Development of a Soft Error tolerant Parallel Hash Table on FPGAs for safety critical applications
Reference persons LUCA STERPONE
Research Groups DAUIN - GR-05 - ELECTRONIC CAD & RELIABILITY GROUP - CAD
Thesis type RESEARCH / EXPERIMENTAL
Description A key component in Artificial Intelligence and Machine Learning applications is the fast execution of quicl search and retrieval of data. Hash tables are a fundamental data structure to perform these operation efficiently. In safety critical applications such as automotive or aerospace, it is not sufficient to have a performant architecture but it is also necessary to guarantee a given level of robustness and protection versus soft errors. In the present thesis, the student will explore, study and development a new architecture for implementing parallel and fault tolerant hash table for SRAM-based FPGAs for aerospace applications. The thesis will be performed in collaboration with the European Space Agency (ESA) and in cooperation with AMD/Xilinx.
Required skills Design of Circuits and Systems in VHDL or Verilog.
Preliminary basic knowledge of FPGA architectures.
Deadline 30/11/2022 PROPONI LA TUA CANDIDATURA