons to show you this is
you're going to go away from this presentation inevitably a week from now or
less you're going to look at some announcement from some PA manufacturer: 48%
efficiency A, 50% efficiency A.
Well, that's all true, right here. You can't use it there. And nobody does. So,
that's not the story. The story is where do you apply it. And that's not the
only story, because the PA is only a part of the power consumption. Right?
There's also a transmitter that you use to feed the signal in to that. Well,
how much does that draw? So, ultimately it's a bigger job to understand the
efficiency of the transmitter than just to look at a single spec sheet on their
top headline.
We're not going to share with you today our curves because frankly they're very
good, but they're also something I'm going to have to hand you over to our
competition right now: let them work for them.
But, I will give you a couple of points on our curves which are very important.
On CDMA, at full power out, it's over 50% efficient. So, what's exciting of
that to the OEMs about that is our 50% is a real 50%. We have no P1 compression
point because we don't put RF into a component and have RF come out of a
component.
There is no compression. It's data in our temps [sp]. So, when we set it for
full power, that's what you get. They can count on that. The reason we give you
the 17 dBm output power is that is the power that most handsets for CDMA and
wideband CDMA operate at most of the time. And in our situation we're at 30%
efficient.
That compares, by the way, with the analog transmit chain at full power they're
going to be in the low 30 percentile, high 20 percentile efficiency. And at the
17 dBm, you're going to see eight to 12%, something like that, efficiency
range.
Here on wideband CDMA range, we're at 44% efficient full power. You're going to
see analog chains in the mid 20s. And at 17 dBm, we're at 22% efficient versus
the analog chains are going to be in the 6%, 8%, 5%, not very efficient.
Audience member: And efficient is usage of
power?
Jeff Parker: Yes, it's power in versus RF
out. So, an the example if I had half a watt going out of my transmitter, and I
was using a watt, that'd be 50% efficient.
So, a couple of things I want to point out. All these charts and graphs that
you just saw on the d2p, those are all from the same d2p. In other words, I
didn't‑‑we don't go to an OEM and say, "We're going to show
you this d2p for wideband CDMA. We're going to show you that d2p for CDMA.
We're going to show you another one for GSM, a fourth for EDGE."
Those all went through one d2p. So, that's the multimode capability. Now, can
you make a more elegant, less expensive d2p that's just single‑band,
single‑mode? Of course you can. So there will be a range, there will be a
family. And there will be single‑mode d2ps, multimode d2ps, single‑band
d2ps, multi‑band d2ps. But I want to point out from our charts we showed
you today, that's from a single multimode d2p.
Our receiver technology is also very exciting. I think just the OEMs get
excited about adopting that after the d2p, but many of you actually use our
WiFi products, you've experienced that for yourself.
One of the things that I think some of the OEMs are beginning to get a hint and
a clue and a good idea about when they test our WiFi products is that the d2d
is going to have the impact of helping people create networks with fewer
dropped calls, better handoff, and better coverage.
Many of you who have experienced our WiFi products, you'll walk out with client
product to the edge of reception and you'll notice that when you come back
toward the access point that it'll reconnect within a couple of steps, versus a
traditional analog radio that you may have to go back halfway to the router.
What does that translate into in the cellular world? It translates into much
better handoffs and much better connectivity. So we think the d2d will come,
but it's going to come after the d2p.
This is just a chart that shows our WiFi performance and what a lot of people
have experienced against the analog radio front end.
Patent protection for our company is obviously very important. It's not our job
to do free research and development for the world. So we spent a lot of time
and effort and money in protecting our patents.
We currently have 22 issued in the
No question that this is why people are excited about our targets, the market
being 3G, because of what we can do for them for battery life and performance.
The size of the market is no big secret. There's going to be over a billion
handsets shipped worldwide starting in the next couple years; it's already
about three‑quarters of a billion today. The value of the semiconductors
that get shipped into that space is almost topping five billion a year if you
add up the power amplifiers, transmitters, and receivers for both the cellular
and any of the WiFi that goes into a handset. Obviously WiFi outside of that is
even bigger, but this chart is just dealing with cellular and WiFi transceivers
and PAs that go into handsets.
The market trends are in our favor. This is the time for ParkerVision
Technologies to really dovetail purposefully into what people are looking for.
They're looking at taking simple handsets and turning them into all sorts of
interesting user devices.
Nokia just made an announcement, I think it was last week, that they predicted
over a hundred million smart phones will ship next year in the industry. That's
amazing if you think about it‑‑a hundred million smart phones. So
those smart phones are just a perfect example of the trend in this whole
industry and why the d2p architecture is so important for that trend.
The challenges of more power consumption because of new features‑‑how
do I get more things on a crowded circuit board, how do I balance size and
battery life, how do I get broadband connectivity in both directions. Today,
broadband cellular is only broadband cellular from the base station to the
handset. It's still fairly narrow‑band back from the base station. We
help people with that significantly. And our advantage really translates in all
those metrics, in not only 3G, but also 3.5G and 4G, which are already starting
to come off the drawing boards now. So our story resonates nicely with OEMs for
all the reasons we talked about earlier.
Consumers, same reason. Today we are very focused on OEM dialogs, although as
we move through OEM dialogs we also begin to have [xx] and manufacturer
dialogues that we're interfacing our components to. A little later, not in the
too distant future, you'll begin to see, I believe, the company engaging
service providers as well in dialogue. The technology is advanced now to the
point where they can really get excited, we believe, hearing what we can do to
help their network and helping influence that back to the OEM, who is trying to
determine what their next platform technology is.
If you take away anything from today's conversation, what I hope you would take
away is ParkerVision is very quickly approaching an inflection point. It's
approaching an inflection point because we have a solution to a problem that
everybody knows about. We're not explaining to these OEMs, "Oh, do you
know you have a problem in battery consumption, or battery life in 3G handsets?
Oh, do you know you have a problem in putting both of those products together?"
They know they have these problems, and so our opportunity really for an
inflection point is now, which is why I thought it was so exciting to be able
to show you those charts of what our component does today in delivering that
performance over all those operating conditions. That's what bites most RF
companies. When they walk in with a good idea, and the OEM really starts
throwing down, they'll find places that it falls apart. They haven't found any
place where our technology falls apart.
Legacy technologies: Guys, they have reached their point of diminishing
returns. I will be very surprised if five years from now any new designs start
that are not digital transceivers. So while you may have an audience out there
yipping and yapping today about, "Oh, analog is the way it's always been
done," it's ludicrous, for all the reasons we talked about and more that
we don't have time to talk about.
The ESP technology will fill those critical gaps and help people with their
goal. Breaking the rules of traditional RF, I can tell you OEMs have never been
more open to hearing about how you're going to help them break those rules. I
think that's our message: we feel we finally have come to our inflection point,
and those are the reasons why.
Business Models, very quickly: The S‑chip, I think you'll also see the
company have OEMs interested in licensing agreements. Some OEMs will see value
in how they can partition the system, in breaking the digital transceiver
technology differently than just on the periphery of plugging into chips as
they exist today. So I think you'll see Parker Vision as both yet the fab and
semi company, but also as a technology provider through IT licensing.
Caller: These agreements with these
OEMs, how is it that our R&D is finally getting exposed?
Jeff Parker: Say that again?
Caller: These fiber agreements with
these OEMs.
Jeff Parker: Right now we're not engaging
any OEM and asking them to help us do research, because then you get into the
unenviable place of saying, "Well, how do you deal with co‑invention?
How do you deal with IP contamination?"
So right now when we think of IP licensing, it's, "Hey, Mr. OEM, we've got
a core we've developed, and we can port that to whatever processor you want.
We'd be happy to license you that core, which you can add to semiconductors
with other system functions or you can run on the fab of your choice." But
that's more along the lines of what we are working towards, not, "Help us
evolve our technology." At least that's not today's current discussions.
I've taken up the entire time including your Q&A. I'm sorry. I tried to go
as quickly as possible, but we do have a couple minutes. If you guys have any
questions, please feel free now, because I would be happy to try to answer them
if I can. [cuts off]