Sampling criterion for nonlinear systems with a bandpass input

Sampling requirements for nonlinear systems with a bandpass input are developed. It is well known that the output of a nonlinear system may have a larger bandwidth than that of the input. According to the Nyquist sampling theorem, the sampling rate needs to be at least twice the maximum frequency of the output to avoid aliasing. However, if the input is a bandpass signal, the spectrum of the output is usually distributed over several frequency bands. In this case, using the bandpass sampling concept, it is possible to sample the output at a much lower rate. Conditions for such a lower sampling rate to exist are derived for nonlinear systems up to the third order. Supporting computer simulation is also provided.