Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Intro (00:01):
This is a Technikon podcast.
Peter Balint (host) (00:07):
5G... it's become a buzzword, a marketing term, maybe even
a sales gimmick. But behind all the glitz and glamour
of the promised tremendous data speeds and ultra low latency
are the people making it happen. And here in Europe,
(00:30):
it's H2020 funded projects that are enabling this research. I'm
Peter Balint from Technikon, and today we look at one
such project called SERENA. SERENA Partners are exploring the use
of high frequencies for data delivery... and the higher the frequency,
(00:52):
the shorter the radio waves are able to travel, so
they must travel smarter. And this is where beam forming
and new types of antennas are coming in. Using gallium
nitride and silicone as a means to develop high-performance millimeter
wave active antenna systems at a commercially viable price point
while remaining energy efficient is at the heart of SERENA.
(01:14):
Today we talk with SERENA Partners, Giuseppe Caire and Thomas Kuehne from the Technische Universität Berlin.
Let's have a listen. Welcome, gentlemen, and thanks for coming
on today.
Thomas Kuehne (01:30):
Yes, thanks for the invitation.
Giuseppe Caire (01:32):
Yeah, thank you.
Peter Balint (host) (01:34):
In the intro, we heard a bit about the big
picture in SERENA, but what is your particular role in
the project? And we'll start with you, Giuseppe.
Giuseppe Caire (01:42):
Yeah. So basically, SERENA is a large project with many partners.
We cover the signal processing part, and the signal processing
part consists of basically what comes after the antennas and
the A-to-D converters, and take the signals and basically make sense
(02:05):
out of them.
Peter Balint (host) (02:07):
And to clarify for our listeners. When you say A-to-D converter,
that means simply analog to digital. Thomas, can you add
anything here?
Thomas Kuehne (02:16):
Yeah. So the the part of signal processing is not
only the strict designing the signal processing, but because of
we are so many partners and also from different fields
of expertise, it's also our task to basically help designing
the overall system and say what requirements the signal processing
(02:41):
has on the other components. And also what is the
limitation of the signal processing so that the others know
what they can expect. So it's also on the cooperation
collaboration side that basically we discuss it together in the project,
and it's our task to basically present the signal processing
(03:04):
and the communication system.
Peter Balint (host) (03:06):
So your team is basically managing expectations when it comes
to signal processing, and this is important because nothing happens
without the signal getting through regardless of hardware or anything else.
Giuseppe Caire (03:18):
Yeah, actually, the signal processing involves going from the digital
domain to the analog domain, and therefore the transmission side
is generating waveforms that will be then amplified filtered and
sent through the antennas. And on the other side, on
the receiver side is collecting the signals from the antennas,
(03:43):
filters and eventually A-to-D converters. And then somehow transferring them
into a digital machine that does the numerical processing of
the signals and therefore analyzes, you know, decodes the signals if
this is a digital communication. But anyway, does the measurement
(04:08):
the representation of the signals that comes from the hardware, so without
that the hardware is... difficult to understand what the hardware
does if you don't see the signals that you eventually
you can actually process them and make sense of them?
Thomas Kuehne (04:25):
So maybe in general, for more a customer viewpoint, what
the signal processing does is it converts the bits from
a video, for example, and converts them to this analog signal,
which is then used by the hardware to to be transmitted.
And this conversion process consists in general of many, many steps,
(04:49):
and some are related to communication systems and we are
basically covering these communication steps. And the better we do that,
the better the communication is. So the more reliable and
the higher the data rate is.
Peter Balint (host) (05:03):
Yeah, reliability and speed from the consumer side, these are
the top issues, and working at higher frequencies gives us
increased bandwidth, thus faster communication speeds. But tell us Giuseppe
about the challenges or downsides of working with these higher frequencies.
Giuseppe Caire (05:20):
Yeah. So it is very challenging because there is something
called the propagation loss that depends on the wavelength. So
the smaller, the wavelength and therefore the higher the frequencies,
smaller the wavelength, the larger is this propagation loss, and
this propagation loss has to be compensated through antenna gain.
(05:45):
So antenna gains means that one has to design, in
this case... in the case of SERENA, smart antenna arrays
that create something called beamforming gain. So the antennas become
very directive. And in order to achieve these directivity, basically
(06:06):
what there is the all the digital and analog processing that,
that that creates that. So it's not only a problem
of amplifying the signal, but is also creating these special
antenna radiation patterns that can be steered towards the target.
And of course, these presents a number of problems... the
(06:27):
technology problem is to design efficient hardware and the algorithms
to to do this beam steering. And also there are
problems in terms of network design because one has to
design particular transmission patterns that allow the base stations and
(06:51):
the users to estimate directions and point beams at each other,
which is a fairly difficult process, especially in the mobility
conditions where beams have to follow the target. In addition
high frequencies suffer from blocking problems so they don't go
to walls that they don't go around buildings like lower
(07:14):
frequencies do. So it means that that you need probably
a more distributed, more dense deployment of access points in
order to have... to guarantee coverage. So there is a
number of challenges at the system level, at the technology
level and at the algorithmic levels that make the exploitation
(07:36):
of very high frequencies is still quite quite at the beginning.
And the direction of 5G and then beyond 5G is
to go higher and higher and so to address these issues.
Thomas Kuehne (07:49):
Yeah, and I think one of the major challenges or
the downsides is that this technology is very new, although
what Giuseppe said is true, and that's why you need those
new technologies. But one has to understand that 5G is
the first big standard to actually introduce these high frequency
into real products. It has been there before, but just
(08:11):
a couple of years ago and was more Wi-Fi systems.
But it hasn't been used. It never really took off.
So where the lower frequencies have been used for 20,
30 years and with radar even longer in commercial systems,
this is the first time where those technologies are used
in commercial systems. And that's what is challenging in actually
(08:35):
using those. So downside is that the experience is much lower,
is much less experienced and with far higher frequencies.
Peter Balint (host) (08:45):
That's a good point. Thomas, I would imagine that without
decades of performance data or real world experience, working in
these high frequencies could be quite challenging. What do you
think SERENA can contribute to the 5G world of tomorrow?
Thomas Kuehne (09:01):
It's actually not maybe a single result which SERENA will
will have, but SERENA is a project with many partners,
and it's actually bringing together different technologies to basically solve
this common goal of having an efficient, high frequency system.
(09:22):
And there's also the interesting part of SERENA is that
there is this cooperation between the many partners and for example,
we have some partners who focus on manufacturing and some focus,
like we are, on the signal processing. And basically all
our results are a bit hidden from the customer because
(09:43):
what we will enable with this cooperation and basically the
that we design a system which has all of these
different fields in mind is that we have a more
efficient system. So the power consumption might be better and
also have a... like a higher integrated system so that
it's actually feasible to to be implemented so that 5G
(10:07):
can have many, many of those millimeter wave or high
frequency base stations. And not only a few because if
they're costly and if they use a high power, maybe
there are few. But if they are really up to
date and they don't use too much power and they
are cheap enough so that they can be sold with a profit,
(10:29):
then they're actually deployed by the providers.
Peter Balint (host) (10:33):
Meaning these marketing aspects must be considered even in the
R&D phases, Giuseppe, do you have anything to add?
Giuseppe Caire (10:40):
Well I generally agree that the SERENA platform is a high
output power steerable antenna array at millimeter waves that is
highly integrated, and therefore it's like a step forward in
bringing this technology to become mainstream in terms of cost
(11:02):
and usability. Then we have to see how the especially
the industry partners seen in SERENA will decide to do,
and this will evolve into real products. That's maybe a
bit early to say, but certainly that's the direction.
Peter Balint (host) (11:21):
That would certainly be a great result coming from SERENA. Giuseppe,
sometimes in the field of communication we hear of beyond 5G.
Is it clear what beyond 5G will actually be?
Giuseppe Caire (11:34):
Yeah, so it's funny because 5G is barely being deployed
and the components of 5G, which is deployed is mainly
the core network and some software features and some radio
(11:58):
access interfaces at the relatively low frequencies. There is not yet,
as far as I know apart of say, beyond lab experiments,
there is not yet a widespread deployment of millimeter waves,
at least for the access. So for for the access
to the users, people use millimeter waves for relaying macro
(12:23):
base station to a smaller base, stations to do like
wireless front haul and things like this, but for access, for
coverage is still pretty much not yet there. Nevertheless, industry
and research are really, you know, thinking about beyond 5G.
Some people there are 6G centers, 6G research hubs popping
(12:47):
out in different European countries. So, I think that the definition
of beyond 5G is still quite vague. For some people,
this means going even to higher frequencies than millimeter wave.
Other people talk about the different type of features that
(13:10):
the ways of going beyond cellular. People talk about cell free,
user centric networks that go beyond the concept of cellular.
There is this concept of open RAN architectures with interoperability
between different vendors so that the telecom operators are not
(13:32):
somehow committed uniquely to a single vendor that provides the
whole set of network components all together. So there are
many different ideas that are in the discussion. So I
think that the contribution of SERENA in this direction will
be somehow demonstration and feasibility of these steerable antenna technology, which,
(13:57):
although it has been around for a while, is not
particularly new in terms of the concepts. But it has
been relegated mainly in military applications or phased arrays in
military applications that are very expensive objects or on research papers.
Thomas Kuehne (14:19):
Maybe I can also add the larger part of SERENA is
investigating frequencies, which will be used in 5G and the 5G,
which might be deployed in a year or two. But
there's also a smaller part of SERENA, which actually focuses
on even higher frequencies. So exactly what Giuseppe said that one
(14:39):
technology or one idea for the beyond 5G world is
is that even using again higher frequencies because of more
available bandwidth. And SERENA even covers this. I mean, it's
not as integrated and not as complete as the 5G part,
but there is already a part of SRENA which investigated
those higher frequencies. And of course, also the signal processing
(15:03):
and the design of the communication system we have proposed
with SERENA might become more and more relevant for higher
frequencies because it becomes more and more different compared to
lower frequency systems. This is also something which can be
used for higher frequencies or which has to be used
(15:25):
more and more for higher frequencies.
Peter Balint (host) (15:27):
And that's most encouraging. SERENA has been running for almost
three years now. What kinds of challenges have you experienced
in that timeframe?
Thomas Kuehne (15:37):
So so what I have mentioned before what the goal
of SERENA is to integrate many partners in many fields
of research. But of course, this can also be challenging
because collaboration is something which needs to be done. And
then you have experts from different fields, so there's often
a bit difference in seeing the problems from different viewpoints
(16:01):
and this collaboration. It was actually a very good collaboration
within the project, but still the partners have dependencies, and
those dependencies are a risk when you see that there
are some delays introduced because maybe something didn't work out
as expected in the beginning. So delays were somehow a
(16:24):
problem for SERENA because it's still research so things can not work.
And then they might need to be redone with something
you learned in the first process where it didn't work.
And then the second version works. But then maybe you
have a partner who is also delayed because he also
needs to wait for the second version. So the challenges
(16:46):
have been exactly in this collaboration that partners depend on
each other and this means that delays propagate and things
cannot be tested before a certain goal has been achieved
and things are tested from a different partner. So that
was basically the main challenge, I would say. But it was
(17:08):
also this collaboration is, I think, at the heart of
SERENA so it's very important, but it was also a
challenge and a risk. But sometimes you... especially in research,
you need to take risks otherwise you cannot go beyond
the state of the art.
Giuseppe Caire (17:24):
Yeah, I think... I don't have much to add to what
Thomas said is that in these multi partner projects the... it's
always challenging to make all the partners work together in
a way that at the end, the integration can actually
(17:48):
be done without too many problems. And also there are
several different technologies here in there that participate... So, yeah,
there were some challenges in in in putting all these
things together.
Peter Balint (host) (18:08):
That's how it is in the research realm. So on
the SERENA website, I see that there is a demonstrator
to be built and seeing as how we're on the
tail end of the project here. Can you give us
a brief update on that endeavor?
Thomas Kuehne (18:23):
So actually, we can see the demonstrator is built from
many blocks. As I said, we have many different partners.
And the main achievements of the project... so this manufacturing
integration and the communication system... those work. So we have
a demonstrator for this very small integrated module where you
(18:46):
have the antennas and the power amplifiers and beamforming all integrated,
and we can also control this module. So this is
the basic block of what our how the proof of
concept system, as we have called it, has envisioned and
this this module works and we have also used it
(19:07):
to to measure communication signals and and to see if
the signal processing works. What is a bit delayed because
of the what we have, the challenges, what we have
talked before is that we use this block and we
basically scale it up to have a nice demonstrator, which
is already into the direction of a product, which is,
(19:30):
of course, not the product, because we have product you
need to support and things like this, but we want
the goal was to have a very a demonstrator which
could actually demonstrate the whole system on a large scale.
And this is unfortunately a bit delayed, but all the
working parts they work and we measure them and we
(19:53):
we got the results. So let's say for the research,
the important parts, they are there and we have demonstrated them,
whereas the final demonstrator to show maybe to the public,
which is easier to understand, is still delayed, but we
are working on it.
Peter Balint (host) (20:14):
Well, this sounds good, and progress was made, and the
results of SERENA could keep Europe on the cutting edge
of next generation communication. And this is, of course, undoubtedly
due in part to the efforts of both of you.
Thanks for all the information about SERENA.
Thomas Kuehne (20:31):
Yes, thank you. Thanks.
Giuseppe Caire (20:33):
OK, thank you very much. It was a pleasure.
Outro (20:39):
This podcast was brought to you by Technikon. The SERENA
Project has received funding from the European Union's Horizon 2020
Research and innovation programme under grant agreement number 779305.