Cost-efficient FPGA implementation of basal ganglia and their Parkinsonian analysis

摘要: The basal ganglia (BG) comprise multiple subcortical nuclei, which are responsible for cognition and other functions. Developing a brain-machine interface (BMI) demands suitable solution the real-time implementation of portable BG. In this study, we used digital hardware BG network containing 256 modified Izhikevich neurons 2048 synapses to reliably reproduce biological characteristics on single field programmable gate array (FPGA) core. We also highlighted role Parkinsonian analysis by considering neural dynamics in design hardware-based architecture. Thus, developed multi-precision architecture based precise using FPGA-based platform with fixed-point arithmetic. proposed embedding can be applied intelligent agents neurorobotics, as well BMI projects clinical applications. Although only characterized models, approach extended more complex neuron models types functional networks. engineer hybrid nucleus.We propose cost-efficient method computational speed.The biorealistic manner.Parkinsonian resource-cost criteria introduced evaluation.A novel multi-clock domain is multi-module