Command: testsuite::test_iaf_ps_psp_poisson_spike_accuracy

(test_iaf_ps_psp_poisson_spike_accuracy) run -> compare with analytical solution


The tests generates a poisson spike train using a random number
generator and random number distribution available in the SLI
programming language. In a second step this spike train is supplied to
a neuron model and the resulting subthreshold membrane potential
fluctuations are compared to the analytical solution. Thus in
contrast to the more advanced test_iaf_ps_psp_poisson_accuracy this
test does not require the interaction of the generator and the neuron
model to work and does not require the availability of a parrot
neuron. In contrast to test_iaf_ps_psp_poisson_accuracy the DC
required to maintain a subthreshold membrane potential is generated by
a dc generator not a property of the neuron model. The
spike_generator used to supply the neuron model with spikes
constraints spike times to the tic grid of the simulation kernel. This
is the temporal resolution in which the computation step size and
simulation times are expressed. Therefore the results of simulations
at different computation step sizes only differ because of limited
machine precision. The difference between the analytical result and
the simulation however is dictated by the number of tics per