Understanding and emulating the brain behaviour is one of the most challenging topic in neuroscience but simulating such complicated models always requires a lot of computation and, as a consequence, a lot of time. In order to efficiently evaluate a model it could be very useful to speed up the simulation process by using software or hardware accelerators. This paper presents an FPGA based accelerator who speeds up the simulation of two different kinds of innovative cerebellar neurons model, such as the Golgi cell and the granular cell, developed by the Physiology department of the University of Pavia. Since the achieved performance is real time compliant, such a circuit could be used, in the future, to bypass some dysfunction of the cerebellum in the human being.
A hardware accelerator for real time simulation of complex neuronal models
MAJANI, ALESSANDRA;DANESE, GIOVANNI;LEPORATI, FRANCESCO
2012-01-01
Abstract
Understanding and emulating the brain behaviour is one of the most challenging topic in neuroscience but simulating such complicated models always requires a lot of computation and, as a consequence, a lot of time. In order to efficiently evaluate a model it could be very useful to speed up the simulation process by using software or hardware accelerators. This paper presents an FPGA based accelerator who speeds up the simulation of two different kinds of innovative cerebellar neurons model, such as the Golgi cell and the granular cell, developed by the Physiology department of the University of Pavia. Since the achieved performance is real time compliant, such a circuit could be used, in the future, to bypass some dysfunction of the cerebellum in the human being.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
