Eindhoven University of Technology
Eindhoven University of Technology (TU/e, www.tue.nl
) is a world-leading research university specializing in engineering science & technology. The Department of Electrical Engineering is responsible for research and education in Electrical Engineering. The TU/e is the world's best-performing research university in terms of research cooperation with industry (#1 since 2009).Electronic Systems group at the TU/e
The Electronic Systems (ES) group (www.es.ele.tue.nl
) comprises three full professors, two part-time full professors, two associate professors, six assistant professors, about 40 PhD candidates and postdocs, and several technical and support staff. The group has excellent infrastructure that includes individual computers, compute servers, state-of-the-art FPGA and GPU farms, sensor- and ad-hoc networking equipment, and a comprehensive range of electronic-design software.
The ES group is world-renowned for its design automation and embedded systems research. It is our ambition to provide a scientific basis for design trajectories of digital electronic circuits, embedded and cyber-physical systems. Our research results in tools and platforms such as SDF3 and CompSOC. The ES group excels in the area of system-level design and performance analysis. A variety of state-of-the-art tools and chips have been developed by our ES group together with our industrial partners, such as NXP Semiconductors, Philips Healthcare and IMEC-Holst Centre. We have strong national and international academic and industrial contacts. Topic Embedded Products
) offers consulting, projects, and products for embedded systems. Topic is based in Best, 10 km North of the TU/E. Topic Embedded Products (https://topicembeddedproducts.com
) designs advanced high-performance FPGA boards and systems on module for applications that have any or all of the following characteristics: safety critical, high dependability, high performance, highly dynamic. The Dyplo framework (https://topicembeddedproducts.com/products/dyplo/
) makes it easy to design dynamic high performance applications using advanced FPGA dynamic partial reconfiguration techniques. Prof. Goossens holds a part-time position at Topic Embedded Products.
The PhD position is sponsored by Topic Embedded Products, and will be carried out in close collaboration between the Electronic Systems group at the TU/E and Topic Embedded Products. The Energy Autonomous Systems for Internet of Things (ZERO) project
The aim of the project is to make IoT devices battery-oblivious; in other words, to create energy- autonomous ultra-low power devices. This means that energy usage has to be drastically reduced, far below what is currently possible. Achieving this requires a holistic design approach, i.e., taking all design layers and system components into account. The ZERO project contains five subprojects. The PhD position is in the DAMC subproject.The PhD position in the Dependable Autonomous Mobile Computing (DAMC) subproject
The DAMC subproject aims to reduce energy consumption of FPGAs while maintaining sufficient computing and communication capabilities when really needed. Processing at minimal energy, whilst obeying the functional and non-functional (performance and dependability) requirements, can only be achieved by addressing:the most energy-efficient hardware and software architectures, for MPSOC FPGAs such as Zynq and Ultrascale, andhardware/software system architectures with dynamic (adaptive and predictive) management of all system resources; e.g., predictive deployment of tasks to balance the required processing power (to fulfill the functional and non-functional requirements) and the energy usage, andreliability including fault tolerance and adaptive applications and architectures.
The PhD candidate will research how to implement adaptive dynamic applications on advanced MPSOC FPGAs using dynamic partial reconfiguration. This requires defining and implementing a software programming model for adaptive dynamic application, the associated run-time middleware to dynamically load and adapt applications running on embedded processors (e.g. ARMs on Xilinx Zynq) and FPGA fabric, and energy and fault/reliability models of the entire system. The focus is on the smart algorithms to load balance and reconfigure applications at run time to minimize the energy and maximize reliability.
The candidate will be a part of the CompSOC team at the TU/E and the Dyplo team at Topic. The candidate will be supervised by Prof. dr. Kees Goossens from the ES group and Dirk van den Heuvel of Topic Embedded Products, the architect of Dyplo. The PhD candidate will be located three days per week at Topic.