1
Guardian Angels
FET Flagship Pilot
Final Report, April 2012
Public Version
Coordinators:
Ecole Polytechnique Fédérale de Lausanne (Switzerland), ETH Zurich (Switzerland)
Partners:
Aridhia Informatics Ltd., Edinburgh (UK), Centre National de la Recherche Scientifique CNRS (France), Centre Technologic de Telecomunicacions de Catalunya (Spain), CNM-CSIC, Barcelona (Spain), CNR, Il Consiglio Nazionale delle Ricerche, Modena (Italy), Commissariat à l’Energie Atomique et aux Energies Alternatives CEA (France), Consorzio Nationale Interuniversitario per la Nanoelettronica (IUNET, Italy), Newcastle University (UK), CSEM Centre Suisse d’Electronique et de Microtechnique SA – Recherche et Développement (Switzerland), Danmarks Tekniske Universitet (Denmark), Forschungszentrum Jülich (Germany), Fraunhofer
‐Allianz Ambient Assisted Living and Fraunhofer-Einrichtung für Modulare Festkörper-Technologien (Germany), GreenTEG GmbH (Switzerland), Grenoble-INP (France), INRIA (France), HiQScreen SARL (Switzerland), IBM Research GmbH (Switzerland), Infineon Technologies AG (Germany), Imperial College London (UK), Institut Català de Nanotecnologia (Spain), Instytut Technologii Elektronowej ITE (Poland), Intel Performance Learning Solutions Limited (Ireland), Interuniversitair Micro-Electronica Centrum VZW (Belgium), Katholieke Universiteit Leuven (Belgium), Lund University (Sweden), National Technical University of Athens (Greece), Nestlé Research Center (Switzerland), NXP Semiconductors Netherlands BV (Netherlands), Philips Eindhoven (Netherlands), Politecnico di Milano (Italy), PSA Peugeot Citroën (France), Rheinisch Westfälische Technische Hochschule (Germany) Aachen University (Germany), Royal Institute of Technology KTH (Sweden), Sanofi Aventis Recherche & Développement (France), Senarclens, Leu & Partner AG (Switzerland), Siemens AG (Germany), Slovenskà technickà univerzita v bratislave (fakulta Elektrotechniky a informatiky) (Slovak Republic), Stichting IMEC Nederland (Netherlands), STMicroelectronics Crolles 2 SAS (France), Technische Universität München (Germany), Technische Universität Eindhoven (Netherlands), Thales SA (France), the SINANO Institute (France), Tyndall National Institute University College Cork (Ireland), Universitat Autònoma de Barcelona (UAB) and Universitat Politècnica de Catalunya (UPC) BarcelonaTech (Spain), Université Catholique de Louvain (Belgium), University of Cambridge (UK), Universidad Complutense, Madrid (Spain), University of Glasgow (UK), University of Liverpool (UK), University Politehnica of Bucharest Institute for Microtechnologies (Romania), Universität Siegen (Germany), University of Southampton (UK), Universiteit Twente (Netherlands), VTT Technical Research Centre of Finland (Finland)
Table of Contents
1 Guardian Angels Summaries ............................................................................................. 4
1.1 Executive Summary .................................................................................................................... 4
1.2 Summary for EU Decision Makers ............................................................................................ 5
1.3 Summary for the General Public ............................................................................................... 6
2 S & T Quality ...................................................................................................................... 9
2.1 Scientific Vision and Unifying Goal ........................................................................................... 9
2.1.1 The big picture: Guardian Angels at the core of the information revolution ......................................... 9
2.1.2 The Vision and Benefits of GA ............................................................................................................. 9
2.1.3 Unifying goal: the science of zero power ............................................................................................ 10
2.1.4 Overview of main objectives: enabling GA autonomous personal assistants ...................................... 11
2.2 Main Scientific Objectives ........................................................................................................ 17
2.2.1 Grand scientific challenges for energy-efficient computation ............................................................. 17
2.2.2 Grand scientific challenges for communications ................................................................................. 18
2.2.3 Grand scientific challenges for low-power sensing ............................................................................. 19
2.2.4 Grand scientific challenges for energy harvesting, storage and power management .......................... 20
2.2.5 Grand scientific challenges for human-machine interfaces ................................................................. 21
2.2.6 Grand scientific challenges for embedded software for low-power computing .................................. 22
2.3 Matching of the GA Proposal with the Flagship Concept ..................................................... 23
2.4 Methodology.............................................................................................................................. 24
2.5 Guardian Angels Demonstrator and Roadmap Executive Summaries ................................ 25
2.6 Work Plan: ramp-up phase and long term ............................................................................. 25
2.7 Resources to Implement the Roadmaps .................................................................................. 25
2.8 Metrics to Measure Progress in the Flagship .......................................................................... 26
2.8.1 Progress monitoring ............................................................................................................................ 26
2.8.2 Quality Monitoring: Quality Assurance Procedure (QAP) .................................................................. 26
2.9 Coordination of Activities and Research Communities ......................................................... 27
3 Implementation ................................................................................................................ 30
3.1 Governance and Scientific Leadership .................................................................................... 30
3.1.1 The Guardian Angels Consortium partners ......................................................................................... 30
3.1.2 Scientific and Technical Leadership and Expertise ............................................................................. 30
3.2 Management Plan ...................................................................................................................... 30
3.2.1 Consortium size and management challenges ..................................................................................... 30
3.2.2 Management structure versus project organisation ............................................................................. 31
3.2.3 Governance and bodies........................................................................................................................ 32
3.2.4 Openness and Flexibility: Open Calls ................................................................................................ 33
3.2.5 Risk Management ............................................................................................................................... 35
3.3 Resources to be Committed ...................................................................................................... 35
4 Impact ............................................................................................................................... 36
4.1 Transformational Impact on Science and Technology .......................................................... 38
4.1.1 Impact on electronic systems ............................................................................................................... 38
4.1.2 Impact on the scientific community .................................................................................................... 38
4.1.3 Availability of vast amounts of data .................................................................................................... 39
4.1.4 Impact of prioritizing energy-efficient technologies ........................................................................... 39
4.2 Impact on Society ...................................................................................................................... 39
4.2.1 Impact on Quality of Life .................................................................................................................... 39
4.2.2 Impact on Health Care ......................................................................................................................... 39
4.2.3 Impact on Human-Technology Interaction .......................................................................................... 42
4.2.4 Impact on Security and Safety ............................................................................................................. 43
4.3 Impact on the Environment ...................................................................................................... 44
4.4 Impact on the Economy ............................................................................................................ 45
4.5 The Impact Analysis Process in the Guardian Angels Flagship ........................................... 47
4.6 User experience groups ............................................................................................................. 48
4.7 Use of Results and Dissemination of Knowledge .................................................................... 49
4.7.1 Dissemination of the Guardian Angels Project ................................................................................... 49
4.7.2 Strategies for raising Societal Awareness............................................................................................ 50
4.7.3 Exploitation and Technology Transfer Plans ...................................................................................... 51
4.7.4 Plan for Using and Disseminating the Knowledge (PUDK) ............................................................... 52
4.7.5 Intellectual Property (IP) ..................................................................................................................... 52
4.8 Education and Training at the European Level ..................................................................... 52
4.9 Potential Ethical and Legal Implications ................................................................................ 53
4
1 Guardian Angels Summaries
1.1 Executive Summary
The "Guardian Angels for a Smarter Life" (GA) FET Flagship project will create intelligent, autonomous electronic personal companions that will assist us from infancy to old age. These devices will be private and secure systems featuring sensing, computation, and communication beyond human capabilities. Three families of life-enhancing demonstrators will offer instant availability of relevant information, interconnectivity between devices fitted with all sorts of sensors, and intentional and intuitive usability: (I)
Physical GAs will have the capacity to monitor the physical and/or physiological status of individuals in health care, rehabilitation, and sports, with an awareness of the context of activity of these individuals. With a strong focus on prevention and early diagnosis, these devices will help keep healthcare affordable and accessible to all. (II) Environmental GAs will observe ambient conditions for environmental threats and communicate with each other to expand their information base. In combination with functionality from Physical GAs, it will be possible to correlate a person’s physical state with the environmental context. (III) Emotional GAs will be able to perceive emotional or affective conditions, and will both support patients and enhance the performance of healthy people, such as with smart-drive assistants for improved safety. The ethical aspects of GA applications will be assessed from the beginning of the project, through interactions between researchers, an ethics board of experts and end users, and privacy and security will be the highest priorities.
The technology-related project goal is the exploration and development of "zero-power" technologies for these electronic personal assistants, so that they will harvest their own energy rather than requiring an external power source. This flagship proposal is driven by the fundamental scientific challenges related to achieving energy efficiency in complex systems, and will impact the development of Information and Communication Technology (ICT) future and emerging technologies through multidisciplinary research targeting long-term goals.
The zero-power requirement of the project has two sides: the acquisition of energy through harvesting, and the development of ultra-low power systems, whose energy consumption is as close as possible to theoretical limits. The devices will have the ability to scavenge energy in very diverse environments and store what is needed for the system functions. This will, however, require disruptive scientific progress in the field of novel concepts and technologies for energy harvesting; both for known types of energy harvesters such as solar, thermal, vibrational, and electromagnetic, as well as for new bio-inspired energy scavengers that are bio-chemical or synthetic photosynthesis-based. The target power density will be up to a few mW per cm
2 (or cm3), achieved by integration of existing and disruptive concepts. At the same time it is indispensable to minimize the energy consumption of the system. The scientific and technological challenges of the project thus include developing an ultra-low energy innovation chain: from materials and devices, to heterogeneous system integration, to software and communication techniques enabling the reduction of energy consumption by up to three orders of magnitude compared to existing state-of-the-art technologies. Compared to existing ENIAC, CATRENE and ICT initiatives, this FET Flagship will not be limited to segments of the nanoelectronics innovation chain, but will instead unify energy-efficient technologies and integrated energy scavengers within one long-term design and integration platform.
The objectives of the project are divided into three main categories: (i) scientific objectives centred on ultra-low power concepts, (ii) novel (nano)technology, materials and device integration solutions (based on advanced silicon platforms but including novel families of nanomaterials such as graphene, carbon nanotubes and/or organic semiconductors on flexible substrates), and (iii) three families of visionary zero-power autonomous systems demonstrators. These categories of objectives are interdependent and are designed to result in a 5
platform of ultra energy-efficient technologies for system integration, which will be applied in an unprecedented number of applications far beyond the GA idea.
As the application scenarios above suggest, the impact of Guardian Angels on society and the economy is very broad. On an individual level, anyone could benefit from a personal assistant acting as a modern day Guardian Angel, focusing on the improvement of health and on personal safety, with the goal of enhancing quality of life in a sustainable society. To achieve this, the GA consortium will work in close cooperation with different social stakeholders, interest groups and future users. Special attention will be paid to energy-efficient, economically feasible and environmentally friendly solutions. Additionally, the consortium will explore novel opportunities for human-machine interaction. Undoubtedly, in the course of the project, further beneficial applications using GA technology will be explored to make our environment more interconnected and smarter, more energy-efficient and safer.
The benefits of developing the enabling technologies for ultra-low energy consumption are widespread and lie beyond the immediate applications in smart systems. The novel electronic GA technologies developed in this project will contribute to the reduction of the energy consumption of systems: the estimated reduction may be worth more than 7% of the worldwide gross domestic product (GDP), as estimated by IBM analysts in 2010.
1 The project will strengthen the leading role of Europe in zero-power technologies by reinforcing the activity of manufacturing in Europe and improving the competitiveness for leading communication, equipment and tools companies and service providers. Overall, the project will result in the creation of new employment in Europe in many fields, including advanced Information and Communication Technology.
1 Dario Gil, presentation at
International Workshop on Future Information Processing Technologies (IWFIPT), 2010.
1.2 Summary for EU Decision Makers
The "Guardian Angels for a Smarter Life" (GA) FET Flagship project addresses challenges which are of utmost importance for Europe’s society, economy and environment and which will have to be tackled in the next 10-15 years.
Health and an ageing society, safety and security, and transportation, will all be addressed by the energy-efficient systems of the GA project. Imagine mobile electronic personal assistants being 1000 times more energy-efficient than they can be today, so that they could be powered by the energy available in your environment without any power plugs. Imagine part of the energy savings transformed into sensing, communications and interface functions of an invisible system that becomes your day-to-day Guardian Angel. This project will address the grand scientific, technological and system engineering challenges in a time frame of 10 years to transform this vision into a reality.
The GA FET Flagship project will develop technologies for extremely energy-efficient, smart, electronic personal companions that will assist humans from infancy to old age. These devices will be private and secure systems featuring sensing beyond human capabilities, computation, and communication. Three pre-defined families of demonstrators will show the feasibility and functionality of the systems: (I)
Physical GAs will have the capacity to monitor the physical and/or physiological status of individuals in health care, rehabilitation, and sports, with an awareness of the context of activity of these individuals. With a strong focus on prevention and early diagnosis, these devices will help keep healthcare affordable and accessible to all. (II) Environmental GAs will observe ambient conditions for environmental threats and communicate with each other to expand their information base. In combination with functionality from Physical GAs, it will be possible to correlate a person’s physical state with the environmental context. (III) Emotional GAs will be able to perceive emotional or affective conditions, and will both support patients and enhance the performance of healthy people, such as with smart-drive assistants for improved safety. The ethical aspects of GA applications will be assessed from the beginning of the project, through interactions between researchers, an ethics board of experts and end users, and privacy and security will be the highest priorities. 6
The contribution of the Guardian Angels FET Flagship on the European economic value chain is completely in line with what the European Commission expressed in its Communication of September 2009: "A significant part of the goods and services that will be available in the market in 2020 are as yet unknown, but the main driving force behind their development will be the deployment of key enabling technologies (KETs). Those nations and regions mastering these technologies will be at the forefront of managing the shift to a low carbon, knowledge-based economy, which is a precondition for ensuring welfare, prosperity and security of its citizens. Hence the deployment of KETs in the EU is not only of strategic importance but is indispensable."
In particular, the supporting zero-power technology platform
itself forms a KET (or can be considered a collection of future KETs working toward a unified goal) that integrates all the levels of the value chain needed to develop the GA autonomous systems. Each layer of the GA systems, from materials, to systems, to applications, has its own impact, all based on one common vision. These impacts, separately and combined, will generate economic and societal benefits for many key domains of strategic importance for Europe.
The project will use a multidisciplinary approach to make disruptive scientific progress. The enabling technologies created in the GA project will substantially advance and expand innovation in Information and Communication Technology, and will manifest Europe’s leading position in low power nanoelectronics and nano/microsystems. Many important European industry sectors such as health, energy, environment, transportation, and security will significantly benefit from these novel technologies.
The project goal is to develop zero-power technologies for these electronic personal assistants, so that they will harvest all the energy they need rather than requiring an external power source. The objectives of the project are divided into three interdependent categories: (i) scientific objectives centred on ultra-low power concepts, (ii) novel technologies, material and device integration solutions, and (iii) visionary zero-power autonomous systems demonstrators.
GA is a visionary project that takes on technological challenges for which no feasible solutions are currently available. GA research is goal-oriented, and the envisioned products have a wide range of applications in different areas. These in turn will require new fabrication equipment and tools as well as methods for the analysis of process results. Here, the potential benefit for European SMEs is especially high. The innovation process is thus not limited to fundamental research in science labs, but also encompasses application sectors and those benefiting from the results. Compared to existing ENIAC, CATRENE and ICT initiatives, this FET Flagship will not be limited to segments of the nanoelectronics innovation chain, but will instead unify energy-efficient technologies and integrated energy scavengers within one long-term design and integration platform. The proposed approach is based on heterogeneous integration of the best existing and emerging materials, technologies and devices for dedicated functions, each used for its best performance and most innovative functionality. It will improve the competitiveness of leading communication, equipment and tools companies as well as service providers.
Uniting 58 partners from 16 countries, the GA consortium is already cooperating closely with companies interested in future technologies and potential products. While established European companies with access to GA technologies will reach new markets, the range of multidisciplinary research results will lead to the creation of new start-ups and hence of new jobs. Since GA is a truly pan-European venture – from Ireland to Poland, from Finland to Italy – the funds, as well as the future benefits, are well distributed all over Europe.
1.3 Summary for the General Public
Guardian Angels are tiny, autonomous, wearable systems that we can choose to integrate into our everyday lives where needed, in a wide range of situations. They are based on networks of intelligent nanosensors designed to safeguard our health, safety and general well-being, and to improve our quality of life. Currently, high energy consumption and the short lifespan of batteries are obstructing further progress in autonomous systems. 7
Guardian Angels will meet the technological challenges of weaving together energy-efficient information processing, sensing, communication, and energy harvesting, into a zero-energy (battery-free) concept.
There will be three families of GA devices, all based on the concept of a smarter life: a lifestyle that benefits from the instant availability of relevant information, whether that information comes from within our own bodies (heart rate, insulin level, the amount of stress we feel, our attention or distraction levels) or outside them (pollutants, pollen, obstacles in our way). The three GA families: (1)
Physical GAs, which can give us information about our physical and physiological status, for purposes including health care, rehabilitation, or sports. If we choose, the information will be communicated securely with doctors or others in our sphere of health care providers. These devices, with their strong focus on disease prevention and early diagnoses, will help keep healthcare affordable and accessible to all. (2) Environmental GAs will focus their sensing on environmental conditions, serving as a sort of 6th sense to allow us to know what is in the air around us. For visually impaired people, they could help fill in visual information by "seeing" for them. In combination with the Physical GAs, it will be possible to correlate our physical state with the environmental context. (3) Emotional GAs will be able to perceive emotional or affective conditions such as stress or attention level, so that we become more self-aware in situations where it can work to our advantage, whether we are driving a car or are in a learning environment. The applications mentioned here are only a fraction of what can be created during the ten-year project, and are a tiny sampling of what is possible.
Strokes, autism, stress and other neurological disorders are affecting a growing number of people worldwide. Their consequences are devastating for the patients, their families, and society. GA will reduce the impact of these disorders by allowing patients to become more autonomous and to participate in activities that weren’t possible before. GA’s easy-to-use, low-power, wearable systems - based on the analysis of behavioural, contextual, and physiological signals (including brainwaves) - will enable independent rehabilitation and management of both motor and cognitive disorders. A stroke patient will play chess with the help of a neuroprosthesis that, upon recognising that he wants to reach for a piece and that his eyes are looking at the knight, will compensate for the missing motor capabilities in order to move the knight. An autistic girl will attend school with the help of "emotional" glasses that, by keeping track of the teacher's and other children’s behaviour as well as of her engagement, will stimulate her to make eye contacts and participate in the activities of the class.
Security and privacy are top priorities in the GA project. The data gathered will be yours; it will always be your decision to keep or to share it. In addition, the ethical aspects of GA applications will be assessed from the beginning of the project, through interactions between researchers, an ethics board of experts, and end users.
The technology-related project goal is to develop environmentally-friendly, battery-free technologies for these electronic personal assistants, so that they will harvest all the energy they need rather than requiring an external power source. GA devices will be 100 times more efficient than existing energy scavengers, harvesting energy from diverse sources including light, heat, and motion. In addition, the circuits and systems that need to be powered by these energy scavengers will use new technologies that consume less energy than any technologies existing today. All this will happen in a system so tiny that you will be able to wear it comfortably, in clothing, temporary skin patches, or "electronic skin."
The economic advantages will be numerous. European industry already has a solid base in the fields of microsystems, low-power electronics and system integration, encompassing communication, software, analytical instruments and fabrication equipment. Guardian Angels will build on this existing expertise, bringing economic benefits to the sector. The range of multidisciplinary research results will also lead to the creation of new start-ups and hence of new jobs. As GA is truly a pan-European venture – from Ireland to Romania, from Finland to Italy – the funds as well as the future benefits are well-distributed all over Europe. 8
In our project vision, GAs will provide data which will allow you to extract relevant information for a smarter life: making life easier when you are well, and maintaining or improving quality of life for those with health problems. Imagine mobile electronic personal assistants being 1000 times more energy-efficient than they can be today, so that they could be powered by the energy available in your environment without any power plugs. Imagine part of the energy savings transformed into sensing, communications and interface functions of an invisible system that becomes your day-to-day Guardian Angel. This project will address the grand scientific, technological and system engineering challenges in a time frame of 10 years to transform this vision into a reality. 9
2 S & T Quality
2.1 Scientific Vision and Unifying Goal
2.1.1 The big picture: Guardian Angels at the core of the information revolution
The
Guardian Angels for a smarter life FET Flagship project is designed as a visionary, science driven, goal-oriented, large-scale, multidisciplinary research initiative, centred on ICT future and emerging technologies. Today everyone is a user of ICT technology and benefits from the tremendous progress offered by silicon-based nanoelectronics. As with past technology waves, computer technology is currently reaching a plateau phase and facing fundamental limits and challenges from energy and economy-of-scale points of view. Our proposal will directly contribute to establishing new generations of ICT technologies and applications, and will unite three major innovation waves of the information revolution: (i) computer technology, (ii) distributed intelligence, and (iii) nanotechnology (see Fig. 1). The Guardian Angels (GA) project vision encompasses extraordinary societal benefits and applications, enabled by the autonomy of future energy-efficient nanoelectronic systems, powered by energy harvesters. Such autonomy requires disruptive progress in the fundamental principles and the engineering of low-energy systems. The project vision prioritises energy efficiency as the main driver for the information revolution, for creating autonomous smart systems that can act as true Guardian Angels for people, offering personalized advice and thus enabling a better life.
Fig. 1: Depiction of technology waves of the industrial and information revolutions.
2 The Guardian Angel FET Flagship vision unites the distributed intelligence, the nanotechnology and the computer waves, setting a common driver for applications: energy efficiency.
2
Adapted from a graph by Norman Poiré, Merrill Lynch.
3
J. Yick et al., Computer Networks, vol. 52, 2008, pp. 2292-2330.
4 H. Chaouchi, ed., The
Internet of Things: Connecting Objects, Wiley-ISTE, 2010.
2.1.2 The Vision and Benefits of GA
GAs are foreseen as interconnected, smart, autonomous systems enabled by energy-efficient nanotechnology, constituting the outer circle of applications depicted in Fig. 2; they can be considered as the future of wireless sensors networks
3 (WSN), and by their functionality they can include components of the internet of things 4. They will be interconnected, not only between themselves, but also through the gateway layer (mobile phones, PDAs, notebooks, tablets) to the inner circle of cloud (high-performance) computing. By their smartness and complexity they will enable personalized advice and assistance, concerning health and interaction with the environment, far beyond what today’s WSN and internet-of-things devices can provide. GA technology will offer unique solutions for new generations of non-invasive biological monitoring, and for future smart apparel with embedded powering and sensing. They will enable unforeseen generations of autonomous robots. The supporting GA zero-power technology platform will impact development within other domains such as 10
environmental, building and industrial monitoring, and efficient transportation. It will offer new progress paths for energy-efficient data processing in cloud computing, and change the way mobile computing interacts with humans’ needs.
Fig. 2: Positioning of GA zero-power technology and applications with respect to cloud computing and gateway (mobile communications and computing) technologies. Guardian Angels are smart, autonomous systems that are beyond wireless sensor networks in terms of functionality and powering, and include a higher complexity than the simple sensor nodes foreseen today for the internet of things. GA will strongly interact with the gateway and cloud layers, and the project will create direct benefits for the future of mobile computing, and more energy-efficient architectures for high-performance computing.
In our project vision, GAs will provide data which will allow us to extract relevant information for a smarter life: making life easier when you are well, helping to efficiently use energy sources, and maintaining or improving mobility and industrial processes without exhausting natural resources. GAs will also play a vital role for those of us who need increasing, or even continuous, support and services due to health problems or reduced mobility or sensory capabilities. Imagine mobile electronic personal assistants being 1000x more energy-efficient than they can be today, so that they could be powered by the energy available in your environment without any power plugs. Imagine part of the energy savings transformed into sensing, communications and interface functions of an invisible system that becomes your day-to-day Guardian Angel. This project will address the grand scientific, technological and system engineering challenges in a time frame of 10 years to transform this vision into a reality.
2.1.3 Unifying goal: the science of zero power
Guardian Angels are defined as future zero-power, intelligent, autonomous systems featuring sensing, computation and communication beyond human aptitudes. The science of zero-power involves exploratory research at the level of novel materials, devices, and system architecture that could enable energy savings by a factor of up to 1000 for the computation, communication and sensing functions. These will be combined with novel, smart, multi-harvesting interfaces able to detect and adapt to the most appropriate sources of energy, which will lead to an improvement of a factor of 100 in the harvesters’ energy output. Fig. 3 summarizes the 11
integration of the main constituting functions into a single GA system, showing the driving force of the R&D, which is the improvement of the energy efficiency by orders of magnitude. As will be shown later, the GA consortium has identified the most credible technology principles and candidates to address these extraordinary challenges.
The zero-power GA is, therefore, the unifying goal of the project
, and is defined as a system’s ability to feed from the energy existing in dynamic environments, by harvesting various types of energy sources. The project will devise novel device concepts and material integration technologies for solar, thermal, vibration, and electromagnetic energy harvesters, and it will explore new bio-inspired, e.g. bio-chemical or synthetic photosynthesis-based energy scavengers, which will target power densities of 10 mW per cm2 (or per cm3) by a combination of different types of harvesters depending on the application context. These battery-free, or in other words, zero-power systems will enable non-intrusive, independent and cost-efficient intelligent devices, which will communicate with each other and will support human beings by sensing biological signals and the environment.
2.1.4 Overview of main objectives: enabling GA autonomous personal assistants
The GA objectives are categorized into three groups:
(I)
Scientific Objectives (SO) – These objectives concern fundamental research and involve the identification of underlying principles, devices, system architectures, algorithms and techniques that can advance the limits of today’s energy consumption for each of the elementary functions composing the GAs, in order to achieve the required system autonomy.
(SO1)
Energy limits of computation: reduce the energy per binary switching from 100,000 kBT down to 10-100 kBT (or from 100 aJ to 0.1 aJ per binary switching event). This research involves new switch concepts (like sub-thermal subthreshold switches), based on different physical mechanisms than conventional field effect transistors. It also involves new architectures. The goal is to devise principles and technologies that allow the voltage supply of logic circuits to be scaled from 1V down to 0.2V, with negligible leakage current, which will offer a power gain larger than 100x.
(SO2)
Energy limits of communication: to reduce the total end-to-end energy consumption of communicating a useful information bit from 10nJ down to 10pJ/useful bit, including transmitter and receiver processing energy, RF front-end energy and transmitted energy. Candidate technologies are combining (1) flexible and adaptive radios, (2) reduced complexity radios, and (3) extreme duty-cycled burst-mode radios. For each research line, innovation is proposed at 4 different granularity levels of the communication sub-system: system level, algorithm level, circuit and antenna level, and device level.
(SO3)
Energy limits of sensing: reduce the energy per integrated sensing event (including the first stage of the read-out interface) from 100 W to 100 nW. Small size and ultra-low energy bring noise limitations, as well as challenges in terms of response time, selectivity and stability. We will explore principles and test paradigms that will allow the reduction of energy consumption by a factor of 1000. Such energy savings, along with nanotechnology, will enable multi-sensing or sensor arrays in a single smart system. The fusion of sensor data with ultra-low power will play a vital role in accomplishing the goals of the project.
(SO4)
Energy limits of human-machine interfaces: the project will address the limits of zero-power human-machine interfaces and multi-modal sensing for affective state categorization. This includes scientific and technical energy challenges for brain-to-machine interfaces (including electroencephalography (EEG) and electrooculograhy (EOG)), in which there is no need to speak, gesture, or type into a keyboard to communicate with machinery.
12
(SO5)
Energy limits of harvesting: the project will explore and push the limits of energy scavengers, which should operate both in outdoor and indoor conditions and achieve levels of energy needed for GA systems 100x higher than the state-of-the-art. Several types of scavenger principles could be combined in a GA system, requiring disruptive materials and devices for solar, thermal, vibration, and electromagnetic, and significant advances in bio-chemical and synthetic photosynthesis scavengers. Research will include novel energy storage elements that could be hybridized with the electronic systems to facilitate a realistic transition to the final zero-power autonomous systems.
Fig. 3: Novel sensing, computation, communication and energy harvesting technologies are the basic blocks that will constitute the GA systems. A dedicated operation system, communication and data security techniques will be combined with the hardware components to enable the GAs. (The energy limits shown are based on theoretical calculations, while the limits shown in the roadmaps are re-evaluated at the system level, corresponding to an energy limit per
function.)
(II)
Technology and Integration Objectives (TIO) for implementing the zero-power technology platform supporting the GA demonstrator systems.
One of the main goals of the project is to develop and implement a zero-power technology platform as a combination of future energy-efficient technologies and disruptive energy scavengers. The main components of the platform that together form the technology and integration objectives:
(TIO1)
Energy-efficient technologies (computation, sensing, communication)
(TIO2)
Highly-efficient energy harvesting and energy storage
(TIO3) Zero-power system design.
(TIO4)
Heterogeneous integration
(TIO5) Software: operating system, communication and power management:
This objective also includes software-hardware interaction and all software-level components: operating system, power management, data security techniques and energy efficient algorithms and software codes.
The zero-power technology platform is designed such that it can enable the implementation of two categories of demonstrators: (i)
full-hardware demonstrators, and (ii) virtual demonstrators, as illustrated in Fig. 4. Early in the project, a full-hardware demonstrator will be the implementation of a GA system using existing mature technology blocks, already available for full integration at a given moment in time, for the experimental proof-13
Generation 1Computing module V1Communication module V1Sensor V1Energy harvester module V1Full-hardware platformsVirtual platformsGeneration 2Computing module V2Comm. module V2Sensor V2Energy harvester module V2V2 new functionalityModel of computing module V2Model of energy harvester module V2Model of comm. module V3Model of sensor V3Generation 3Computing module V3Comm. module V3Sensor V3Energy harv. module V3V3 new functionalityModel of new V2 functionalityModel of new V3 functionalitynanopowerharvesterRadioFront-endDSPEnergyharvestingSensingActuatingCommunicating
of-concept of all GA functions with energy efficiency. Then, the virtual demonstrators will create the link between full-hardware demonstrators using existing technologies, and emerging technologies that have achieved measurable device or elementary circuit block characteristics, without featuring full integration. At that point, calibrated libraries of models for multi-scale and multi-physics simulation will make it possible to evaluate the impact of an emerging energy-efficient technology at the system level, in order to facilitate benchmarking and the selection of the most promising emerging technologies for the next generation of full-hardware GA demonstrators.
Fig. 4: Full-hardware and virtual platforms for demonstrators in the GA project.
(a) (b)
Fig. 5: (a) Components of the zero-power technology platforms driven by energy efficiency for system autonomy. (b) One GA node and its key components.
(III)
Zero-Power System Objectives (SysO)
GAs are personal companions, electronic devices or electronic systems, which are small and thus inconspicuous and nonintrusive; they are autonomous and thus easy to use; they are personalized, secure, and under full control of their users, and thus safe and trustworthy. Guardian Angels will actively assist humans from their infancy to 14
old age in complex life situations and dynamic environments by offering access to an augmented reality that includes biological and environmental signals. Three families of GAs are targeted, each constituting measurable objectives of the proposed systems; progress on these families will be reported in terms of
full-hardware and virtual demonstrators in a well-defined time frame (year 3, year 7, year 10). Fig. 5(b) shows the main components of a wireless, autonomous sensor system node. In this section we describe some of the main characteristics of the three GA families that form the system objectives.
(SysO1)
Physical Guardian Angels
The Physical Guardian Angels are quasi-invisible, zero-power body area networks or, if appropriate, implantable devices, monitoring vital health signals and offering the necessary information for taking appropriate actions to preserve human health. They will acquire a well-defined view of the state of a person’s health adapted to individual needs, by using a real-time, ultra-low-power, multi-parametric combination of non-intrusive, bio-signal sensors (ECG, accelerometers, gyroscopes, pulse oximetry, etc.) to allow for early warning and thus enhancement of the quality of life. They can employ future GA technologies such as electronic skin or wearable networks of sensors with wireless interfaces. These systems will be compatible, from the communication point of view, with existing gateways (such as mobile phones) to serve as smart parts of a future vision of the
internet of things5. First medical applications of Physical Guardian Angels will deal with monitoring of the elderly for frailty assessment, exploring metabolic diseases, and offering prevention solutions, early diagnosis and response to therapeutic interventions.
5 http://www.iot-visitthefuture.eu/fileadmin/documents/researchforeurope/
(SysO2)
Environmental Guardian Angels
The Environmental Guardian Angels extend their abilities from the body to monitoring of our daily environment, featuring zero-power, bi-directional interfaces, full battery-free operation, disruptive scavengers (biochemical, thermoelectric, synthetic photosynthesis), personalized data communication, and algorithms permitting decisional processes. These devices will offer access to an augmented reality including alerts for hazards, e.g. electromagnetic or ionizing radiation, extended UV exposure, the concentration of allergens, pollens and harmful gases. Moreover, they will be designed as real personal assistants to protect our children and maintain quality of life for elderly people. These sophisticated GAs will guard people from diverse environmental dangers, including pollution and catastrophic events, rendering our environment safer. These devices will feature complex, energy-efficient communication technologies (based on novel materials like graphene), both from GA to GA, and from GA to other gateways, offering complete networking capabilities. The environmental Guardian Angels will be developed for various scenarios: (1) as smart air quality companions for indoors and outdoors, (2) as enablers for the inclusion of visually impaired and blind people in society and (3) as a trusted personal device for complex disaster management. Extensions of these GAs can be foreseen for other types of smart monitoring, for industrial, environmental and transportation applications.
(SysO3)
Emotional Guardian Angels
The Emotional Guardian Angels are intelligent personal companions with zero-power human-machine interfaces to sense the body’s reactions to one’s emotions or state of mind, correlated with environment and context, in order to provide objective and holistic information to improve services, benefitting both society and the economy. Emotional GAs will offer two major categories of services: (1) enhancing the performance or well-being of healthy people and (2) supporting patients to ensure that they live as normal a life as possible, or even recover lost motor functions and cognitive abilities. They are expected to play an important role in society, and will form a completely new generation of devices, not even imaginable today, based on human-technology interaction. They could assist people in capacities such as smart automobile driving assistants or air traffic controls, providing feedback if the user is too tired or emotional to control a vehicle, or using their inter-GA communication interface to avoid accidents. They could potentially aid elderly people with Alzerimer’s disease. They could provide quadriplegic sufferers with greater control of their environment, by enabling nonverbal 15
decision-making and communication. These GAs could also help families and educators of those with autism spectrum disorders to understand and use alternative means of nonverbal communication. Emotional Guardian Angels may be the first intelligent systems of their kind for maintaining or even extending quality of life for patients who suffer from physical and mental health problems; they may interpret intentions and communicate with people in a completely new way, disruptive in comparison with existing technology.
(IV)
Governance, Management and National Matching Objectives (Gv&MgO)
The objectives listed under this section are designed as a part of the GA strategy for efficient yet dynamic governance and management along the ten-year lifetime of the FET Flagship. The governance and management of the GA FET Flagship shall be conducted based on an operational model that takes two key boundary conditions into account: (1) a 30-month ramp-up phase funded as a CPCSA FP7 instrument, followed by a more flexible, adapted funding mechanism under FP8, (2) the matching of the FET Flagship by national states and/or by combined ERA-NET+ instruments. The following main objectives are proposed to encompass both the high-risk science and the high-impact ICT engineering specific to the concept of FET Flagships, with a goal-oriented governance structure.
(Gv&MgO1)
Implement a goal-oriented and milestone-based strong internal governance. The governance should be capable of defining and enforcing the major scientific milestones of the project. For this purpose, a Scientific Steering Board will include key scientific representatives of core EU partners. A Board of Directors will professionally implement the strategic direction into a work programme with clearly-defined major milestones for 3 major periods of scientific and technical reporting. In addition, a Board of National State Representatives will help the project integrate research activities within the member states (notably by adopting the FET Flagship’s milestones) and achieve maximum national impact and the needed matching.
(Gv&MgO2)
Implement a strong management capacity of the leading houses. The objective is to implement a professional management structure at all levels. The organisational model will co-integrate scientists and expert managers from outside academia and will follow a roadmap-based strategy, creating the necessary conditions for interactive work with a FET Flagship spirit. This includes the implementation of a Project Office with public relations and dedicated departments for finance, legal affairs, intellectual property rights, communication, project evaluation and internal auditing, and the development of organizational strategies to recruit and remunerate suitable management staff. The GA leading houses are EPFL and ETHZ; the two Swiss Federal Institutes of Technology, the major components of the ETH domain in Switzerland, will define their joint management with the establishment of a non-profit Guardian Angels Swiss Foundation and a clear split of responsibilities and tasks.
(Gv&MgO3)
Hybrid operation with a well-defined balance between a project, a program, and a dynamic partnership. The GA FET Flagship will have an early identification of key contributors to the zero-power technology platform and system demonstrators and applications. This will be reflected in clear roles in the CPCSA, from the ramp-up phase that will bring together Core and Associated Partners. The Openness of the Consortium, initiated in the Pilot Phase with 2 calls attracting 30 new partners, will continue with one call during the ramp-up phase and three other calls through the end of the FET Flagship. The GA FET Flagship will create the right conditions for goal-oriented networking, where the selection of the new partners will be based on excellence criteria needed to achieve concrete priorities in research, as defined by the project governance and the scientific and technology objectives.
(V)
Dissemination, Exploitation and Intellectual Property Objectives (DEO)
(DEO1) Define and implement a dissemination strategy covering both scientific and non-scientific communities. The scientific dissemination will include the very early identification of high-impact scientific journals, conferences and technical events, and a strategy for coordinated publication of
16
topical high-level reviews specific to the flagship roadmaps. Specific dissemination channels (media, social networks, Wikipedia, etc.) for a larger audience will be set. Additionally, GA partners will act as everyday knowledge disseminators through their work in universities, their teaching and as members of governing committees. They will be able to impact the course syllabi and research activities to focus on future and emerging technologies in the information and communication sector. In order to enable collaboration between companies, the project will use open innovation principles. Additionally, the long-term scientific roadmap will be made public.
(DEO2)
Define and implement a flagship-specific exploitation strategy and associated intellectual property rules for transferring knowledge and new technologies to industry, in order to apply the resulting technologies widely for the benefit of society and the economy. Given the complexity of the FET Flagship and the particularly high involvement of industry in GA, a specific Consortium Agreement will be designed and adopted. It will include the possibility to define IP-protected sub-projects together with a pre-competitive knowledge zone where the project partners will be able to share advances in research (lying between fundamental basic research conducted mainly in universities, and proprietary research performed in corporate laboratories) concerning the zero-power technology platform. A unique feature of GA will be to install an Intellectual Property and Exploitation Committee that, via a concrete mechanism, will create and support the creation of a minimum of 10 start-ups emerging from the project. Additionally, the industrial partners within GA will drive an effort for standardization of the technology platform components.
(VI)
International Collaboration Objectives (ICO)
In the global picture of advanced nanoelectronics and smart systems, globalization is a reality and international collaboration is a must for success and wide impact; therefore, the GA Flagship has defined some concrete objectives for international collaboration outside Europe.
(ICO1)
Setting a collaborative effort with leading partners from the USA. This particular objective is motivated by the fact that some of the leading research institutions in the US have needed particular expertise in the research field proposed by the GA project, and some of them are mother companies of some of the European partners. The objective is to identify clear mechanisms for concrete collaborations (beyond the co-organization of scientific events) with some already-identified US partners (MIT, Stanford, UC Berkeley, IBM and Intel) that will have national fund-matching to contribute to the GA scientific and technological challenges. We will also target particular win-win collaborative efforts with complementary US initiatives having a similar size as our FET Flagship such as the One Mind for Research initiative of the medical community.
(ICO2)
Setting a collaborative effort with leading partners from South Asia and Japan. Some specific technologies and application innovations will be the subject of a collaborative effort defined outside Europe, and supported by matching with national programs in Japan and/or South Asia. This action will ensure an even more global impact of the GA Flagship by considering world-wide societal aspects and the interest of these technology-hungry regions of the world for novel electronic technologies for new services. European companies will benefit from the new markets and business resulting from these collaborations.
(ICO3)
Setting a collaborative effort with other world-wide partners. Under a more generic collaborative effort, GA will consider collaborative efforts for new ideas for applications of the GA demonstrators, and/or for enlarging the project vision with a priority on solutions addressing the issues of sustainability and the role of affordable (low-cost) energy-efficient technologies.
17
Fig. 6: The energy dissipation per logic operation is approaching the fundamental limit. On-chip communication consumes orders more than the switch (e.g. inverter). The energy consumption of today’s systems with von-Neumann architecture (e.g. for a 64-bit floating point operation) is dominated by capacitive losses in memory interconnects. New computing architectures such as IBM’s cognitive computing project SyNAPS could reduce energy consumption by orders of magnitude.
1940196019802000202010-1910-1610-1310-1010-710-410-110-11021051081011101410171020 Energy [Joule]YearTodayenergy / IBM SYNAPSE eventenergy / logic operationinverter only incl. on-chip comm.energy / floating point op. Energy [kT]3kT ln(2)
2.2 Main Scientific Objectives
2.2.1 Grand scientific challenges for energy-efficient computation
Guardian Angels will include logic and memory functionality for many purposes, including processing the output signals of the various sensors, decision algorithms, running the operating system, and running specific software for energy optimization, communication, and other uses. The ambition to attain zero-power systems requires a radical (1000-fold) reduction of the energy consumption for computation.
Fundamental scientific limits:
Thermodynamics and quantum mechanics set fundamental limits for the energy transfer during binary switching. For the relationship between switching energy E and transition time td, the Heisenberg uncertainty principle requires that E h/td. The minimum energy required to preserve a binary state can be estimated from the Boltzmann probability as Ebmin= 3 kBT ln(2) ≈10-21J (T=300K).6,7 Irreversible or many-to-one operations such as AND or ERASE require dissipation of at least Ebmin for each bit of information lost. In principle, reversible or one-to-one logical operations such as NOT can be performed without dissipation, as shown by Landauer.8 The drawback of reversible or adiabatic computation is that system switching speed is proportional to the energy dissipation; hence to achieve significant energy savings, prohibitively low speeds may be required. A detailed discussion of the ultimate limits of a computer was proposed by Lloyd9: he suggests that the speed per logical operation is limited by its energy, and the amount of information that can be processed is limited by the number of degrees of freedom of the computing system.
6
V.V. Zhirnov et al., Proc. IEEE, vol. 91, 2003, pp.1934 – 1937.
7
J .D. Meindl et al., Science, vol. 293. 2001, pp. 2044 – 2049.
8
C.H. Bennet, Int. J. Theor. Phys., vol. 21, 1982, pp. 905-940.
9
S. Lloyd, Nature, vol. 406, 2000, pp. 1047-1054.
10 R.W. Keyes, IBM J. Res. Develop. vol. 32, 1988, pp. 24-28, data updated by T. Theis and R. Keyes, IBM Research 2010.
11
Y. Ye et al., IEEE J. Sol. State Circuits, vol. 36, 2001, pp. 239-248.
12
J.J. Welser et al., J. Nanopart. Res., vol. 10, 2008, pp. 1 - 10.
13
G. P. Boechler et al., Appl. Phys. Lett., vol. 97, 2010, p. 103502.
State of the art:
Historically, the advancements in digital technology have reduced the dissipated energy per operation by roughly one order of magnitude every five years. Today’s advanced CMOS technology operates at energies on the order of 104 - 105 kBT per binary switching event using MOSFET switches and von-Neumann architectures.10 Several (quasi-)adiabatic circuit designs have been implemented. Typically, power reduction compared to standard CMOS lies within one order of magnitude. However, maximum transition frequencies lie in the 100MHz regime and considerably larger circuit footprints are required.11 The need for alternatives to charge-based logic are being explored.12,13
Challenges to be addressed:
It is the goal of the project to propose, demonstrate and exploit highly efficient computation technology, devices and software to reduce this energy by a factor of 1000 while safely operating one hundred times higher than the fundamental kBT ln(2) limit. Combining device and architecture research with the goals of maximizing energy efficiency while maintaining realistic operating speed, room temperature operation and cost figures of merit will enable truly useful applications. 18
Fig. 7: Fundamental limit for the minimum transmitted bit energy versus the maximum allowed signal attenuation.
506070809010011012010-410-2100102104Bluetooth (IEEE 802.15.1), ZigBee (IEEE 802.15.4)UWB-ZigBee (IEEE 802.15.4a) Transmitted energy (pJ/bit)Signal attenuation (dB)Region for which communication is impossibleEb = 10(L/10)ln(2)kT 1 pJ/bit
2.2.2 Grand scientific challenges for communications
An essential feature of zero-power Guardian Angels autonomous systems is the ability to communicate wirelessly using minimal energy per information bit. In a typical state-of-the-art sensor node, the wireless communication, necessary to upload sensed data and receive configuration settings, often consumes a significant part of the available energy. The GA project therefore aims at a radical energy reduction for communication by a factor of 1000 compared to current state-of-the-art. The goal is to achieve a transmitted energy per bit of 1pJ, approaching the fundamental scientific limit, with a total system energy of 10pJ/bit.
Fundamental scientific limits:
A fundamental performance limit for the required transmitted energy in a wireless communication system can be deduced from the Shannon capacity theorem14, which assumes that the transmitted signal is Gaussian using an infinite block code interval, the channel is constant and perturbed by additive white Gaussian noise (AWGN), and the detector knows the deterministic channel state. One of the most fundamental implications of the capacity theorem is that communication is not possible below the received signal-to-noise ratio per bit of -1.6 dB in the wideband limit.15 Assume optimistically that only the thermal noise with density N0 = kBT affects a communication device. At room temperature, the minimum required transmitted bit energy versus the allowed signal attenuation resulted from the capacity theorem is shown in Fig. 7 (assuming that the receiver is noiseless), showing that communication with 1pJ/bit is feasible when transmit-receiver attenuation is limited below 85dB.
14 C. Shannon, Bell System Technical Journal, vol. 27, 1948, pp. 379-423.
15 S. Haykin, John Wiley & Sons, Hoboken, NJ, 4th ed., 2001.
16
C. Bachmann, IEEE Communications Magazine, vol. 50, issue 1, 2012, pp. 20-27.
17
Report from CATRENE Working Group on Energy Autonomous System: "Energy Autonomous Systems: Future Trends in Devices, Technology, and System", p. 32.
State of the art:
A general objective in the design of wireless communication systems has been the information-theoretic capacity limit discussed above. In order to achieve this goal in AWGN channels, very powerful coding schemes were developed including iterative turbo and low density parity check decoding methods. The effect of multipath fading was reduced by using time, frequency, or spatial diversity techniques. These approaches, however, all aim to approach the Shannon limit at the expense of a large increase in computational complexity, hence possibly negatively affecting the overall energy consumption of the transmitter and receiver. Until recently, very little attention has been paid to minimising total energy consumption. State-of-the-art systems typically consume less than 1nJ/bit in terms of transmit power, though their overall energy per bit goes up to 10 to 100nJ.16,17
Challenges to be addressed:
In order to achieve zero-power systems, the GA project will aim at a reduction of the overall system energy for communications. To this end, the target energy per bit metric encompasses all energy spent in the transmitter (signal processing and analog front-end), the receiver (signal processing and analogue front-end), as well as the actual energy transmitted into the air. Moreover, this energy per useful bit metric differs from the plain energy per bit metric, as it incorporates all MAC and network layer overhead, as well as all energy spent in sleep, stand-by or listen mode. Under this definition, the GA project targets an energy per useful bit of 10pJ for communication over a 2m distance. Cross-level and cross-domain research will allow the assessment of trade-offs between transmitted energy and signal processing energy, between transceiver front-end performance and back-end complexity, between active modes and sleep states, and between the upper and 19
Fig. 8: Plot of different accelerometer devices or technologies in the resolution-power system of metrics. Resolution here is defined as noise spectral density. Accelerometers are distinguished based on three criteria: Status {(c)ommercial or (r)esearch}, Conditioning {(A)nalog, (D)igital or None(X)} and Principle {(C)apacitive, piezo(R)esistive, (p)iezoelectric, (o)ptic, resonant(f)} (see legend letters).
lower communication layers (network, MAC, PHY). Additionally, the communication subsystem should be truly adaptive and linearly scalable in terms of system parameters like data rate, communication distance, number of nodes, etc., to enable dynamic adjustments towards minimal energy under all circumstances.
2.2.3 Grand scientific challenges for low-power sensing
In Guardian Angels, sensors will collect physical, physiological, chemical and biochemical data. Multiple sensor data will be fused to interpret and monitor a person’s physiological and emotional status, in relation to the actual environmental and social/situational context. By definition, a GA sensor refers to the functional device transforming real world information to electronic information. To afford several sensors per GA, the power consumption target per sensor element/function was set to 100 nW. This power consumption target includes readout electronics, signal conditioning and AD conversion, but excludes digital signal processing such as linearization, pattern recognition and sensor fusion.
18
18 Power consumption of digital circuits is considered in section
2.2.1 about "energy-efficient computation"
19
Ekinci et al., J Appl Phys, vol. 95, 2004, pp. 2682-2689.
Fundamental limits in low-power sensing:
Limits in sensor systems are well represented by the signal-to-noise ratio (SNR), which incorporates the sensitivity, bandwidth and noise spectral density of a sensor and determines its limit of detection (LOD) and resolution. Miniaturization reduces the power and energy needed to drive the sensor due to the smaller number of charge carriers requiring transport, or simply by avoiding heater elements. However, it has been observed that electronic and mechanical noise, in particular 1/f noise, increases with smaller device dimensions. In nanostructures, excess electronic noise (carrier number and mobility fluctuations) can span from μz to kHz, posing severe constraints on signal evaluation techniques, which are not discussed here. Consequently, the limits in sensor systems should be described by SNR, LOD or resolution – in ambient conditions – with respect to the size and power consumption of a sensor.
State of the art:
In Fig. 8, we use accelerometers as representatives of a class of highly developed sensors to visualize the state-of-the-art for this category of sensors and the trade-off between power consumption and the noise spectral density (which gives the LOD when multiplied with the square root of the bandwidth) without any loss of generality for other sensor categories. Today’s low power solutions are still 2-3 orders of magnitude away from the objectives in GA.
Challenges to be addressed:
The grand scientific challenges for GA sensors are both to explore low power transducer concepts for the GA functions mentioned above and also to push the power consumption of the sensors down to the order of magnitude of 100 nW per function. It is obvious from the analysis above that pushing the front (Fig. 8, dashed line) requires a non-incremental paradigm shift, such as nano-devices and –materials, for example. However, as soon as one of the relevant dimensions of a sensor reaches the nano scale, noise increases considerably, SNR approaches unity and the sensor characteristic itself becomes nonlinear. This trend has been captured explicitly in the model equations governing nano-electro-mechanical systems in particular,19 but it probably applies to other sensors. Importantly, though engineers attempt to avoid non-linearity and noise, biological sensors typically operate in a noisy, non-linear regime. For example, our visual system, which allows detection of a single photon, operates over 14 orders of magnitude. Similarly, sound 20
pressure detection in our inner ears allows detection of both a flying mosquito and the noise of thunderstorms. Inspiration by biological principles is expected to improve ultra-miniaturized sensors in important ways, and allow the GA challenges to be met. Indeed, scientists have already begun adopting similar mechanisms in their sensors.
20,21 Another strategy worth pursuing would be to combine several functions within one device, as has been demonstrated with a CNT that serves as all essential components of a radio,22 and predicted for piezoelectric nanowires who could supply their own energy source.23
20
Almog et al., Appl Phys Lett, vol. 90, 2007, pp. 013508-1/3.
21
Karabalin et al., Phys Rev Lett, vol. 106, 2011, pp 094102-1/4.
22
Jensen et al., Nano Letters, vol. 7, 2007, pp. 3508-3511.
23
Xu et al., Nanotechnology, vol. 22, 2011, p. 105704.
2.2.4 Grand scientific challenges for energy harvesting, storage and power management
Research within the consortium will address the fundamental challenges involved in converting different forms of energy available in the environment (solar, thermal, chemical, and mechanical) into electric energy, and efficiently storing and managing the converted energy to power the autonomous, zero-power Guardian Angels.
Fundamental limits
|
State of the art
|
Challenges to be addressed in Guardian Angels
|
Energy Harvesting (EH)
Solar Cells
|
Shockley–Queisser (SQ) limit of 33.7% efficiency for single junction solar cells
|
Single crystal Si solar cells approach limit but are costly.
SOA Efficiency of thin film solar cells:
Amorphous Si: 12-13%
CIGS : 19-20%
CdTe: 16%
DSSC(liquid): 12-13%
DSSC(solid): 6-7%
Polymer: 8%
|
Surpassing SQ limit with new device architectures (tandem solar cells) and new materials (multiple exciton generation, etc.);
Development of light trapping structure for bulk heterojunction polymer & DSSC cells
|
Thermoelectric
|
Carnot efficiency:
(1-T
cold/Thot), T in K.
4% for T
hot=37°C and Tcold = 25°C
|
Efficiency; dimensionless figure of merit ZT = 1 at room temperature
|
Synthesis of room temperature thermoelectric materials with ZT 3.0;
Integration of superlattice, quantum dot structures
|
Biofuel Cells
|
Maximum cell voltage due to fixed concentrations of glucose (e.g. 8 g cm
-3 in blood) and dissolved oxygen (e.g. 2 g cm-3 in blood) and fixed operating temperature ( 37 oC)
|
4.8 W/mm
2 at 0.60V and 6.5 W/mm2 at 0.13V in different physiological conditions
|
Development of 3D structures to allow direct electron transfer between the electron and the enzyme to improve the operating cell voltage
|
Mechanical
|
Electro-mechanical energy conversion efficiency: spring constant, damping (Q factor in resonant systems) and electro-mechanical coupling coefficient. Band width for a single generator
|
Mechanical to electrical conversion efficiency of 50%;
Device efficiency of 30%;
Available mechanical power 10 μ for 0.1 cm
3 device
|
Low frequency, wideband / non linear, small optimal load, resonant/non resonant, vibration harvesters, High k spring structures, nano-engineered epitaxial and nanowire materials (flexo, multi-ferroics, lead free materials). MEMS to NEMS hybrid generators
|
Energy Storage (ES)
Batteries/ Microbatteries
|
Anode and cathode material determines capacity and cell voltage;
Lithium diffusion and electronic transport in micro-energy storage applications
|
Lithium-based materials (~400 mAh/g and 4.5 V);
Thin film batteries, dense sintered cathodes
|
New materials such as iron fluorites;
Integrated on chip solution,
3D nanostructuring of electrodes (e.g. CNTs)
|
