JANUARY 2017EDITION

Preface January 2017 The highlights of 2016

Vision: Technologies to extendsemiconductor scaling

Vision: Optimizing technology for IoTsystems – adding fingerprints andbrains

Intro: Enabling the intuitive Internet ofThings

Vision: Interplay between chip anddigital technologies crucial for the IoT

Vision: DNA sequencing becomingroutine

Vision: A new era in mobility

Imec MagazineEdition | January 2017

1/57

Vision: A living lab for smart citiesworldwide

Vision: The Fourth IndustrialRevolution is underway

Vision: Renewable energy and ICT for asustainable energy system

Vision: Imec digs into the goldmine ofFlemish high-tech

Vision: Helping companies – small andbig – translate innovative ideas into

market-ready solutions

2/57

Making the impossible possible – together

“Fantastic growth figures, significant technological breakthroughs in numerous areas,the merger with digital research center iMinds: 2016 was an incredible year for imec.Still, more than ever, we are looking to the future; a future driven by radicalinnovation; a future where, together with our customers and partners, we strive tomake the impossible possible…” – Luc Van den hove, General Director and CEO ofimec.

2016 – an incredible year for imec

When I look back at 2016, I can draw only one conclusion: 2016 was an incredible year for imec. Tobegin with, we were once again able to present fantastic growth figures, and that applies to all thesegments in which we are active. What’s more, these growth figures resulted in an increase ofroughly 35 million euros in turnover (an increase of 7 to 8 per cent per annum). That is huge – andmoreover it is a trend we have been able to record for several years running.Secondly, we have initiated and successfully completed countless innovation projects with you –our loyal partners and customers – thereby achieving various technological breakthroughs in a vastrange of fields. You will find a short overview of these further on in this annual report. These, too,are achievements we can be proud of!

Edition | January 2017

General

Preface January 2017Each month our CEO reflects on the events in his (professional)life and discusses some of the articles featured in this month’smagazine issue.

3/57

Another significant achievement in 2016 was the opening of the new cleanroom. Thanks to thisultramodern lab covering more than 4,000 m² (as a result of which we expand our cleanroominfrastructure to more than 12,000m²) we can continue to conduct research into new technologiesthat make microchips even smaller, more powerful and more energy efficient, using the mostadvanced chip production equipment.

And finally, of course, there was the merger with the Flemish digital research center iMinds. It was amerger dictated by opportunities; moreover, it was a merger that has been completed at a speednobody would have believed possible at the beginning of the year. It is a marriage between chip andsystems technology on the one hand, and digital technology on the other – equipping us to createeven greater local and international impact.

That is also the message we aim to convey through the double interviews in this edition of imecMagazine. In each article, a hardware expert and a software expert discuss the most importantachievements in their domain in the past year. At the same time, they look at what the future holdsfor us.

Our mission: to make the impossible possible – together

Of course, it is always nice to look back on a successful year. But at imec, what really counts is thefuture. And there as well, the merger with iMinds will generate significant value.

After all, our rapidly-evolving society is in great need of disruptive solutions in areas such as smartcities, health(care), industry 4.0, mobility and energy: solutions that require hardware and softwareto be inextricably linked. It is exactly that multidimensional R&D environment that we are currentlycreating at imec – by combining our nanoelectronics expertise with iMinds’ knowledge in researchdomains such as data science, cryptography and smart network technology.

At the same time, we realize that this knowledge cannot come from within imec alone. Hence, wewant to continue extending our collaboration with local university research groups, andinternational universities. After all, our worldwide ecosystem is fundamental to confirm and expandour international leadership and to address the growing need for innovation capability in ourindustry.

Collaborating closely with Flemish (industry) partners obviously remains a key focus area as well,providing them with expert guidance throughout their disruptive innovation trajectories. As theworld’s pioneering research center for nanoelectronics and digital technology, we are in anexcellent position to stimulate and realize high-tech innovation at a local, European and globallevel. As such, we can really lay a solid foundation for solutions that contribute to a better life foreveryone.

4/57

On top, we will continue to support venture initiatives that put high-tech solutions on the market,through programs such as imec.istart and imec.xpand; programs that are currently being reinforced.

We commit wholeheartedly to all these points in 2017. They will enable us to continue to respondflexibly to the rapidly changing needs of our society and make the impossible possible – togetherwith you; with imec, your strategic innovation partner par excellence, at your side!

Luc Van den hove,General Director and CEO of imec

5/57

Imec and iMinds become one

We announced it back in February and by September 2016 it was finally a done deal: the mergerbetween the research organizations imec and iMinds. This merger creates a unique research centerthat is at the very summit in the world of nano-electronics. It also excels for its knowledge insoftware and ICT. The expanded innovation center – which now operates under the imec name –puts this knowledge to good use to develop disruptive technologies in areas such as healthcare,smart cities and mobility, logistics and Industry 4.0 and energy.

Edition | January 2017

General

The highlights of 2016Life is busy! So you might not always have the time to keep upwith imec's latest news and achievements. On this page you canfind a quick overview of what imec has been doing in the pastmonth.

At imec, research and innovation in the fields of chiptechnology, ICT and software remain our essential focusareas. We work with industry to translate this expertiseinto applications for better health, smart cities, smart

mobility and efficient logistics and factories – notforgetting sustainable energy supplies. Here are 10

memorable highlights from 2016.

6/57

Our cleanroom expands

On 11 March, imec welcomed Flemish Minister-President Geert Bourgeois, Flemish Minister PhilippeMuyters and Mayor Louis Tobback as they officially opened the new cleanroom space. This brand-new high-tech lab, which is over 4000m2 in size, will be used for researching future generations oftechnology that will make microchips more powerful, perform better, reduce them in size and makethem more energy-efficient – all using the very latest and sophisticated chip productionequipment. The new cleanroom is an extension of the existing 300mm cleanroom, which now

covers an impressive 7200 m2.

RVO-Society blows out 15 candles

Fifteen years ago, RVO-Society was established in memory of Roger Van Overstraeten, the founderand first general manager of imec. The aim of RVO-Society is to encourage young people, agedbetween 5 and 25, to become interested in technology and science. RVO-Society does this bytranslating the knowledge found in innovative companies and research organizations (such as imec)into fun projects and educational activities for youngsters.

More information at www.rvo-society.be.

Another birthday: Holst Centre turns ten

In April, Holst Centre celebrated its tenth anniversary. This independent research center wasfounded by imec and TNO, with support from the local, regional and national governments ofBelgium and the Netherlands. The focus of Holst Centre’s research is on wireless autonomoussensors and flexible electronics. Over the past ten years, the number of researchers has risen from 5to 200. Today, Holst Centre works with 50 industrial partners.

Chip design receives boost at imec Florida

July saw the inauguration of imec Florida, a chip design center for photonics and high-speedelectronics. At the opening of the new facility a collaborative agreement was signed with theUniversity of Central Florida (UCF), Osceola County and the International Consortium for AdvancedManufacturing Research (ICAMR).

7/57

Prestigious ERC grants for three of our scientists

In 2016, Kris Myny (imec), Bart Vermang (IMOMEC) and Piet Demeester (iMinds - UGent) wereawarded European ERC research grants for young scientists. These awards are among the mostprestigious research grants for European scientists. Kris Myny received 1.5 million euro to improve avery promising type of thin-film transistor and to develop a new way of designing circuits that willsignificantly reduce energy consumption.

Bart Vermang was awarded a grant of 2 million euro. His ERC project is about improving CiGS(e)solar cells by using sophisticated technologies that are also applied to silicon solar cells. His work isaimed at producing more sustainable cells that deliver between 23% and 26% greater efficiency.

Piet Demeester received 2.5 million euro to develop a working ATTO prototype over the next fiveyears. ATTO is a wireless technology that can be used to provide every individual object in a largegroup of moving objects with a superfast mobile connection of 100 Gbps, which is a connectionwhere the signal delay is also kept to the minimum (under 10 microseconds). This is the basis for awhole series of new mobile applications that require a high level of calculating power quickly –such as intelligent ‘swarms’ (large groups, densely packed) of robots.

imec.academy brings top speakers to imec for ‘Nanotech for Health’ course

Genuine innovation is found by interfacing different disciplines: this is something that they havedefinitely understood at imec academy. For that reason, a ‘Nanotech For Health’ course is run eachyear for engineers and bio-specialists. Once again this year a totally new program was put togetherfor this 4-day course, which featured authoritative speakers from both the business and theacademic world. In addition to a unique range of courses, imec.academy also organizes trainingcourses specifically tailored for companies. Imec partner companies and staff can also consult theextensive library of recordings of the courses that have been presented in the past. Because lifelonglearning is a must in this rapidly changing world! More information: www.imec-academy.be

American Cancer Society praises imec team

Each year, imec’s ‘Levensloop’ relay team organizes a number of different programs to raise fundsfor ‘Levensloop Leuven’ and the Stichting tegen Kanker (Anti-Cancer Foundation). In 2016, imec alsoreceived the ‘Best Relay for Life Team’ award, selected from more than 500,000 teams involved inRelay for Life worldwide. The imec team won the award not only for the large amount of money itraised, but also for the emphasis it places on making people aware of their health and promoting ahealthy lifestyle. And not forgetting the amazing passion and enthusiasm that all of the volunteersbring to the work they do on the day of the event!

8/57

There were other support and solidarity programs in 2016, too: 15 teams raised money forschoolgirls in Kuria (Kenya) who attend the Visa Academy, a school for girls escaping female genitalmutilation. Imec staff also showed their support for the victims and families affected by theearthquakes in Italy, raising 3,280 euros to rebuild schools in the affected areas.

Gordon Moore wins imec ’s lifetime of innovationaward

You could call Dr. Gordon Moore the founding father of the semiconductor industry, after all howmany people have a law with their name on it that has governed and guided an entire industryworth hundreds of billions of dollars, created millions of jobs and has changed the way we all live ?Our CEO, Luc Van den hove, was honored that he could visit the co-founder of Intel Corporation,and the author of Moore's law, in the beautiful surroundings of Moore’s Hawaiian home. And he hada truly well-deserved present for him: this year’s imec “the Lifetime of Innovation award”, which isgranted every year at the Imec Technology Forum (ITF) in Brussels.

In 2017, ITF and iMinds The Conference will join forces. This first XL edition of ITF Belgium will takeplace on May 16-17, 2017 in Antwerp. Central theme is the unique combination of nanoelectronicsand digital technologies.

The imec.istart portfolio has over 100 start-ups

In addition to its research activities, imec helps researchers, young entrepreneurs and start-ups totake their ideas to market successfully. One of the tools developed to do this is imec.istart. Whenstart-ups are selected for this program, for the following 12 to 18 months they are able to call onfinancial support, professional coaching and a whole range of facilities customized to suit the needsof the aspiring start-up business.

Since the program began in 2011, more than a hundred entrepreneurial projects have been givensupport. This number is despite the very strict selection criteria applied by which only 1 out of every5 applications is approved. Between them, these start-ups have already generated more than 450full-time jobs and a total turnover in excess of 20 million euro (in 2015). In all, they have also raisedover 50 million euro in follow-up funding. Last year, imec.istart was placed 4th in the world by UBIGlobal in the category for ‘Top University Business Accelerators’ after more than 500 incubationprograms were screened from all over the world.

9/57

Introduction

The explosive growth of data traffic fuels the demand for ever more processing power and storagecapacity. Moore’s Law continues to be necessary, but innovations are needed beyond this law tohelp managing the devices power, performance, area and cost. An Steegen reveals some of thesecrets of semiconductor scaling – a pipeline full of materials, device architectures and advancedtechniques that promise to further extend semiconductor scaling.

Edition | January 2017

Semiconductor technology & processing

Vision: Technologies toextend semiconductorscaling

10/57

The end of happy scaling?

Data traffic explosion, fueled by the Internet of Things, social media and server applications, hascreated a continuous need for advanced semiconductor technologies. Servers, mobile devices, IoTdevices... they drive the requirements for processing and storage. An Steegen: “At the same time,this trend is also creating more diversification. IoT devices for example will need low-power signalacquisition and processing, and embedded non-volatile memory technologies. For mobile andserver applications, on the contrary, further dimensional scaling, continuous transistor architectureinnovations and memory hierarchy diversification are among the key priorities.” But will we be ableto continue traditional semiconductor scaling, as initiated by Gordon Moore more than 50 yearsago? An Steegen: “For a long time, we have lived in the happy scaling era, where every technologynode reshrinks and redoubles the number of transistors per area, for the same cost. But the last10-12 years, we have not been following that happy scaling path. The number of transistors stilldoubles, but device scaling provides us with diminishing returns. We’ve seen these dark periods of‘dark silicon’ before, but, fortunately, we’ve always managed to get out of these periods. Again, thetechnology box will provide new features to help manage power, performance and area node bynode as we move to the next generation.”

The technology box for dimensional scaling

On the dimensional scaling side, extreme ultraviolet lithography (EUVL) is considered an importantenabler for continuing Moore’s Law. An Steegen: “Ideally, we would need it at the 10nm node, wherewe will start replacing single exposures with multiple exposures. More realistically, it will hopefullybe ready to lower the costs for the 7nm technology. At imec, we already showed that EUVL iscapable of printing 7nm logic dimensions with one single exposure.” Still, issues need to be resolved,related to, for example, the line-edge roughness. An Steegen: “At the same time, to enhancedimensional scaling, we increasingly make use of scaling boosters, such as self-aligned gate contactor buried power rail. These tricks allow a standard cell height to be reduced from 9 to 6 tracks,leading to a bit density increase and large die cost reduction - a nice example of design-technologyco-optimization.”

11/57

Improving power/performance in the front-end of line

FinFET technology has been the killer device for the 14 and 10nm technology nodes. But for the7-5nm, An Steegen foresees challenges. “At these nodes, FinFET technology can’t meet the 20%performance scaling and 40% power gain anymore. To go beyond 7nm will require horizontal gate-all-around nanowires, which promise better electrostatic control. In such a configuration, the drivecurrent per footprint can be maximized by vertically stacking multiple horizontal nanowires. In 2016,at IEDM, we demonstrated for the first time the CMOS integration of vertically stacked gate-all-around Si nanowire MOSFETs. Vertical nanowires, although requiring a more disruptive process flow,could be a next step. Or junction-less gate-all-around nanowire FET devices, which, as shown at the2016 VLSI conference, appear as an attractive option for advanced logic, low-power circuits andanalog/RF applications.” Further down the road, from the 2.5nm node onwards, fin/nanowiredevices are expected to run out of steam. An Steegen: “Sooner or later, we will need to find thenext switch. Promising approaches are tunnel-FETs, which can provide a 3x drive currentimprovement, and spin-wave majority gates.” Spin-wave majority gates with micro-sized dimensionshave already been reported. But to be CMOS-competitive, they must be scaled and handle waveswith nanometer-sized wavelengths. An Steegen: “In 2016, imec proposed a method to scale thesespin-wave devices into nanometer dimensions, opening routes towards building spin-wave majoritygates that promise to outperform CMOS-based logic technology in terms of power and areareduction.”

Extending or replacing Cu in the back-end-of-line

Looking ahead, it might as well be the interconnect that will threaten further devicescaling. Therefore, the back-end-of-line (BEOL) and the struggle to keep scaling theBEOL needs attention as well. “We look at ways to extend the life of Cu, for examplewith liners of rubidium (Ru) or cobalt (Co). On the longer term, we will probably needalternative metals, such as Co for local interconnects or vias”, says An Steegen.

12/57

The future memory hierarchy

Besides a central processing unit, memory to store all the data and instructions is another keyelement of the classical Von Neumann computer architecture. The ever increasing performance ofcomputation platforms and the consumer’s hunger for storing and exchanging ever more data drivethe need to keep on scaling memory technologies. Besides this scaling trend, existing memories thatmake up today’s memory hierarchy are challenged with the need for new types of memory. AnSteegen: “STT-MRAM, for example, is an emerging memory concept that has the potential tobecome the first embedded non-volatile memory technology on advanced logic nodes foradvanced applications. It is also an attractive technology for future high-density stand-aloneapplications. It promises non-volatility, high-speed, low-voltage switching and nearly unlimitedread/write endurance. But its scalability towards higher densities has always been challenging.Recently, we have been able to demonstrate a high-performance perpendicular magnetic tunneljunction device as small as 8nm, combined with a manufacturable solution for a highly scalable STT-MRAM array.” The future memory landscape also requires a new type of memory able to fill the gapbetween DRAM and solid-state memories: the storage class memory. This memory type shouldallow massive amounts of data to be accessed in very short latency. Imec is working there onMRAM and resistive RAM (RRAM) approaches.

Beyond classical scaling – towards system-technology co-optimization...

A challenge for traditional Von Neumann computing is to increase the data transfer bandwidthbetween the processing chip and the memory. And this is where 3D approaches enter the scene. AnSteegen: “With advanced CMOS scaling, new opportunities for 3D chip integration arise. Forexample, it becomes possible to realize different partitions of a system-on-chip (SoC) circuit andheterogeneously stacking these partitions with high interconnect densities. At the smallestpartitions, chips are no longer stacked as individual die, but as full wafers bonded together.” Anincreased bandwidth is also enabled by optical I/O. In this context, imec continues its efforts torealize building blocks (e.g. optical modulators, Ge photodetectors) with 50Gb/s channel data ratefor its Si photonics platform.

Moore’s Law will continue, but not only through theconventional routes of scaling.

13/57

An Steegen: “We have moved from pure technology optimization (involving novelmaterials and device architectures) to design-technology co-optimization (e.g. the useof scaling boosters to reduce cell height). And we are already thinking ahead about anext phase, system-technology co-optimization. And to keep computing powerimproving, we are exploring ways beyond the classical Von Neumann model, such asneuromorphic computing, a brain-inspired computer concept, and quantumcomputing, which exploits the laws of quantum physics. There are plenty of creativeideas that will allow the industry to further extend semiconductor scaling...”

Biography An Steegen

An Steegen is imec’s Executive Vice PresidentSemiconductor Technology & Systems. In thatrole, she heads the research hub’s efforts todefine and enable next-generation ICTtechnology and to feed the industry roadmaps.Dr. Steegen is a recognized leader insemiconductor R&D and an acclaimed thoughtleader and speaker at the industry’s prominentconferences and events. An Steegen joinedimec in 2010 as senior VP responsible for imec’sCORE CMOS programs in logic and memorydevices, processing, lithography, design, andoptical & 3D interconnects. Before, she wasdirector at IBM Semiconductor R&D in Fishkill,New York, responsible for the bulk CMOStechnology development. While at IBM, Dr.Steegen was also host executive of IBM’s logicInternational Semiconductor DevelopmentAlliance and responsible for establishingcollaborative partnerships in innovation andmanufacturing. Dr. An Steegen holds a Ph.D. inMaterial Science and Electrical Engineering,which she obtained in 2000 at the KU Leuven(Belgium) while doing research at imec. She haspublished more than 30 technical papers andholds numerous patents in the field ofsemiconductor development.

14/57

Introduction

The IoT is fast becoming a multilevel system of systems spanning the globe. But to realize thegrowth path that is forecasted, we’ll need optimized and specialized hardware, capable of, amongstothers, sensoring at ultralow power, guaranteeing a system’s security during its full lifetime, andlearning from huge amounts of data. Imec’s Diederik Verkest and Ingrid Verbauwhede talk about thenext step: how technology can be further optimized to solve specific system and applicationdemands. As examples, Ingrid proposes hardware-entangled security and Diederik explains imec’sefforts in neuromorphic processing.

A heterogeneous chip future (Moore on steroids)

“Until recently,” says Diederik Verkest, “we concentrated almost all of our scaling effort on thesmallest unit of a chip, the transistor, whatever that unit was used for afterwards. Next, to stay onthe course predicted by Moore’s Law, we co-optimized technology with lower-level design unitssuch as e.g. memory cells. Now we’re working our way up in the system hierarchy, always lookinghow we can optimize technology to better implement a function. So naturally, we also arrive at thekey functions needed for the future IoT, such as a failsafe security. And we are also eying specializedprocessors for e.g. neuromorphic computing, complete subsystems to tackle specific, hardproblems.”

Edition | January 2017

Internet of Things, Heterogeneous integration

Vision: Optimizingtechnology for IoT systems– adding fingerprints andbrains

15/57

Verkest adds that he is excited about imec’s recent merger: “This is a great opportunity for bothsides. My new colleagues are application experts in domains such as bio-informatics or security.They can help us screen technology and direct us to the solutions that are best fit to solve the hardproblems in their domains. And reversely, as application experts, they will learn from all thehardware opportunities that we are considering. What I see happening going forward, is a muchmore intimate, structured interaction between hardware and application R&D, greatly speeding upthis system/technology co-optimization.”

Secure chips with unclonable fingerprints

Today already, electronics are embedded in many objects in our environment. Think of your car’skeys, security cameras, smart watches, or even implanted pacemakers. “This makes securityconsiderably more complex than it used to be," says Ingrid Verbauwhede. "Existing cryptographicalgorithms demand a lot of compute power, so they run mainly on high-end platforms. But mostmicrochips in the IoT are small, lightweight, low-power and have a limited functionality. Sotraditional cryptography doesn’t fit well. Our ambition is to make chips that are inherently moresecure through the way they are designed and processed."

In 2016, Ingrid Verbauwhede and her team received a prestigious European ERC research grant fortheir Cathedral project.

“This grant is at once a recognition for what we have been doing, and a great support going forward.A support that will allow us to independently look for the best solutions.”

Ingrid’s team is exploring various ways of doing that:

“In the past, R&D looked at dedicated design methods for e.g. low-power chips. We now want to dothe same to better secure chips. Chips e.g. that don’t leak information while they are computing sothey are more resistant against side-channel attacks. And another of our focus points isimplementing future-proof cryptography, algorithms that will protect a system during its longlifetime, even if it is attacked by future quantum computers.”

Asked for her plans for 2017, Ingrid Verbauwhede points to the direct access her team now has totechnology processing at imec’s fabs:

16/57

“One of the characteristics of today’s chip scaling is process variation: each chip is slightly differentfrom all others. From a reliability perspective, that is a nuisance. It requires engineers to take extrameasures so that computations remain predictable. But there is also an upside that we want toexploit: the variations are like a fingerprint, a way to uniquely identify each chip without expensivecalculations. It is what we call a physically unclonable function (PUF). And if you tie that function tothe software running on the processor, you have another layer of security which is well-suited forIoT devices.”

Smart chips with brain power

Our brains are formidable computing wonders, using only a fraction of the power of traditionalcomputers to obtain comparable results. Therefore, engineers are eager to mimic the brain on chip,e.g. to speed up deep learning from massive amounts of data, or low-power image recognition.

“But to do so,” says Diederik Verkest, “we have to replicate the brain’s architecture, a tightinterconnection of an enormous number of relatively primitive processing nodes (the neurons) andtheir interconnections (the synapses). That is usually done with some type of crossbar architecture,wires laid out in a matrix (or cube), so that each input line connects with all outgoing lines. At acrossing of two lines, there is a switch that implements the synapses. They contain the intelligenceof the system, the ability to hold data, process and learn from experience. So they should be madeprogrammable and self-adaptable.

Work on this emerging domain at imec started some two years ago, partly embedded in theEuropean Horizon2020 project NeuRAM3. In 2016, we have selected an architecture and screenedoptions to implement the self-adapting synapses. We are convinced that our concept is uniquelysuited to tackle the problem, so we’ve taken out a patent and are now building a proof-of-concept.In 2017, we will tape-out a first chip and package it into a neuromorphic computing system that wecan test against neuromorphic application simulators with growing numbers of neurons.”

These brain-on-chips may not be exact copies of our brain circuits, but nature teaches us that it isphysically possible to build much better computers than we do today. Computers that we need tomake sense from the enormous amounts of data that the IoT will generate. But also for theintelligent sensors and robots of the connected world. Small, low-power, long-lasting devices thathave to stand their ground among an ever growing stream of data, continuously adapt themselvesto their environment, even learn and become smarter over their life time.

17/57

Biography Diederik Verkest

Diederik Verkest is director of imec’s INSITEprogram. After earning a Ph.D. in micro-electronics engineering from the KU Leuven,Diederik joined imec in 1994, where he has beenresponsible amongst others for hardware/software co-design. In 2009, he started imec’sINSITE program focusing on co-optimization ofdesign and process technology for sub-14nmnodes. The program offers the fabless designcommunity insights into advanced processtechnologies and provides a platform forfoundries and fabless companies to discussdirections for next generation technologies. Diederik Verkest published over 150 articles ininternational journals and at internationalconferences. Over the past years he has beeninvolved in numerous technical conferences.He was the general chair of, DATE, the Design,Automation, and Test in Europe conference in2003. Verkest is a Golden Core member of IEEEComputer Society.

18/57

Biography IngridVerbauwhede

Ingrid Verbauwhede is professor at the KULeuven (Belgium) in the imec-COSIC researchunit where she leads the embedded systemsand hardware group. She is also adjunctprofessor at the electrical engineeringdepartment at UCLA, Los Angeles (USA). Ingrid Verbauwhede is an IEEE fellow, memberof IACR and she was elected as member of theRoyal Academy of Belgium for Science and theArts in 2011. Her main interest is in the designand design methods for secure embeddedcircuits and systems. She has published around 70 papers in international journals and 260 papers at international conferences. She isalso inventor on 12 issued patents.

19/57

Thanks to microchip technology, every object – machines, buildings, vehicles, personal appliances –can be connected to the Internet to provide us with a continuous flow of useful information. It isexpected that by 2020, this Internet of Things (IoT) will connect billions of devices. Imec develops the building blocks and digital technology for an ‘intuitive’ IoT – an Internet ofThings that discretely runs in the background, yet is instrumental to increasing our wellbeing andcomfort. Concretely, imec’s intuitive IoT consists of ‘thinking objects’ – networked sensors that constantlymonitor the environment, provide status reports, receive instructions, and take short-term andlong-term actions based on intelligent processing of the gathered data. This intuitive IoT willinteract with us and learn from our habits, preferences and health, … It will help us make better-informed decisions while taking our privacy and security concerns into account. And it will help uscreate a more sustainable and safer world at large.

Edition | January 2017

Internet of Things

Intro: Enabling the intuitiveInternet of Things

Such an intuitive IoT will have enormous potential. But inorder to fully unlock its economic and societal value, a

number of challenges need to be tackled.

20/57

First, we need further innovation in hardware. We need single chip integrated wireless sensormodules, containing sensors, actuators, processing, storage and communication abilities. Andsecondly, we have to find solutions for the security and privacy considerations that werementioned before, and solutions to cope with the tsunami of data that is being generated. Pleaseread our experts story on this subject on the next pages.

Imec plays a leading role in studying and solving some of these issues. It has the critical mass torealize breakthroughs that will make the intuitive IoT a game changer in various application domains– such as smart health(care), smart mobility, smart cities, smart industries, and smart energy. Theseapplication domains are highlighted elsewhere in this annual overview.

21/57

Introduction

There is still a great deal more innovation required in hardware if the intuitive Internet of Things isto be achieved. This includes advanced sensor modules and wireless communication chips thatenable sensors to talk both with each other and with their environment. But if we want to see areliable, secure and efficient IoT, we will also have to make a lot more effort in terms of digitaltechnologies and software. Rudi Cartuyvels, Executive Vice President Smart Electronics at imec, andDanny Goderis, Executive Vice President Smart Applications at imec and the former CEO of iMinds,explain the main developments in their fields and give us a taster of what the interaction betweenmicrochip and digital technology can create.

Enabling the wireless connectivity of sensors

A functional Internet of Things is inextricably linked with reliable wireless communication thatenables the various sensors, equipment and machines to ‘talk’ both with each other and with thecloud.

Edition | January 2017

Internet of Things, Data science and data security

Vision: Interplay betweenchip and digitaltechnologies crucial for theIoT

22/57

Rudi Cartuyvels: “For that to happen you need wireless communication chips for theIoT’s sensor networks. The sensors in these networks will be powered by batteries andthat’s an area that poses an enormous challenge for us: at the moment we only havevery low power available to supply the communication chips. In addition to that, thewireless communication chips have to be capable of transmitting data over arelatively long distance (we’re talking about kilometers here). They also need tocomply with the communication standards proposed for the IoT, just as for Wi-Fi andBluetooth Low Energy.”

Imec and Holst Centre have built up many years of experience in designing ultralow-powercommunication chips.

Rudi Cartuyvels: “We again had excellent results in 2016, such as at the ISSCCconference. Our achievements included developing a transceiver that is optimized forIoT applications and which is in line with the low-power, long-distance Wi-Fi protocolIEEE802.11ah. We are also working on solutions that meet the Bluetooth Low Energystandard for IoT applications, as well as looking at combined solutions thatincorporate Bluetooth and Wi-Fi on the same chip.”

One obvious solution would be to use the cellular phone network that is available everywhere.

Rudi Cartuyvels: “At the moment we are not yet able to use this technology in sensorsbecause it uses too much power. However, we are working on solutions that complywith the narrowband IoT standard and which ultimately will enable sensors tocommunicate over long distances via the cellular network.”

23/57

Sending large volumes of data wirelessly

Data speeds ranging from kilobits to megabits per second are sufficient for sensor networks. Butwhat happens if large volumes of data have to be sent wirelessly and at high speed? Researchers ofimec - UGent - IBCN (a former iMinds research group) are working on ultrahigh-speed wirelessconnectivity that uses ATTO technology. Danny Goderis: “ATTO technology is a development fromwireless ‘small cells’ technology in which large numbers of antennas each cover a limited area (orcell) to make fast wireless broadband connections possible. Professor Demeester’s team aims tocheck whether this technology can be used to provide each object in large groups of movingobjects with a superfast mobile connection of 100Gbits per second and as little signal delay aspossible.” Professor Demeester was awarded an ERC grant for this work in 2016. He will use thefunds to develop the technology further over the coming years. Initially, the technology will beused in production environments to enable flexible swarms of intelligent robots to work inharmony with humans.

Danny Goderis: “But we also hope to lay the foundations for a whole series of othermobile applications that need high levels of fast calculating power. And I see ATTO asan enabler for the ‘mass customization of products’, a trend that I believe in strongly.Mass customization, as opposed to mass production, will enable customers to adaptproducts to their own requirements. Intelligent robots that can be reconfiguredquickly – and preferably wirelessly – will be needed to do this.”

An alternative way to transport huge amounts of data makes use of millimeter wave technology.

Rudi Cartuyvels: “We will need to be able to send large amounts of data from thesensor networks to the cloud, at very high speeds and wirelessly. We are looking atwireless solutions that will reach speeds of up to 20Gbits per second, using millimeterwave technology in 60GHz. We use beamforming for this, which enables directionalsignal transmission between a transmitter and a receiver, at very high speed. In 2016 atISSCC we worked with the Vrije Universiteit Brussel and Holst Centre to present alow-power demonstration chip in 28nm CMOS technology for 60GHzcommunication. We were able to achieve data speeds of almost 5Gbits per secondover a distance of 1 meter. At the imec campus, we demonstrated data rates of1.5Gbits per second over 100m distance. We are currently working on solutions thatwill enable even higher data speeds and a longer range.”

24/57

Testbed

Ultimately, the technologies used for the IoT (such as sensor platforms and wireless communicationtechnologies) must also be linked with each other and be rolled out to scale in a genuine testenvironment. Rudi Cartuyvels: “At imec we have a great deal of expertise in developing integratedwireless sensor modules, for example for measuring ambient gases, fluids or body parameters. But itis also very important to know how these technologies behave in ‘real life’. As a result, working withthe former iMinds (now merged with imec) is very important. They have a lot of experience insetting up testbeds that enable technologies to be validated.” The flagship project is the City ofThings in Antwerp, Belgium, in which researchers will roll out a hundred gateways and a multitudeof sensors in a city infrastructure by the end of 2017.

“But the software is also important,” says Rudi Cartuyvels. “In the end, we also need tohave a software platform capable of controlling and managing the hardwarecomponents. One of the challenges here is the heterogeneity of the network: a lot ofdifferent technologies will have to work together as part of the same network.”

Here again our expertise in digital technologies will come into its own.

Danny Goderis: “In the Internet of Things there will be far more ‘any-to-any’connectivity than in a conventional network. To be able to run and manage thatnetwork, you need sophisticated tools. Our researchers are developing theseoperations management (or OM) systems. We are working on a plug-and-play designthat will give us a simple way of plugging in sensors and wireless interfaces so thatthey can then be programmed, upgraded, monitored or managed. We produced aprototype for this in 2016.”

Safety first

One of the major challenges facing the IoT is protecting the data that will be collected by all ofthese sensors in our home, on our body or in our car.

Rudi Cartuyvels: “In the future, data security will become extremely important.Furthermore, standards will have to be created that people will have to comply withbefore a device can be connected with a networked environment. Data protectionwill be a combination of both software and hardware security.”

25/57

However, security at a sensor or chip level is not so straightforward in view of their limitedcalculating power and battery content.

Danny Goderis: “There are already methods in place today, such as complexcryptographic algorithms, for protecting devices against hackers. But these forms ofsecurity require a great deal of calculating power and energy. At imec - KU Leuven -COSIC (a former iMinds research group) new ways of securing microchips are beinginvestigated. In 2016 Ingrid Verbauwhede was awarded a European ERC AdvancedGrant that she will use over the coming five years to develop new protectivemechanisms for making electronics more resistant to abuse.”

This is another fine example of what the interplay between hardware and software can create...

26/57

Biography Rudi Cartuyvels

Rudi Cartuyvels is Executive Vice PresidentSmart Electronics at imec. He is passionateabout delivering industry-relevant innovationsin electronic microsystems and nanosystems,enabling novel applications in the field of IoT,healthcare and energy markets. Imec’s SmartElectronics delivers innovative electronicsystem platforms to imec’s global partnernetwork, enabling sensing and wirelessconnectivity for smart cities and vehicles,wearable electronics for health and lifestyle,microfluidic and electro-optical componentsfor medical diagnostics, thin-film electronicsfor flexible displays and smart tags,photovoltaic energy generation for smartbuildings, solid state batteries and GaN-on-Sipower electronics. Rudi Cartuyvels holds aMaster Degree in Electrical Engineering fromthe KU Leuven, Belgium. He started his careerat imec in 1990 in advanced CMOStechnologies. He was appointed Director ofInterconnect Technologies in 2001 and has heldseveral management roles in semiconductortechnology development and smart electronicsystems. He was appointed Executive VicePresident in 2016 and directs imec’s R&D inSmart Electronics.

27/57

Biography Danny Goderis

Danny Goderis is Executive Vice PresidentSmart Applications at imec and was formerlyCEO of iMinds. Before he joined iMinds in 2012,Danny Goderis worked in research, venturingand strategic marketing. He ran Bell Labs in theBenelux, which is the Research and Venturingorganization of Alcatel-Lucent (one of iMinds'strategic partners). The areas of research thereincluded fixed access (DSL, fiber optics), homenetworking, (3D) video research, ICT Web 2.0 &telecoms applications and cloud computing.Before joining Alcatel-Lucent, Danny workedwith several Belgian universities in a range ofdoctoral and post-doctoral positions. Hereceived his doctorate in Science and aMaster’s in Physics from KU Leuven and is theauthor of more than 50 publications. Dannyalso has a Master’s in Business and MarketingManagement.

28/57

Introduction

A new age has dawned: the age of the 1000-dollar genome (which means a full genome can besequenced for 1000 dollars)! It’s an important milestone for medicine and our experts Yves Moreauand Pol Van Dorpe, each in their own individual field, contribute to this. Moreau spends his timeporing over the mass of data that comes with DNA sequencing, while Van Dorpe puts chiptechnology to good use to make DNA sequencers cheaper and more efficient.

A genetic passport for everyone?

There are 23 pairs of chromosomes in every cell in your body. These are made up of the DNAcontaining the code by which all of our inherited traits are established. This code consists of around3 billion ‘letters’ or base pairs. Moreau: “Within a foreseeable space of time, determining someone’sgenetic code will be as normal as doing a CT scan. Every child will be able to have his or her geneticpassport from birth. It will become the foundation for a healthy life: am I likely to suffer fromcertain diseases and how can I avoid them? What medication works best for me? And so on.”

Edition | January 2017

Life sciences, CSR

Vision: DNA sequencingbecoming routine

29/57

And although your genetic code is established from birth, you can still be ‘sequenced’ several timesduring your life. Van Dorpe: “DNA sequencing will become so accurate that you’ll be able to use ablood sample to go looking for ‘rare’ DNA, such as tumor cells or even viruses. With tumor cellsthere are two types of DNA found in the blood: the DNA of ‘circulating’ tumor cells (that havebroken away from the original tumor and are spread via the blood), and ‘cell-free’ DNA (from dyingtumor cells). And we can learn a great deal from this. For example, you can find out (1) that you havea tumor, (2) which is the primary tumor (hence where it should be) – all without scans/operations,and (3) the DNA analysis can also tell you what specific treatment is appropriate. All this is referredto as a ‘liquid biopsy’, which is the ability to take a biopsy of the tumor without having to cut intothe tumor, but based just on a blood analysis.”

“In addition to our own cells, with our own genetic information, the amount of bacteria our bodyhouses is over 10-fold the number of human cells. There is increasingly more evidence that thebacteria population determines to a great extent our health. A genetic analysis of your intestinalflora can be of interest in guiding your health.”

Moreau: “As far as technology is concerned, we are taking enormous strides forward by finallybringing the 1000-dollar genome within reach. But, of course, there is also a societal side to all that.When is it justified to carry out a genetic test? When and how should your doctor explain to youabout any genetic abnormalities you may have? Or do you opt not to have a genetic passport madeat all? These are all questions that ethics specialists are considering at the moment.”

Terabytes of data

Yves Moreau is a member of the STADIUS research group at KU Leuven. STADIUS is a group from theformer iMinds (now integrated within imec) specializing in data analysis for detecting rare geneticdiseases. Moreau: “We develop algorithms to deduce from the 3 billion ‘letters’ of your geneticcode whether you have a rare genetic disease. There are approximately 7,000 rare diseases thataffect 4 to 8% of the population. If you can determine from a young age what specific disease achild is suffering from, it is easier for the family to come to terms with it. They then know how thedisease will develop and whether there is any likelihood that a subsequent child will also have thesame abnormality. In the end, researchers will be able to use the abnormal genetic code to gain aninsight into the mechanism behind the disease and develop better treatments as a result. 1000dollars for a genome perhaps still sounds like a lot, but it’s not when you add up the costs of all thetests and visits to the doctor that were needed previously in the search for the correct diagnosis ofa rare disease. Not to mention the pain and distress that the family experiences by not knowingwhat is going on with their child.”

Determining the genetic code is still not a routine task, but what happens when it does? Moreau:“We are going to see an explosion of data. And that quantity of data is one of the main challengesfacing us: how are we going to be able to store all of those terabytes of data securely and at anaffordable price? And having done that, how will we analyze and interpret it?”

30/57

“Our research group has spent 2016 working mainly on solutions for determining which differencesin the genetic codes between different individuals (around 3 million differences out of the 3 billionbase pairs) are important and which ones are not. Which differences end up causing diseases? Usingour algorithms we are able to identify mutations that may be the cause of a particular disease. Fromthere on, doctors study it more in detail.”

The more genetic data can be collected, the more insight can be gained by doctors on diseases andtheir therapy. Moreau: “Over the past year we have also been looking at how genetic informationcan be shared securely. Storing everything in a single central database would not be a good idea,because there is always the danger of hacking. Which is why we have developed a model in whichdata is stored locally, but can be viewed and consulted remotely from anywhere. For instance, adoctor is able to retrieve the genetic data of all of the patients in the world with a specific disease,although he will not be able to access an individual genome containing personal data.”

From 1000 dollars down to 100 dollars for a genome?

Scientists are over the moon with the 1000-dollar genome; but for ordinary people that still soundslike a lot. Especially as it’s a cost that the health funds aren’t reimbursing (yet). But how can we drivedown the cost of a genetic passport even further? For the answer we can ask Pol Van Dorpe. He hasa leading role in imec’s research group that produces photonics solutions for life scienceapplications. Van Dorpe: “This 1000-dollar figure is relative. Currently it represents the cost ofsequencing an entire genome if you could run a DNA-sequencer continuously for a whole year,thereby reducing the cost per genome. But the cost of the equipment involved is falling; thequantity of DNA that can be analyzed per hour – the throughput – is also increasing. We are usingour expertise in nanophotonics, CMOS sensors, integration and chip production to improve thisequipment. And we are working on it with various major players, too. For example, PacificBiosciences has been able to reduce the cost of its equipment by 50% and their chips are generating7 times more information as the result of our working together.”

“In addition to lowering the cost of the equipment and increasing throughput, we are also makingprogress towards ‘long reads’ – in which long strands of DNA can be read. This compares withearlier techniques in which the DNA was divided into small pieces, multiplied and then read. Theadvantage of ‘long reads’ is that we can put together a person’s entire genetic map much moreaccurately and quickly than before. It also makes it possible to detect structural variations. Theseare pieces of code that are repeated in the genome at specific places. Any abnormalities in thispattern may also lead to health problems.”

“There are various technologies on the market, as well as equipment in all shapes and sizes – fromlarge DNA-sequencers like they use at Pacific Biosciences, down to small handheld devices such asthe MinION from Oxford Nanopore Technologies. So different technologies will live side by side,each one for a different specific market. Small, wearable devices may also be important if you wantto research diseases such as ebola in remote areas, whereas the large equipment is useful forproducing complete genetic maps quickly and accurately. Chip technology is particularly importantfor making these systems more compact, faster and less expensive.”

31/57

Biography Yves Moreau

Yves Moreau is Professor at imec - KU Leuven(STADIUS). Yves Moreau received his PhD at KULeuven ESAT-SCD in 1998. Between 1998 and2005, he was a postdoctoral researcher (FWO-Vlaanderen) and assistant professor at ESAT-SCD, developing bioinformatics research. Since2004, he has been a lecturer and professor atESAT-SCD, coordinating SymBioSys, the KULeuven Center for Computational SystemsBiology. He is also the program director of theMaster of Bioinformatics. He conducts researchinto computational methods for diagnosis anddisease gene discovery in congenital geneticdisorders and teaches several bioinformaticscourses, mainly focusing on probabilisticmodels in computational biology. Yves isassociate editor for IEEE Transactions onComputational Biology and Bioinformatics. Heis also a co-founder of two spin-off start-upsfrom the university: Data4s, now part ofNorkom Technologies, specializing indatamining for the banking industry, andCartagenia, specializing in IT solutions forclinical genetic diagnosis.

32/57

Biography Pol Van Dorpe

Pol Van Dorpe received his PhD from thefaculty of engineering of KU Leuven for hiswork in the field of spintronics. Afterwards hewas appointed as a postdoctoral fellow ofFWO-Flanders (2006-2012), based at imec,where he focused on metal-basednanophotonics and plasmonics for biosensorsand energy-harvesting. During this period, Polworked for some time at Stanford University,where he set up worldwide collaborations withrenowned scientists in this field. His work hasled to over 100 peer-reviewed papers in high-impact factor journals and has attracted morethan 5000 citations. Since 2012 he has held theposition of part-time associate professor in thephysics department at KU Leuven and he is aleading member of staff in the life sciencesdepartment at imec, where his main researchfocus is applying integrated photonicsconcepts to enable novel applications in thelife sciences field.

33/57

Introduction

As technological developments open the door to safer, greener and cheaper mobility solutions, andmore opportunities arise for companies to strengthen their position in the market, new players arealso introduced, causing a shift in the sector’s traditional structure. Wim Van Thillo and ErikMannens shed light on imec’s role towards more sustainable and efficient transportation options.

The revolution in the automotive sector

Imec’s vision of smart mobility includes comfortable and safe transportation for everybody, andaims at solving challenges such as traffic jams, car accidents and lengthy commute time. To achievethis, imec makes use of its leading role and expertise in sensor technology to develop automotive-oriented radar sensors, a program led by Wim Van Thillo and Massimiliano Maranella.

For cars to drive autonomously, they need to be able to understand their surroundings, assess trafficand detect objects. Thanks to imec’s 79 GHz sensor chips program – “the next generation of radarsin the automotive industry” – unlicensed high-resolution radar systems are able to operate withoutinterference from existing applications. “We want to create lower power, lower cost, smaller size,higher accuracy radars,” affirms Wim Van Thillo.

The development of self-driving cars has been gaining a lot of traction, introducing new players inthe manufacturing sector, such as Google, Apple or Uber. This, together with the increasedconsolidation the automotive sector has been witnessing, represents a significant transformation,which constitutes both a challenge and an opportunity for imec researchers.

Edition | January 2017

Smart Mobility, Radar technology

Vision: A new era in mobility

34/57

“We work with partners to help conduct our research and develop our technologies. The increasedconsolidation of the chip industry has been a challenge in the past year, since it means fewerplayers are available to support our projects,” states Wim Van Thillo. “On the other hand, thetraditional structure of the sector is also changing, with new players without any previous links tothe sector entering the race for the self-driving car. This shift can come as an opportunity for imecto go deeper in the field of smart mobility.”

The next step towards the smart, autonomous car

Detecting obstacles is only a first step towards a truly autonomous car. The next challenge is makingsure it can identify the obstacle (i.e. other vehicles, road signs, pedestrians) and automatically actupon it. Moreover, it should also be able to communicate with other cars on the road.

Achieving such level means gathering data from all individual cars’ sensors, the environment andother external information sources, analyzing it and interpreting it in order make intelligent, real-time decisions. Both Wim Van Thillo and Erik Mannens (imec’s expert in Data Science) agree that thisis where imec’s Smart Applications (former iMinds) expertise in the domain of Data Fusion will becrucial and how the merger between imec and iMinds increases imec’s role in this domain.

“Each car produces an average of 4.000 GB of data per day,” discloses Erik Mannens. “However,more than its amount, the real challenge lies in the variety of data that needs to be collected andprocessed, such as radars, lidars or cameras.”

According to Erik Mannens, two aspects need to be taken into account: safety and reliability. On theone hand, a self-driving car must be able to interpret data accurately and fast enough in order toassure the safety of everyone – passengers, other vehicles, bikers, pedestrians. Moreover, the car’ssoftware and hardware components must be flexible, in order to adapt to today’s fast-pacedtechnological evolution. “The average lifetime of a car, from drawing board until it’s discontinued, isaround 20 to 25 years - of which the first ten years are reserved for R&D prior to manufacturing,”states Erik Mannens. “This means that the cars that are launched on the market today need to beable to cope with the technical developments of the next 10 to 15 years, which is primarily afirmware and software update problem.”

Privacy, ethical and security issues need to be considered as well. Imec’s security experts are lookinginto new encryption methods to prevent connected cars from being hacked. At the same time,imec’s legal and ethical experts are studying which laws and regulations should be in place to ensurewe can all enjoy the benefits of a smart, connected world to the fullest.

35/57

Enabling citizens to make smart commuting decisions

Smart mobility is more than developing self-driving cars. It’s also about offering users informationon all the transportation options they have at their disposal, thus enabling them to make informed,real time decisions on their route planning.

To achieve that, data from private citizens (e.g., Is there a car or a bike at their disposal? Do theyhave special accessibility needs?), public transport companies, the government, and other thirdparty entities – such as Uber, BlaBlaCar or other ride sharing services – needs to be publiclyavailable and adjustable to users’ needs. “The more contextual data we have, the more accurateinformation we can provide,” states Erik Mannens. The key is using Linked (Open) Data, a solutionthat Erik Mannens and his team at imec – UGent – IDLab have been working on for the past years.

As part of his PhD, imec – UGent – IDLab’s researcher Pieter Colpaert is now working on LinkedConnections, a technology that opens up transport data in a standardized format to facilitate itsuse by developers in third-party applications. The ultimate goal is to be able to combine data fromdifferent sources for more complex route planning. “Up until now, only data from the Belgiannational railway company NMBS is available. Our challenge in the upcoming months is to get moreservice providers on board and to develop a solution for citizens to automatically add theirpersonal contextual input, such as their spatio-temporal history on commuting, route preferences,or special needs,” reveals Erik Mannens.

Erik Mannens has also been working with federal and regional governments in Belgium on severalother open data projects. The latest one involved the Department of Mobility and Public Works ofthe Flemish Government (MOW) to help them use traffic information collected from the nationalhighways as Linked Open Data so it can be re-used by anyone who wants to create a route planningapp or use it in their route planning system. The successful results opened the door for follow-upprojects in 2017.

36/57

Biography Wim Van Thillo

Wim Van Thillo received his master degree inelectrical engineering and his undergraduatedegree in business economics from the KULeuven, Belgium. He obtained a PhD degreefrom the same university based on his researchin imec’s wireless communications group. In2008, he was a visiting researcher at UCBerkeley’s Connectivity Lab. From 2012 to 2014he led imec’s 79 GHz radar research program.Since January 2015 Wim is responsible forimec’s R&D in cellular and WiFi transceivers, 60GHz communications, 79 GHz radar and 140GHz sensors.

Biography Erik Mannens

Erik Mannens is a professor at imec – UGent –IDLab, a research group conductingfundamental and applied research on Internettechnology and Data Science. Since 2008, he ispaving the Open Data path in Flanders. Hestood at the cradle of the first hackatons and isa founding member of the Open KnowledgeFoundation (Belgian Chapter). He is frequentlyinvited as Open Data evangelist at national andinternational events and actively participates inW3C’s eGov and Data On The Web workinggroups.

37/57

Introduction

“The essence of a smart city is not that it is crammed full of new technology,” says Pieter Ballon,smart cities expert and director of imec-VUB-SMIT. “First and foremost it is a city in which thequality of living has been lifted to a new level, fulfilling the practical needs and expectations of thepeople who live there.” With Kathleen Philips, program director perceptive systems at imec andHolst Centre, he explains how imec is working with the city council and a number of committedcompanies to make the city of Antwerp a living lab that may become a benchmark for urbanenvironments worldwide.

Evangelize and build gateways

“The goal of our ‘City of Things’ project is to examine how a smart city can be developed as part ofa realistic framework – working in close collaboration with the residents and the city council,” saysPieter Ballon. “Over the past year we have laid the foundations for doing that, both in terms of theorganization and of the technology to support this smart city.”

A first requirement for the success of a complex project like this is to make sure all the stakeholdersare on the same page. “You need to evangelize a lot for that to happen,” says Pieter Ballon, “Youhave to make sure that everyone is given the same level of expertise so that at the end of the daythey feel the same about it. Part of that process involved my book, which was published in 2016 andis already in its fourth edition. So the topic is very much alive in Flanders. My aim was to enthusethe policymakers, while at the same time ensuring that every city didn’t start crafting its ownsolution. Because when that happens, everything gets fragmented, with systems that don’t talk toeach other and cities that spend years tied to certain solutions and suppliers.”

Edition | January 2017

Smart Cities

Vision: A living lab for smartcities worldwide

38/57

Once the project is up and running, 100 gateways will have been put in place across Antwerp. Thesewill literally be the digital entry gates to the city and its inhabitants, complete with supportingtechnology. Gateways through which tens of thousands of wireless sensors, worn by the locals orattached to vehicles, the traffic infrastructure or buildings, can send their data to a whole range ofapplications. At the moment, we already have around 20 of these gateways in operation. The restwill be installed gradually. Everyone will be able to use them to develop or evaluate a specific smartapplication. The technology is heterogeneous, not based on a single solution or protocol, but on aseries of standards that are suited to these applications, such as Zigbee, WiFi, Cellulair, LoRa, SigFoxand others.

Senses for the city

“Imec specializes in making environments smart, so we have also incorporated some of thetechnology that we have developed,” says Kathleen Philips. “For instance, there are currently twoBelgian Post Office vans going about their business every day around Antwerp equipped with ourwireless multi-parameter sensors. These can now take readings for around 20 parameters, such asCO2 and NO2. And soon they will be able to measure fine pollution particles. The aim is to arrive atmore refined measurements, with hundreds of sensors installed all over the city, aboard a largernumber of vehicles driving around. These readings can be linked to weather forecasts andknowledge of the local terrain and mobility. Eventually, they should make for a very detailed map ofthe city indicating e.g. air quality down to the level of street corners.”

“Other technologies that fit in with this scenario are our new ion sensor and 60 GHz wirelessbackhaul. The first of these, the ion sensors, have already been installed at a water pollutionsampling station in the river Scheldt. The 60 GHz solution for backhaul networks will be developedto expand the available communication bandwidth quickly and seamlessly when major events arebeing staged or emergencies occur.”

The sensors and gateways are only a first layer, the senses of the smart city. Behind it all is aninfrastructure for processing and analyzing big data in real-time. Like the gateways, thisinfrastructure is also technology-neutral and offers everyone who wants to use it a set of tools forconverting data into useful knowledge. Kathleen Philips: “Take the sensor data relating to air quality.One of the scenarios we are considering for 2017 is an app for cyclists and pedestrians indicating thehealthiest and safest route to a particular location.”

Changing behavior

All of this infrastructure will only form the foundation of the new digital city. To make it reallysmarter and more pleasant, the inhabitants will have to be involved. That is why the project will berecruiting a large number of volunteers from 2017 on. These will be put to good use, taking part intest panels that evaluate and gradually improve the new applications at every stage of theirdevelopment – from initial idea to prototype. That way the city will become a genuine living lab.

39/57

Pieter Ballon: “In 2017 we intend to further expand this ‘Living Lab’ approach. Until now we havebeen asking people how they use a particular application and what they think about it. And we askthem whether they would be prepared to pay for applications (and how much). But now we wantto test whether – and how – we can influence people’s behavior using applications. How can asmart app encourage people to behave more sustainably or more healthily? And how can it do thatwithout forcing them or taking some of their comfort away? This type of behavioral change is thebasis for many new business or governance models: the so-called outcome-based economy. It willalso be one of the big challenges facing us in the years ahead.”

In the meantime, the government has already taken up some of the ideas and concepts of ourexperts. In 2017 one of the tasks given to imec is to work on the standards for open data, as well ason the first pilot projects for a smart Flanders. “This does not mean that all cities are required toimplement the same policy,” says Pieter Ballon. “It simply means that everyone is given a number oftools that support their own policies and areas of emphasis in a smarter way.”

Living lab for the world

There’s an international component here, too: imec has been selected to take part in the European‘Horizon 2020’ project called SynchroniCity. As part of the project, Antwerp will become one of theEuropean reference zones, along with 6 other European cities. The aim is to establish large-scalepilot projects featuring experiments with new IoT services, based on the real needs of citizens. TheAntwerp reference zone will focus mainly on mobility and logistics.“We have set up this City of Things project from within iMinds, where we had a great deal ofexpertise in digital technologies, as well as in setting up large-scale living labs,” says Pieter Ballon.“But now that we are part of the major global player that is imec, the whole story can take on a fargreater dynamic. We were a pioneer in this area in Flanders, working to harmonize things from anorganizational and technological point of view. But it was still on a fairly small scale – especially ifwe look at the challenges facing large conurbations worldwide. Now, with imec, our approach canattract an international following. That is one of the breakthroughs – plus the fact that we can alsocall on truly innovative solutions for both the software and hardware layers of our solution.”

“All around the world there are many smart city initiatives going on and the challenge there is todifferentiate yourself,” adds Kathleen Philips. “Yet in the past year in discussions with partners,people have said that they are very interested in our particular approach. This has to do mainly withthe breadth of our approach, with us working on open technological innovation in a living lab, whileat the same time also on applications, scenarios, business models, social organization andawareness. We are the only ones doing that for the time being.”

40/57

Biography Pieter Ballon

Pieter Ballon is a director of imec-VUB-SMITand professor at VUB Brussels (Belgium). Pieterspecializes in living lab research, businessmodelling, open innovation and the mobiletelecommunications industry. He has beeninvolved in numerous local and internationalR&D projects in this field, and has publishedwidely on these topics. Currently, he isinternational Secretary of ENoLL, the EuropeanNetwork of Living Labs. Pieter holds a PhD inCommunication Sciences and a MA in ModernHistory.

Biography Kathleen Philips

Kathleen Philips is program director at imecand Holst Centre for infrastructure and person-centric focused Perceptive Systems. Kathleenjoined imec in 2007 and has held positions asprincipal scientist, program manager for ULPWireless and program director for PerceptiveSystems. Before that, she was a researchscientist at Philips Research for over 12 years.She holds a PhD in electrical engineering, hasauthored and co-authored over 60 papers andholds various patents.

41/57

Introduction

For An Jacobs and Andy Lambrechts, the Fourth Industrial Revolution is underway. And not onlythat: they are right in the thick of it, guiding several imec customers and industry partners day in,day out, throughout their smart innovation projects. Professor An Jacobs provides coaching on thebasis of her expertise in the adoption of new technologies in the workplace; Andy Lambrechts andhis team are the brains behind smart, integrated vision applications that are pouring into themanufacturing industry at full force. Together, they reflect on the concept of “smart industries” andon what imec has to offer in this area.

From collaboration between humans and machines to customized visionapplications

“One of the most important topics we worked on in 2016 was the collaboration between humansand machines and the question of how we can make industrial labor more ergonomic, morepleasant,” An Jacobs begins. “And actually that is an issue that can be examined within a broadercontext, specifically that of bringing industrial production closer to home again. After all, one of thecharacteristics of the evolution towards a smarter industry - what we refer to as ‘Industry 4.0’ - isthe need for personalized, customized products. By committing to an optimized collaborationbetween humans and machines - to be able to meet the demand for unique products with a highdegree of efficiency and quality - we aim to strengthen local industrial activity.”

Edition | January 2017

Smart Industries

Vision: The Fourth IndustrialRevolution is underway

42/57

“On the other hand, we are also wholeheartedly committed to vision applications for qualityinspection and detailed process monitoring, where our technology reveals far more detail than wewould be able to detect with the naked eye. Over the past few months, the greatest challenge inthat area has been to make the technologies launched by imec in recent years more robust. Thatwas a response to specific demands from our customers. The fact that we can help them all theway from the introduction of technology through to its implementation in an industrialenvironment is a truly satisfactory experience,” Andy Lambrechts adds. “A good example of this canbe found in microscopy, where we can identify different types of white blood cells in a verycompact and scalable way. But we can also provide solutions to facilitate follow-up treatment forburn injuries, for example, or for monitoring and inspecting crops and food with hyperspectralcameras. In all these areas, we help our customers solve their industrial or medical challenges withaffordable equipment - and we do it really quickly, too.”

Incidentally, helping customers with concrete solutions - rather than developing a prototype - hasturned out to be an important requirement within the context of smart industries.

Andy Lambrechts tells us why. “This is why we decided a couple of years ago to produce - in limitedseries - the products our customers (often in high-value markets) require. Simply because there is noone else they can turn to. In addition to our technological expertise, we add to the mix ourexperience with materials, sensors, optics and software. A scope as broad as this really is unique.We have consciously invested in this in recent years, precisely because they are looking forsolutions that work and are easy to pick up. And we have clearly reaped the rewards of thatstrategy in 2016, which translates into customers who keep coming back to us with new requests.Personally, I believe that is one of the most important achievements of the past year.”

According to An Jacobs and Andy Lambrechts, the merger of imec and iMinds, which wascompleted in 2016, is excellent news for imec’s partners and customers as well.

“Customers in the medical sector, for example, are not just in search of equipment to identify andsort different types of cells; solutions for this market also need to take security and privacyrequirements into account. Customers active in the food industry, on the other hand, may benefitfrom the combination of our hyperspectral cameras and the data mining expertise that the mergerwith iMinds has brought on board,” Andy Lambrechts explains.

“If you want to take the step - as Andy just said - from prototypes to concrete products, the livinglab methodology developed by iMinds might also be very valuable. That methodology can help toaccurately define the requirements of a given technology, check certain assumptions in due timeand produce a product that is genuinely relevant – both locally and internationally,” An Jacobs adds.“This is yet another area in which the broadened imec now occupies a unique position. A goodexample is the imec.icon ‘ClaXon’ project, which involves Audi Brussels and a number of otherpartners. In this project, we investigated the interaction and collaboration between humans andwhat we call ‘cobots’ (collaborative robots) in the manufacturing industry. In the course of theClaXon project as well, prototypes were used several times – so that their effectiveness couldrapidly be assessed.”

43/57

The broadened imec covers the entire smart industries value chain

“Obviously, in the next few years, we are going to continue building on what we have achieved in2016. In terms of collaboration between humans and machines, for example, numerous possibilitiesin health care can be explored,” An Jacobs says. “At present, this sector primarily makes use of‘passive’ robots that require constant control. But we are going to change that. By adding artificialintelligence, we can teach robots to take ‘the unexpected’ into account. Interaction with people -who are unpredictable by nature - is a good example of this. And we can also use this knowledge toboost further development in production environments, where we want to make robots respondmore flexibly to new situations, such as a new model of car, for example, that has to roll off theproduction lines. Fact is that reprogramming production lines is an expensive, time-consuming taskat present - which will in many cases result in an entire factory coming to a standstill. We are goingto take significant steps in tackling all these challenges in the next few years. Safety will also be afocal point in this.”

“In addition, we are clearly going to continue helping our customers and partners to resolvedifficult problems at systems level. The interesting thing about that is that it allows us to apply ourinnovations in markets where you wouldn’t initially expect this, and where our sensors are part of afar greater whole,” adds Andy Lambrechts.

Finally, An Jacobs and Andy Lambrechts agree that the synergy between hardware and software aswell will continue to be exploited.

“In the past, covering the entire value chain in projects was often a problem - but that will becomea lot easier for the broadened imec. This is also important for all our customers and partners: afterall, decisions constantly have to be made in terms of both hardware and software; you can nolonger consider them separately,” An Jacobs points out.

“For example, each market demands specific algorithms to deal with sensor data. Our sensor groupsdo not have the knowledge to build this up for each application separately. But the former iMindsgroups do have that knowledge. It will enable us to fulfill our customers’ wishes even more end-to-end,” Andy Lambrechts concludes.

44/57

Biography Professor AnJacobs

Professor An Jacobs co-ordinates the digitalhealth projects within imec - VUB - SMIT. Shehas expertise on the adoption andappropriation of new technologies from aprofessional and an end-user point of view. Atthe University of Brussels, An teachesqualitative research methods inCommunication Sciences. At iMinds (now partof imec) she has coordinated and participatedin various European and Flemish researchprojects and work packages with a focus ondigital technologies development, use andappropriation in health and wellbeing.

Biography Andy Lambrechts

Andy Lambrechts received a MSc and PhD.degree in electrical engineering from LeuvenUniversity. He is currently leading imec’sIntegrated Imaging team and is working ontechnologies such as hyperspectral imaging andlens-free microscopy. Andy is also involved in avariety of other activities that combine imec’sprocess technology with systems and softwareknowledge to enable new applications in thevision domain.

45/57

Introduction

Jef Poortmans and Matthias Strobbe are creating the new energy network. Poortmans is producing anew generation of batteries and solar cells that are cheaper and more efficient and which operateperfectly in everyday circumstances. Meanwhile, Strobbe is developing smart algorithms that alignenergy supply and demand, and wireless technology for connecting devices. Both Poortmans andStrobbe agree that to achieve a true sustainable energy system they need more than justtechnological innovation. It is just as important to ensure the various partners on the energy marketare working together, as well as to put the appropriate regulations in place and to develop market-based models that deliver a proper ROI for the various players involved.Solar cells that not only feel good in the lab, but also on your roof!

To a large extent, the new energy network will be based on renewable, distributed sources ofenergy, such as solar, wind and water. Jef Poortmans is scientific director photovoltaics (PV) at imecand knows better than anyone what is at stake in the research into solar energy: “Up until five yearsago, our researchers were working mainly on ways of reducing the initial investment cost ofphotovoltaic solar energy, expressed in euro per Wp. And their work (and that of others) was anunprecedented success: thanks to technological innovations, the right incentives and the upscalingof the industry as a whole, the cost of photovoltaic cells fell by a factor of five in the space of tenyears. But where previously the focus was on reducing investment costs, attention has nowswitched to reducing the effective electricity cost of a PV system in terms of eurocents per kWh.”

Edition | January 2017

Smart Energy

Vision: Renewable energyand ICT for a sustainableenergy system

46/57

‘Kilowatt-peak’ refers to the power that the panels produce under standard conditions; ‘kilowatt-hour’ refers to the actual power of the solar panels under real-life circumstances. Maybe a treecasts shadow over your roof at certain hours of the day, the sky is partly cloudy, or you live in a hotand humid climate. Poortmans: “We are now heading towards a situation where solar cells are beingdeveloped to suit a specific type of climate: solar cells for deserts, the tropics, temperate climatesand so on. Smart solar cells that are also able to work well in changing weather types and that caneven know in advance how the weather is likely to change. All of this will enable us to push theprice per kWh down even further.”

“But the greatest challenge is definitely increasing the value of the system. This is something we areworking towards in close collaboration with our EnergyVille partners,” says Poortmans. “The extentof your solar energy value is defined not only by your own system, but also by how it is integratedand applied within the energy network. For example, if you are producing at a time when there is asurplus of electricity available, then your PV energy does not have much value. When that happens,you need to work with batteries or other solutions to gear the production peak to demand. This isnot purely a technical problem and can be a challenge when there are a lot of different partiesinvolved. Hence also the high level of complexity.”

“Last year we achieved very good results with batteries. Our researchers developed a solidelectrolyte for Li-ion batteries with a record ion conductibility of 15mS/cm. These batteries are asafer alternative to today’s batteries with their liquid electrolyte.”

Poortmans also emphasizes that esthetics are very important: “Solar cells need to form a seamlesspart of a building by being integrated into the roof or into big glass walls.”

Virtual batteries

Matthias Strobbe knows just about everything about the ‘seamless integration of PV’ – if not inarchitectural terms then certainly in relation to the electricity grid: “Our research team, underProfessor Chris Develder, is busy developing smart algorithms and communication technology forthe new energy network. Our aim is to be able to integrate renewable energy sources seamlesslyinto the network. The problem is that the production of energy based on sun, wind and water isvariable and does not always correspond to demand. You can, as Jef says, work with batteries forstoring energy, but you can also use ‘virtual’ batteries. For example you can use a boiler as a thermalbuffer, or you can recharge an electric car when there’s a surplus of energy available and you canadjust the temperature of a cold store one degree higher if there is a shortage of energy, and so on.”In other words: demand side management is equally important.

47/57

How flexible is your energy usage?

An important condition for the scenario above to work is that you have a clear picture of energyusage and especially of the level of flexibility in place to delay or bring forward this energyconsumption where required. Strobbe: “In past years we have gathered a mass of data about energyusage in homes and business, as part of the LINEAR project. Based on this we have set up statisticalmodels that give us a clear picture of the consumption of equipment and appliances and the timeat which that usage takes place. These models serve as input for our smart algorithms, which aredesigned to harmonize energy production and energy usage.”

“Last year we also worked on technology for wireless communication between sensors. Sensors areessential in the energy network of the future for enabling appliances and machines to communicatewith one another, as well as with the network. This includes things such as ventilation, heating andsun protection. These are all systems that need to be able to ‘talk’ to each other to create anenergy-efficient building. Our research group is working on standardization and making wirelesstechnology more economical.”

Solar cells with two sides

A great deal also happened in 2016 in the area of imec’s solar cell research. Poortmans: “A veryimportant trend in PV is the development of two-faced or bifacial solar cells. These are cells thatare capable of collecting light on both their upper side and lower side and converting it toelectricity. This also enables the light reflected from the ground to be put to good use and henceimprove the energy output. These solar panels can also be placed upright, for example in solarpower stations in the desert. This gives you a much flatter production profile and as an additionalbonus, the panels don’t collect so much sand on their surface, which is a common problem with thesolar panels currently installed in deserts. Our researchers have produced a bifacial solar cell withefficiency in excess of 22%. They have also developed a new plating technique for metalizing thefront and back of the solar cell. To put in the correct perspective, with typical reflection, thiscorresponds to an ‘effective efficiency’ of 26%.”

“Another new kid on the block is the perovskite cell. Last year, we achieved 19% efficiency withthese. In terms of perovskite modules, a number of world-best results have also been achieved –and certainly not forgetting that we have now produced semi-transparent PV modules.Transparency is important for integrating the modules in glazed façades, or if you want toincorporate the perovskite cell above a silicon solar cell. In this latter case, efficiency levels of up to30% and more can be achieved (at exposure level 1 sun).”

48/57

A sunny future?

The merger of imec and iMinds is definitely a positive step forward for the new energy network andwill certainly boost the EnergyVille offering. Poortmans: “Our areas of expertise complement eachother perfectly. Technically speaking, I believe that we can achieve a smart energy network by 2030,although good regulations will be essential to cover all interests and make sure we are all heading inthe right direction. For instance, this is to avoid situations where electricity users wanting to be self-sufficient with solar cells and batteries find themselves having to pay a ‘disconnection’ fee to theenergy provider.”

Strobbe: “There are indeed a great many areas of expertise required to achieve a smart energynetwork. One of these areas is security, which is a field in which imec - KU Leuven - COSIC excel. Ifyou are going to connect all appliances with each other and also with the network, they could alsobe hacked. And as data will be collected about energy usage, there are all sorts of things thathackers could deduce about the day-to-day ins and outs of families and businesses. So the smartenergy network of the future must be secure, respect people’s privacy and also be reliable. It’s OKfor a multimedia application to drop out, but you wouldn’t tolerate it if that happened to yourelectricity network.”

Biography Jozef Poortmans

Jozef Poortmans received his degree inelectronic engineering from the CatholicUniversity of Leuven, Belgium, in 1985. Hejoined imec and worked on laserrecrystallization of polysilicon and a-Si for SOIapplications and thin-film transistors. In 1988 hebegan working on his PhD on strained SiGelayers. He was awarded his PhD in June 1993.Afterwards he joined the photovoltaics group,where he became responsible for the advancedsolar cell group. Within this framework hestarted the thin-film crystalline Si solar cellsactivity at imec. He also coordinated severalEuropean Projects in this area as part of the 4thand 5th European Framework Programs. In 2003he became cluster coordinator of Europeanprojects in the field of thin-film solar cells. In1998 he initiated the organic solar cells activityat imec, which was complemented by anactivity dealing with III-V solar cells, whichbegan in 2000. At the present time, Jozef isscientific director photovoltaics at imec.

49/57

Biography Matthias Strobbe

Matthias Strobbe received his M.Sc. degree andPh.D. in Computer Science Engineering fromGhent University in July 2004 and June 2011respectively. He was affiliated with the IDLabresearch group at Ghent University and imec(formerly iMinds) from 2004. He has worked assenior researcher and project coordinator inthe field of smart energy grids since 2011. Hejointly coordinates research projects in theteams of professors Chris Develder and IngridMoerman at imec – UGent – IDLab. In 2016Matthias also became an IoT businessdeveloper for smart energy, smart buildings anddigital manufacturing.

50/57

Introduction

In Flanders a lot of research is carried out into new high-tech technologies – often with impressiveresults. But only a small portion of these innovations filter through to the market, for example asnew products or upgrades. Roger Lemmens, director of imec’s incubation and innovation services,explains how imec is helping to bridge this gap, with more support for the whole innovationprocess and with a comprehensive offering of software and hardware expertise from 2017 on.

Over the years, iMinds (now incorporated into imec) has developed a number of services to helpyoung start-up companies bring new digital technology to the market. And also imec proper had arange of services to help entrepreneurs, albeit more focused on developing hardware. Now, afterthe merger of the two research centers in October 2016, the new imec has the highest level ofexpertise in both software and hardware. And thanks to its 3,500 researchers, it also has significantleverage in fields such as healthcare and the IoT, as well as in smart cities, mobility and logistics.

“In 2016, the iStart operation (now imec.istart) was responsible for launching 20 new digital start-ups,” says Roger Lemmens. “That brings the total number of companies launched and supported byiStart to well over 100 since the program began in 2011. In 2016 alone, our portfolio of start-ups hasraised almost 20 million euro in follow-up funding. This demonstrates that our approach is able togenerate a great deal of value.”

Edition | January 2017

imec.cityofthings

Vision: Imec digs into thegoldmine of Flemish high-tech

51/57

Also ICON (now imec.icon) – the support for demand-driven, cooperative research involvingmultidisciplinary teams from companies and universities – again saw a good number of successfulprojects. One example is iFest, a project aimed at developing a new generation of festival braceletsto provide a more enriching festival experience, based on built-in communication and sensorfunctionality.

Another highlight of 2016 was the launch of the City of Things project in Antwerp, an ambitiousliving lab relating to the organization and technological set-up of a smart city environment for thefuture.

The Leuven innovation team for Flanders made a major contribution to a number of sectorinitiatives. One example is the Flemish food industry’s i-FAST platform. i-FAST brings innovativetechnologies and analysis methods to the companies involved in food processing. Through workingwith a sector, imec is able to bring innovation to the attention of whole groups of companies at atime, companies that are often not aware of the rapid developments in high-tech. The Leuven-based group also organized a range of workshops for the industry and provided practical support toa number of Flemish SMEs to help prepare their products for market. These included the start-upsAirobot (a module for measuring distances in drones) and Sensolus (a GPS tracker), as well asHydroko, which is developing a smart water meter solution that e.g. will be rolled out in Antwerpby 2020.

“What we have been doing until now with these programs was to provide a helping hand, withresearch and business expertise as well as with financial assistance where applicable,” saysLemmens. “We now intend to expand this approach, helping companies through the entireinnovation process, from the idea and application prototypes right through to a successful product.We call this our digital incubation and innovation services (DIS). Central to this is the ‘Living Lab’methodology. In short, this involves calling on panels of end-users or professionals at every stage ofthe innovation process. These provide feedback that is used to improve the ideas, plans orprototypes in a continuous cycle of co-creation. Such an interactive development results inproducts that better meet the needs and expectations of users – and so hopefully gain a quickerfoothold in the market.”

In 2017, DIS will be rolled out as an integrated service to Flemish and international industry partners.Roger Lemmens: “In the new, extended imec, we can now call upon all departments, such as imecIC-link or imec Taiwan, to offer added value for DIS.”

Central to the new approach from March 2017 onwards will be the ‘De Krook’ site in Ghent. Thisultramodern building is equipped for the digital future. It includes a data wall and areas where userscan design and test technology in various stages of development. It also has around 250 intelligentsensors spread across the site. These send their data to an open platform that is similar to the Cityof Things platform that imec is rolling out in Antwerp. De Krook is designed to be the physicalshowcase for the digital incubation and innovation services, enabling imec to work closely withstrategic partners such as UGent and industrial players involved with digital innovation.

52/57

Another leading light is the City of Things project in Antwerp. Roger Lemmens: “Antwerp is ourbiggest Living Lab, with an open data network in which businesses at all stages of the innovationprocess can experiment and validate their solutions by e.g. making a structured use of the expertiseof our user panels. That City of Things project will become our Smart City flagship – an example ofwhat we could also realize in other cities or environments.”

“Finally in 2017,” continues Lemmens, “we offer two complementary services for start-ups.Imec.istart will focus on Flemish high-tech start-ups that will be supported by experiencedentrepreneurs and experts towards growth and an international breakthrough. Imec.xpand, oursecond initiative, is designed to support more capital-intensive start-ups: initiatives that will takeimec’s knowhow and technology to market, with products that use the unique power ofnanotechnology and miniaturization. More than in the past, we will focus on acceleration, ongrowing quickly. This will involve more vertical support, with services and experts specializing in aspecific fields such as healthcare.”

Biography Roger Lemmens

Roger Lemmens is imec’s director of incubationand innovation services. He studied CivilEngineering Mathematics at TU Eindhoven andobtained an MBA at the Vlerick BusinessSchool. Roger has built his career at varioussoftware companies, such as Real Dolmen,Pietercil (Netherlands) and NMC (Germany). Sixyears before joining iMinds, Roger was CEO ofxperthis, a Belgian company with almost 200staff, developing, implementing and supportingsoftware for the healthcare sector.

53/57

Introduction

Imec is not conducting research in an ivory tower: our research agenda is – to a large extent – beingdetermined by your requirements, i.e. the needs and innovation opportunities identified by ourpartners and customers. To address those needs, imec has a number of instruments at its disposal –such as our prototyping, product development and production capabilities, through which we cantranslate innovative ideas into market-ready solutions. Another example includes the imec.IC-linkoffering, which supports companies with the design, production and validation of custom-madechipsets – even in small quantities.

Importantly, these services can be accessed by a wide variety of customers – from start-ups, tosmall and medium-sized companies as well as multinationals. Yet, a prerequisite to provide thesedifferent audiences with the best possible guidance and support is the availability of good insightsinto their needs and concerns. That’s why we sat together with Max Mirgoli, Executive Vice-President Worldwide Strategic Partnerships at imec, to find out what has been keeping you awakeat night in 2016 and how those challenges translate into short and mid-term opportunities;opportunities that we can address together.

Edition | January 2017

General

Vision: Helping companies –small and big – translateinnovative ideas intomarket-ready solutions

54/57

Feedback from the field

“When talking to customers in the last couple of months, three key topics kept coming back,”observes Max Mirgoli. “The first set of questions was voiced by our partners and customers in thechip making industry. They have become used to a cadence of two-year refreshment cycles inwhich semiconductors get increasingly powerful while realizing a better power budget. But thatcadence has started to slow down for cost reasons. Hence, they are eager to discover the nextgeneration of technology options, and get more info on when those will become available. Andthey’re clearly looking at imec to come up with the answers.”

“Secondly, in the smart systems domain interesting things are happening as well,” he adds. “Look atwhat is ongoing at the crossroads of the life sciences and the semiconductor industries, where chiptechnology has become fundamental to enabling the next generation of smart life scienceapplications. And the same is happening across all application areas that make use of Internet ofThings technology – whether in agriculture, robotics, etc. All players involved are trying tounderstand what will be the impact of this – in terms of semiconductor feature requirements andshipment volumes.”

According to Max Mirgoli, imec is instrumental to driving innovation in this space. A recent exampleof how imec is making a difference here, includes last year’s launch of miDiagnostics – an imec andJohns Hopkins University collaborative effort that aims at a more efficient healthcare. Leveragingthe partners’ respective expertise in nanoelectronics and medicine, miDiagnostics develops labs-on-chip that require only drops of blood for the on-chip detection of cells, proteins, nucleic acids,and/or small molecules.

“A final question I received on a regular basis, relates to the potential of combining hardware andsoftware expertise,” Max Mirgoli continues. “In essence, that’s not surprising: after all, it is thecombination of hardware and software that has led to some of the most significant technologicalbreakthroughs. Yet, in practice we see that hardware and software are still largely being decoupled– both in research and the industry. But again, in 2016, imec has been able to address thisopportunity – thanks to the merger with digital research center iMinds. And mark my words: this isa marriage made in heaven! Combining the expertise of iMinds in domains such as cryptography,data mining, data analytics and data connectivity with our world-leading position innanoelectronics will enable us to dramatically increase the pace at which we – and our customers –can innovate.”

Taking the next steps together – in 2017 and beyond

So, building on this knowledge, how can you call upon imec in 2017 to take the next steps –whether that means translating innovative ideas into market-ready solutions, or reinventing yourexisting product portfolio?

55/57

“It is important to recognize that innovation is no longer the prerogative of big, establishedcompanies. Increasingly, start-ups and SMEs have an important role to play here as well. And thegood thing is: imec’s innovation services can serve both worlds,” claims Max Mirgoli.

“Obviously, companies in the chip making industry can continue to rely on us as their reliable,strategic partner to help mitigate the risks of redesigns and other mistakes as they investigate theuse of new materials and new device concepts or the potential of future technologies, such asextreme ultraviolet (EUV) lithography.”

Secondly, customers and partners can rely on imec’s expertise in coming up with new, miniaturizedand different technology pieces – and integrating those. The colleagues from Holst Centre (set upby imec and TNO), for instance, are helping companies in the digital health realm – from concept todevice. One example includes a health patch that was launched in November 2016 – the first of itskind to track physical and cardiac activity, while monitoring bioelectrical impedance. The healthpatch integrates unique technologies and components from imec and its industrial partners.

Another example of how imec speeds up product development, thereby tuning products toconsumers’ needs, can be found in imec’s City of Things testbed in Antwerp (Belgium). Building onimec’s living lab methodology, the project leverages interactive user research (such as panelfeedback and co-creation) and innovative business modelling within a real-life, large-scale setting todevelop, test and scale up digital innovations.

“Leveraging that knowledge and those innovation capabilities is something that SMEs and start-upsare highly interested in,” concludes Max Mirgoli. “Moreover, many companies have great ideas for innovative solutions in a wide range of domains –but what they usually lack, are the devices on which to run these solutions. That’s where we comein as well. Through our imec.IC-link activities, for instance, via which we help start-ups and SMEsacquire the specific, low-volume chipsets they require – at a fraction of the mass cost. Or throughimec Florida, which focuses on integrated circuit (IC) design research for a broad set ofsemiconductor-based solutions such sensors and imagers, and via which we can now also providethe US market with low-cost access to advanced foundry services.”

“And let’s not forget our system prototyping and low-volume offering from imec Taiwan, whichworks together with today’s visionaries to turn their great ideas into valuable, bonafide products.One example in this domain includes the development of an innovative bathroom mirror thatallows people to see themselves in front and rear view simultaneously; an innovation developed in2016 in collaboration with two leading Belgian design companies.”“In 2017 and beyond, our partners and customers can continue to count on those assets to take thenext steps – together with imec.”

56/57

Biography Max Mirgoli

Max Mirgoli is the Executive Vice-President ofWorldwide Strategic Partnerships at imec. Hehas over 26 years of experience in thesemiconductor industry, has held variousleadership roles in the industry, and has an in-depth knowledge of the semiconductorindustry’s dynamics.Prior to joining imec, Mr. Mirgoli was the COOand Managing Director of ICOS Vision Systems.Before that, he served as the CorporateExecutive Vice-President and General Managerof the Automation and Controls Division ofMatsushita Electric Works (Panasonic). Mr. Mirgoli holds an MBA from Notre DameCollege and a Master’s of Science Degree inElectrical Engineering.

57/57