BIOMIMETIC CORTICAL NANOCIRCUITS - Bio-Inspired and Nanoscale Integrated Computing - page 466

1

AP1
PSP1
PSP2

0.8

AP1

0.6

0.4

0.2

0

0 1 2 3

Time in seconds

4 5 6

× 10 − 11

Figure 17.4. The effect of presynaptic action potential variation on the PSP.

potential. Likewise, the speed of the carbon nanotubes allowed us to scale the
delays significantly, with about 1ms in the biological neuron scaling to about 2 ps
in the nanotube neuron [76].

17.4.1. Experiments with the Nanotube Synapse
We simulated the synapse with the ion channel electomotive force, E-channel, at
0.8V, and with reuptake control, R, at 0.3V. The ion pump gate voltage was set to
0.4V representing typical operation. The postsynaptic potential appearing at the
dendritic trunk is shown in Figure 17.3. This potential is approximately 14% of the
action potential and the duration is about six times as long as the action potential,
similar to EPSP's described in the literature [77]. While the control voltages were
tuned to yield reasonable operation, the input voltage was varied in duration to
remain at a maximum of 0.8V for 2.5 ps and an acceptable PSP still resulted. This
sensitivity testing revealed that the circuit is not highly sensitive to the presynaptic
action potential's gross characteristics. This second action potential with increased
duration and lowered magnitude is shown in Figure 17.4. The typical PSP from
Figure 17.3 is reproduced here, along with the PSP resulting from the modified
action potential.

1

AP
PSP
PSP2
PSP3
PSP4

0.8

0.6

0.4

0.2

0

0 1 2 3

Time in seconds

4 5 6

× 10 − 11

Figure 17.5. The action potential and resulting PSPs with reuptake control
varying from 0V to 0.9V.