Real-time and Networked Embedded Systems
While invisible in many cases, the presence of real-time and networked embedded systems is being spreading with a staggering pace during the past years, consuming the largest part of the world-wide production of microprocessors / microcontrollers, bridging the physical and information worlds and seamlessly connecting smart miniaturized objects to the globally connected systems of the Industrial Internet of Things (IIoT). This is the backbone of the various Cyber-Physical Systems (CPS), which are identified as the emerging “neural system" in society and economy today, and constitute one of the principal dimensions of the 4th Industrial Revolution.
Main current research topics of interest include:
- Operating systems and middleware, resource management, resource usage tracking and optimization for resource constrained real-time & networked embedded systems
- Wired/wireless real-time industrial communications and networks, fieldbuses, wireless sensor/actuator networks (WSANs) and RFID
- Energy consumption modeling and management, dynamic reconfiguration, QoS, heterogeneity and interoperability
- Modelling, simulation and development tools for real-time & networked embedded systems, wireless sensor networks and distributed applications
- Hardware/software co-design, synthesis, reconfigurable hardware (FPGAs), real-time network- and system on chip
- Low-overhead, efficient algorithm implementations in software and reconfigurable hardware (FPFAs), including application specific processing, network protocols, compression and encryption, and HW/SW architectures for resource constrained networked embedded systems
- Embedded System security – Security protocol design and development.
- Secure embedded system hardware and software design including reconfigurable hardware, resistance against software/hardware side channel attacks
- Reference architectures, frameworks and performance evaluations for edge and fog embedded computing in IIoT
- Real-time & networked embedded systems and sensor networks support for digital twins and edge-based analytics
- Machine and Deep learning on Embedded Devices on the Edge
- Online learning from streaming data, Online learning for multimodal information fusion in IoT systems
- Distributed Architectures for Adaptive Systems, Deep learning technologies for distributed real - time embedded systems
- Distributed optimization and statistical learning for dense wireless sensor networks
- Artificial Intelligence tools for driver state monitoring and scene analysis in connected and automated vehicles
- 3D Sensing and Computing for Augmented Reality/tele-immersion applications
- New applications and use-cases of IIoT-enabled, real-time & networked embedded systems and wireless sensor/actuator networks for smart environments in process and manufacturing industry, transportation, energy, health, agricultural, environmental, space and seismological/geophysical sectors. Design, implementation and performance evaluation
ISI has been active in research and development in the area of Real-Time and Networked Embedded Systems and their industrial applications since its establishment, participating in many European and national projects, including flagship ones, such as EU project R-Fieldbus (first steps in real-time radio communications and wireless multimedia streams in harsh industrial environments), or others such as EU projects LOCCATEC (earthquake emergency response and rescue teams support), BETSY (adaptive wireless multimedia streams in handhelds), uSWN (generic and reusable paradigms for optimal deployment of Wireless Sensor Networks – WSNs), WSN-DPCM (WSN Development, Planning, Commissioning and Maintenance), ASPIS (in-tunnel wireless embedded networking infrastructure), SOLDER (Spectrum overlay through aggregation of heterogeneous dispersed bands), HEAP (High Efficient Adaptive Multi-processor Framework), ERA (Embedded Reconfigurable Architectures), as well as national projects, such as the Educational Seismograph and MODSENSE (modular WSN architecture for heterogeneous applications like seismic networks or building energy management systems), WelCOM (Wireless Sensors for Engineering Asset Life Cycle Optimal Management), PALMIS (IIoT and cyber-physical Systems for condition-based maintenance of production assets), MERIT (RFID-based Safety in Marine Transportations), ISRTDI (IoT Components and Systems for Secure Smart Building Environments), I3T (IIoT-based Cyber-Physical Systems in Smart Energy and Smart Manufacturing environments), DAMAZO (Demand Response Energy Management System), DynaCargo (cargo-centric transport management system), KATHODIGOS (WSN based smart parking), SELIDA (RFID based library management system), MyOlivGroveCoach (Analysis, modeling and multi-spectral sensing for the predictive management of Verticillium wilt in olive groves) and DEEP EVIoT (Deep embedded vision using sparse convolutional neural networks).