education
MSc thesis project proposal

DOME algorithms & machines for the SKA radio telescope

Topic: SKA is a future radio telescope consisting of thousands of elements, producing images with hundreds of teraflops computational complexity. This requires special algorithms and machines

Project outside the university

ASTRON-IBM (Dwingeloo)

The Project

The Dome project is a unique collaboration between IBM Research Zurich and ASTRON (Netherlands Institute for Radio Astronomy) to perform fundamental research towards building the international Square Kilometre Array (SKA), the world's largest and most sensitive radio telescope: http://www.dome-exascale.nl/. Dome concentrates on three domains:
  • Sustainable supercomputing (extreme performance within a limited energy budget)
  • Extreme streaming (real-time processing of massive amounts of data)
  • Nano-photonics.
The goal of the Dome Algorithms & Machines team is to analyze the fundamental design tradeoffs in building exa-scale processing systems. Currently, a set of ASICs for low-power processing of antenna data in the early stages of the telescope's pipeline is being defined.

Assignment

Dome offers an internship or master thesis opportunity (3 to 9 months) with the ASTRON & IBM Center for Exascale Technology in Dwingeloo, The Netherlands. The project goal will be to work out the implementation details of the beamformer ASIC (containing FIR filters, FFTs and beamformers), optimizing it for power and area, and make it conform with SKA requirements. The large data volumes in the SKA put large requirements on the compute system. As the telescope will be located in the desert, power is scarce and low-power designs are crucial.

For the student, the opportunity exists to work in an international project, together with researchers and engineers from both ASTRON and IBM Research in the Netherlands and Switzerland.

Requirements

MSc student in Electrical Engineering or an equivalent field, with a background in signal processing and hardware design, and, ideally, experience with implementations of signal processing algorithms in hardware (FPGA, ASIC). Given the international nature of the project, the student should be able to communicate fluently using the English language.

Contact

prof.dr.ir. Alle-Jan van der Veen

Circuits and Systems Group

Department of Microelectronics

Last modified: 2014-08-27