Ricerca CERCA

High throughput implementation of baseband algorithms by means of an NVIDIA graphic processing unit


Reference persons GUIDO MASERA

External reference persons Maurizio Martina

Description In several application contexts, running simulations or data processing models can take extremely long time to complete and requires expensive computers or computing clusters. A new solution to process computational intensive applications and computer based simulations several times faster than a current top of the line workstation exploits the massive parallel processing capabilities available in graphics processing units (GPUs) such as NVIDIA or other multi-core hardware environments. The objective of this thesis is the use of a GPU platform for accelerating fixed point hardware oriented models for the implementation of error correcting decoders.

Deadline 19/10/2014      PROPONI LA TUA CANDIDATURA