Skip to main content
. 2015 Aug 11;9:8. doi: 10.3389/fnbot.2015.00008

Figure 7.

Figure 7

Closed-loop simulation platform and neural network architecture for the Bioinspired BMI (B-BMI). (A) Closed-loop simulation platform for the B-BMI. The signal synthesizer (Synt-A) generates Poisson-distributed (inverted) pulse signals to simulate extracellular recordings from 18 primary motor cortex (M1) neurons and provides inputs for 18 analog channels of the DAQ device. SNN simulation tasks run the B-BMI by simulating the MSNs, which receive synapses from M1 neurons of Synt-A. A button press provides external input for the TTL control module to initiate a trial. Synt-A receives inputs from TTL control module to adjust the firing rates of the M1 neurons according to the selected target or the out-of-trial (baseline) status. (B) Tuning map of the M1 neurons of Synt-A (B, baseline; L, left; R, right direction) and neural network architecture for the B-BMI in closed-loop simulations. Red/green lines indicate synaptic connections between M1 neurons and left/right action MSNs.