The objectives of the ESDA group include all aspects of Research and Development of Electronic Systems, and Research and Development of Tools and Algorithms for Design Automation. Current emphasis is given to the area of Reconfigurable Systems.
The research on Reconfigurable Computing gives emphasis both to the use of Reconfigurable Embedded Systems, e.g. using Field Programmable Logic Arrays (FPGAs) and SoC FPGAs as embedded platforms. These platforms allow the design of dedicated hardware/software solutions with high-performance and low cost when compared to software-based architectures and/or embedded Graphics Processing Units.
• High-Performance Embedded Computing
High-Performance Embedded Systems require configurable hardware/software Systems on a Chip (SoCs) integrating multi- and many-core embedded processors combined with specially designed hardware accelerators.
• Custom Programmable Hardware Architectures (e.g. for IoT)
With the advent of the Internet of Things, geographically distributed sensors collect a high volume of data which needs to be analyzed and transmitted over the network.
Embedded processors combined with specially designed hardware accelerators process data at the sensor location, to avoid overflowing servers with data and wasting a lot of energy in the transmission.
Custom programmable hardware solutions can be used to dynamically create useful computational datapaths, consuming orders of magnitude less power than conventional solutions.
• Resilient Architectures for Dependable (Embedded) Systems
With the increase of complexity in embedded systems, it has become more difficult to assure that they will operate as intended for their expected life. Moreover, as we depend more on them every day, some of them are becoming critical. Hence, there's a need for innovative methodologies and architectures to design reconfigurable systems, more resilient and tolerant to faults, with less expensive implementations, in terms of circuit resources, power and time, than traditional designs (e.g including redundancy).
The envisioned applications include spectral analysis using FFTs or other transforms, image processing, data algorithms such as deep learning and k-means clustering and data processing algorithms from consumer electronics.
This research impacts the most on areas with a very limited budget for area, power and tolerance to errors, such as (aero)space and automotive industries.