Congratulations on the nice paper!
Can you perhaps rerun the experiments with high-frequency triangle waves instead of noise? The best is to use a frequency similar to that of the high- frequency end of your noise bandwidth.
According to the results of my calculations, which I presented in the public debate session about Stochastic resonance at the Fluctuation and Noise (FaN) Symposium in Florence (2007), using triangle waves instead of noise would produce the best signal-to-noise ratio at the output of a threshold stochastic resonator and would not result in any stochasticity in the output signal.
Laszlo Kish (formerly L.B. Kiss) http://engineering.tamu.edu/electrical/people/lkish
None declared