Cognitive Electronics (COEL) Project
1 like153 views
This document discusses a proposed project for an upcoming H2020 call focused on developing adaptive smart working environments to support active and healthy aging. The project would aim to improve data collection methods for monitoring musculoskeletal disorders (MSDs) using miniaturized wearable and implant sensors for real-time posture and load monitoring. Both local processing for early warnings and remote cloud-based analytics of big sensor data using machine learning are proposed. Relevant technologies, stakeholders, and potential partner institutions across Estonia, France, UK are outlined to support the transdisciplinary nature of the project.
Cognitive Electronics (COEL) Project
- 1. 1 Cognitive Electronics (COEL) Project Muhammad Mahtab Alam ERA-Chair Holder (COEL) Associate Prof. PhD, MIEEE, MIET, Ceng Thomas Johann Seebeck Department of Electronic (TJS-ELIN) School of Information Technology, Tallinn University of Technology Tekk Tour Digital Wallonia. 06-11.2017, Mons, Belgium This work has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 668995. This material reflects only the author's view and the EU REA is not responsible for any use that may be made of the information it contains.
- 2. Thomas Johann Seebeck Department of Electronic (TJS-ELIN) 1.Semiconductor materials based structures and devices Development of diffusion welding (cold bonding) technique for realization of improved high quality metal contacts or wafer based anisotropic hetero-polytypic interfaces (specific wide bandgap heterojunctions); Lab-on-Chip solutions for medical applications (e.g. Heater design, THz measurements for detection of content of droplets, etc.). 2.Impedance Spectroscopy and its applications Bioimpedance application in biological environments (e.g. Medical and rescue applications like the check of the quality of tissue surgery); Impedance and eddy current measurements for material parameters detection (e.g. Constractive surgery applications, Euro-coins quality). 3.Cognitive Electronics Sensorics and (wireless) sensor networks, 5G, Internet-of-Things, Wearable Electronics and Communication, Low- Power, etc. (e.g. rescue vests for life guards, smart city solutions, etc.). Head of T.J. Seebeck Department of Electronics: Prof. Toomas Rang Staff: ca. 45, PhD students: ca. 35 Budget: ca. 2.8 MEUR yearly
- 3. • Key Objectives: • To improve the research productivity of 13 EU member states • Improve the impact and quality of the research • Problem solving approaches and applied research • To build new collaborations • Budget: 2.5 Million Euros (Project ends 31 November 2019) • Research and Development Focus: • Internet of Things - Wireless Sensor Networks • Communication Technologies • Wearable Wireless Networks • Ultra Low Power/Energy Neutral and Harvesting • Smart Sensor-Signal Processing • Wireless Communications • LTE-A/5G – device to device communication. • Public Safety Networks – Connectivity Issues • Cognitive Radio – Distributed Sensing and Estimation COEL ERA-Chair (Cognitive Electronics) EU H2020 Widespread Program 2016-2019
- 4. Glimpse of Our Competencies (Mahtab and Yannick) RESEARCH DOMAIN TOPICS COVERED STANDARDS PLATFORMS COLLABORATION S WIRELESS BODY AREA NETWORKS/ HEALTH MONITORING/PUBLIC SAFETY NETWORKS Optimized Joint PHY-MAC system design: (specifications). MAC Layer: (Scheduled Access, CSMA/CA). Throughput-Channel Aware MAC. Coexistence: Time-shared, freq. hopping – Only non-collaborative. Realistic Performance Evaluation: MAC, Coexistence, Channel Models (Biomechanical deterministic and dynamic models) IEEE 802.15.6 (Strong Experience) Shimmer Sensor SDKs Network Simulator (WSNET – Cross layer design) CEA-LETI (Grenoble, France) INTERNET OF THINGS - WIRELESS SENSOR NETWORKS Accurate Hybrid Energy Model: real-time measurements + analytical analysis MAC Protocol: Traffic-aware dynamic MAC (TAD-MAC). Heuristic Modeling of TAD-MAC IEEE 802.15.4 (Good Experience) MSP-430 University College Cork (Cork, Ireland) DIGITAL SIGNAL PROCESSING Adaptive Systems, Multi-standard transceiver designs, SDR (Multi- rate processing techniques – Channelizers, baseband processing) IEEE 802.11 a/b/g/n UMTS DSP/FPGA-based design GateHouse (Aalborg, Denmark) RESEARCH DOMAIN TOPICS COVERED STANDARDS PLATFORMS COLLABORATI ONS EMBEDDED SYSTEMS H/W and S/W Co-design: (Design space exploration, allocation, partitioning, scheduling). Metrics for Characterizing Algorithms: Execution-time, area, power FPGA-Technology: Altera, Xilinx. Quartus II, ISE DSP Builder, System Generator Department of Computer Engineering (TTU) LABSTICC - France INTERNET OF THINGS - WIRELESS SENSOR NETWORKS Cognitive WSN: Frequency and Power Allocation Algorithm, Reinforcement learning Energy Harvesting: Solar, thermal, RF Mesh Networks: IPv6 Sensor Node Platforms Department of Computer Engineering (TTU) DIGITAL SIGNAL PROCESSING Hardware implementation of baseband signal processing SDR, ADS-B, LTE Aalborg University – Denmark
- 5. Project Idea for Upcoming H2020 Call SC1-DTH-03-2018 Adaptive smart working and living environments supporting active and healthy ageing
- 6. What is Musculoskeletal disorders ? • Musculoskeletal disorders (MSDs) are injuries and discomforts that affect the human body’s movement or musculoskeletal system. • The factors contributing to MSDs include but are not limited to: • work postures and movements, • repetitiveness and pace of work, • force of movements, vibration, temperature, • increased pressure (e.g., to increase productivity). • Certain workplace conditions such as: • the layout of the workstation, • the speed of work and the weight of the objects being handled etc.
- 7. MSD in EU and Worldwide • MSDs are one of the most prevalent occupational disease in the European Union (EU) • Throughout Europe MSDs affect millions of workers and cost employers billions of euros • Based on the latest survey from the Labour Force in Great Britain, 41% of the work illness is due to MSDs • similarly, it is the dominant health related problems among the working population in the EU • Such facts and statistics are spread worldwide as reported by the Canadian Centre for Occupational Health and Safety in and Safe Work Australia in and so on
- 8. Existing Works • Existing studies on various occupational workers includes: • computer and office workers, • nurses, caregivers and paramedical staffs, dentists, • cashiers etc. • They cover the discomforts and disorders: • in neck/shoulder, fingers/wrist/hands, upper and lowers back pains etc. • These studies rely on huge set of data collections across various professionals among different countries and continents. • However, the existing data collection methods are dominantly based on “surveys and questionnaires” which are often not very accurate. Mostly, the estimated percentage of particular positions and postures are obtained based on mere guess. “There are serious concerns regarding the reliability/accuracy/real-time requirements of the existing data collection methods”
- 9. Our Motivation (1/2) • To improve such data collection methods portable sensing platform could provide: • miniaturized wearable and implant sensors for real-time monitoring • obtain accurate positions, postures and orientations, • load measurements on different bones and so on. • To exploit the data collections at various levels: • First, at the local level, POCT and diagnostics with alarms/ warnings. • The local level processing and decision process: • exploit deployed infrared cameras to monitor and keep track of the movements • to perform “Gait analysis” early warnings and cautions can be provided to the workers.
- 10. • Another important aspects are: • remote level monitoring, processing diagnostics and early detections • Exploiting deployed sensors and devices to transmit to the remote cloud for long- terms predictions and estimations. • machine learning algorithms and techniques for big data analytics Our Motivation (2/2)
- 11. Technologies Sensing (TUT) Low-power near sensor processing (TUT, CNRS-UBS) Data communication (TUT, CNRS-Rennes 1) “Cloud“ (IRT-bcom, CNRS-UBO) Security and privacy (IRT-bcom) National IT infrastructures (e.g. X-Road in Estonia) Digital adaptive services and solutions: smart work environments for older adults (but not only) Remain actively involved in professional life Sustain and renew their work and personal life related skills Support independent active and healthy lifestyles while taking into account reduced capabilities due to age- related health risks and conditions Stakeholders Employees (older ones, but not only) Employers (workplaces such as factories) Insurance (national health insurance + private) Unions Policy makers Medical experts Ergonomics (TU, Karolinska) Kinesiology and Biomechanics (TU) Re-education (TU, Karolinska) Social experts (LU) Behaviour Sociology Psychology Gender and culture Laws Participatory design TUT: Tallinn University of Technology (Thomas Johann Seebeck Department of Electronics) TU: Tartu University (Institute of Sport Sciences and Physiotherapy) 11 AllthesepointssupporttheTrans-disiplinaryaspect(seecall) 1 CNRS Team: Lab-STICC-UBS (Diguet) / lab-STICC- UBO (Boukhobza) / IRISA-Rennes 1 (Sentieys, Berder) IRT-bcom LU: Loughborough University Others ?